[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-Little-Podi--Collaborative_Perception":3,"tool-Little-Podi--Collaborative_Perception":61},[4,18,26,36,44,53],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":17},4358,"openclaw","openclaw\u002Fopenclaw","OpenClaw 是一款专为个人打造的本地化 AI 助手，旨在让你在自己的设备上拥有完全可控的智能伙伴。它打破了传统 AI 助手局限于特定网页或应用的束缚，能够直接接入你日常使用的各类通讯渠道，包括微信、WhatsApp、Telegram、Discord、iMessage 等数十种平台。无论你在哪个聊天软件中发送消息，OpenClaw 都能即时响应，甚至支持在 macOS、iOS 和 Android 设备上进行语音交互，并提供实时的画布渲染功能供你操控。\n\n这款工具主要解决了用户对数据隐私、响应速度以及“始终在线”体验的需求。通过将 AI 部署在本地，用户无需依赖云端服务即可享受快速、私密的智能辅助，真正实现了“你的数据，你做主”。其独特的技术亮点在于强大的网关架构，将控制平面与核心助手分离，确保跨平台通信的流畅性与扩展性。\n\nOpenClaw 非常适合希望构建个性化工作流的技术爱好者、开发者，以及注重隐私保护且不愿被单一生态绑定的普通用户。只要具备基础的终端操作能力（支持 macOS、Linux 及 Windows WSL2），即可通过简单的命令行引导完成部署。如果你渴望拥有一个懂你",349277,3,"2026-04-06T06:32:30",[13,14,15,16],"Agent","开发框架","图像","数据工具","ready",{"id":19,"name":20,"github_repo":21,"description_zh":22,"stars":23,"difficulty_score":10,"last_commit_at":24,"category_tags":25,"status":17},3808,"stable-diffusion-webui","AUTOMATIC1111\u002Fstable-diffusion-webui","stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面，旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点，将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。\n\n无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师，还是想要深入探索模型潜力的开发者与研究人员，都能从中获益。其核心亮点在于极高的功能丰富度：不仅支持文生图、图生图、局部重绘（Inpainting）和外绘（Outpainting）等基础模式，还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外，它内置了 GFPGAN 和 CodeFormer 等人脸修复工具，支持多种神经网络放大算法，并允许用户通过插件系统无限扩展能力。即使是显存有限的设备，stable-diffusion-webui 也提供了相应的优化选项，让高质量的 AI 艺术创作变得触手可及。",162132,"2026-04-05T11:01:52",[14,15,13],{"id":27,"name":28,"github_repo":29,"description_zh":30,"stars":31,"difficulty_score":32,"last_commit_at":33,"category_tags":34,"status":17},1381,"everything-claude-code","affaan-m\u002Feverything-claude-code","everything-claude-code 是一套专为 AI 编程助手（如 Claude Code、Codex、Cursor 等）打造的高性能优化系统。它不仅仅是一组配置文件，而是一个经过长期实战打磨的完整框架，旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。\n\n通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能，everything-claude-code 能显著提升 AI 在复杂任务中的表现，帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略，使得模型响应更快、成本更低，同时有效防御潜在的攻击向量。\n\n这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库，还是需要 AI 协助进行安全审计与自动化测试，everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目，它融合了多语言支持与丰富的实战钩子（hooks），让 AI 真正成长为懂上",152630,2,"2026-04-12T23:33:54",[14,13,35],"语言模型",{"id":37,"name":38,"github_repo":39,"description_zh":40,"stars":41,"difficulty_score":32,"last_commit_at":42,"category_tags":43,"status":17},2271,"ComfyUI","Comfy-Org\u002FComfyUI","ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎，专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式，采用直观的节点式流程图界面，让用户通过连接不同的功能模块即可构建个性化的生成管线。\n\n这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景，也能自由组合模型、调整参数并实时预览效果，轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性，不仅支持 Windows、macOS 和 Linux 全平台，还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构，并率先支持 SDXL、Flux、SD3 等前沿模型。\n\n无论是希望深入探索算法潜力的研究人员和开发者，还是追求极致创作自由度的设计师与资深 AI 绘画爱好者，ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能，使其成为当前最灵活、生态最丰富的开源扩散模型工具之一，帮助用户将创意高效转化为现实。",108322,"2026-04-10T11:39:34",[14,15,13],{"id":45,"name":46,"github_repo":47,"description_zh":48,"stars":49,"difficulty_score":32,"last_commit_at":50,"category_tags":51,"status":17},6121,"gemini-cli","google-gemini\u002Fgemini-cli","gemini-cli 是一款由谷歌推出的开源 AI 命令行工具，它将强大的 Gemini 大模型能力直接集成到用户的终端环境中。对于习惯在命令行工作的开发者而言，它提供了一条从输入提示词到获取模型响应的最短路径，无需切换窗口即可享受智能辅助。\n\n这款工具主要解决了开发过程中频繁上下文切换的痛点，让用户能在熟悉的终端界面内直接完成代码理解、生成、调试以及自动化运维任务。无论是查询大型代码库、根据草图生成应用，还是执行复杂的 Git 操作，gemini-cli 都能通过自然语言指令高效处理。\n\n它特别适合广大软件工程师、DevOps 人员及技术研究人员使用。其核心亮点包括支持高达 100 万 token 的超长上下文窗口，具备出色的逻辑推理能力；内置 Google 搜索、文件操作及 Shell 命令执行等实用工具；更独特的是，它支持 MCP（模型上下文协议），允许用户灵活扩展自定义集成，连接如图像生成等外部能力。此外，个人谷歌账号即可享受免费的额度支持，且项目基于 Apache 2.0 协议完全开源，是提升终端工作效率的理想助手。",100752,"2026-04-10T01:20:03",[52,13,15,14],"插件",{"id":54,"name":55,"github_repo":56,"description_zh":57,"stars":58,"difficulty_score":10,"last_commit_at":59,"category_tags":60,"status":17},4487,"LLMs-from-scratch","rasbt\u002FLLMs-from-scratch","LLMs-from-scratch 是一个基于 PyTorch 的开源教育项目，旨在引导用户从零开始一步步构建一个类似 ChatGPT 的大型语言模型（LLM）。它不仅是同名技术著作的官方代码库，更提供了一套完整的实践方案，涵盖模型开发、预训练及微调的全过程。\n\n该项目主要解决了大模型领域“黑盒化”的学习痛点。许多开发者虽能调用现成模型，却难以深入理解其内部架构与训练机制。通过亲手编写每一行核心代码，用户能够透彻掌握 Transformer 架构、注意力机制等关键原理，从而真正理解大模型是如何“思考”的。此外，项目还包含了加载大型预训练权重进行微调的代码，帮助用户将理论知识延伸至实际应用。\n\nLLMs-from-scratch 特别适合希望深入底层原理的 AI 开发者、研究人员以及计算机专业的学生。对于不满足于仅使用 API，而是渴望探究模型构建细节的技术人员而言，这是极佳的学习资源。其独特的技术亮点在于“循序渐进”的教学设计：将复杂的系统工程拆解为清晰的步骤，配合详细的图表与示例，让构建一个虽小但功能完备的大模型变得触手可及。无论你是想夯实理论基础，还是为未来研发更大规模的模型做准备",90106,"2026-04-06T11:19:32",[35,15,13,14],{"id":62,"github_repo":63,"name":64,"description_en":65,"description_zh":66,"ai_summary_zh":66,"readme_en":67,"readme_zh":68,"quickstart_zh":69,"use_case_zh":70,"hero_image_url":71,"owner_login":72,"owner_name":73,"owner_avatar_url":74,"owner_bio":75,"owner_company":76,"owner_location":77,"owner_email":78,"owner_twitter":79,"owner_website":80,"owner_url":81,"languages":82,"stars":83,"forks":84,"last_commit_at":85,"license":82,"difficulty_score":86,"env_os":87,"env_gpu":88,"env_ram":88,"env_deps":89,"category_tags":92,"github_topics":93,"view_count":32,"oss_zip_url":82,"oss_zip_packed_at":82,"status":17,"created_at":110,"updated_at":111,"faqs":112,"releases":148},7050,"Little-Podi\u002FCollaborative_Perception","Collaborative_Perception","This repository is a paper digest of recent advances in collaborative \u002F cooperative \u002F multi-agent perception for V2I \u002F V2V \u002F V2X autonomous driving scenario.","Collaborative_Perception 是一个专注于自动驾驶领域协同感知技术的开源论文汇总库。它系统性地整理了近年来在车与车（V2V）、车与基础设施（V2I）及车与万物（V2X）场景下，多智能体协作感知的最新研究进展。\n\n在自动驾驶中，单车感知常受限于视距遮挡和传感器盲区，难以应对复杂路况。Collaborative_Perception 通过汇集前沿学术成果，帮助从业者理解如何利用车辆间的通信共享数据，突破单点感知局限，实现超视距的环境理解，从而提升驾驶系统的安全性与鲁棒性。\n\n该资源特别适合自动驾驶算法研究人员、高校学者及相关领域的开发者使用。它不仅按字母顺序梳理了大量核心论文，还精心分类了“方法与框架”、“数据集与模拟器”等关键板块。此外，库中独家收录了丰富的学习资源，包括权威的综述文章、顶级会议的技术教程视频以及专用的代码库（如 V2Xverse），为用户提供了从理论入门到代码复现的一站式支持。对于希望深入探索协同感知机制、追踪技术前沿或寻找实验基准的专业人士而言，这是一个极具价值的知识导航工具。","# Collaborative Perception\n\nThis repository is a paper digest of recent advances in **collaborative** \u002F **cooperative** \u002F **multi-agent** perception for **V2I** \u002F **V2V** \u002F **V2X** autonomous driving scenario. Papers are listed in alphabetical order of the first character.\n\n### :link:Jump to:\n- ### [[Method and Framework](https:\u002F\u002Fgithub.com\u002FLittle-Podi\u002FCollaborative_Perception#bookmarkmethod-and-framework)]\n- ### [[Dataset and Simulator](https:\u002F\u002Fgithub.com\u002FLittle-Podi\u002FCollaborative_Perception#bookmarkdataset-and-simulator)]\n\nNote: I find it hard to fairly compare all methods on each benchmark since some published results are obtained without specified training and testing settings, or even modified model architectures. In fact, many works evaluate all baselines under their own settings and report them. Therefore, it is probably to find inconsistency between papers. Hence, I discard the collection and reproducton of all the benchmarks in a previous update. If you are interested, you can find a bunch of results in [this archived version](https:\u002F\u002Fgithub.com\u002FLittle-Podi\u002FCollaborative_Perception\u002Ftree\u002F1be25908aea0a9f635ff4852b3a90729cf2b6aac).\n\n\n\n## :star2:Recommendation\n\n### Helpful Learning Resource:thumbsup::thumbsup::thumbsup:\n\n- **(Position)** When Autonomous Vehicle Meets V2X Cooperative Perception: How Far Are We? [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.24927)], Research Challenges and Progress in the End-to-End V2X Cooperative Autonomous Driving Competition [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.21610)], Automated Vehicles Should be Connected with Natural Language [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.01059)], Collaborative Perception Datasets for Autonomous Driving: A Review [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.12696)], Collaborative Perception for Connected and Autonomous Driving: Challenges, Possible Solutions and Opportunities [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.01544)], V2X Cooperative Perception for Autonomous Driving: Recent Advances and Challenges [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.03525)], Towards Vehicle-to-Everything Autonomous Driving: A Survey on Collaborative Perception [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.16714)], Collaborative Perception in Autonomous Driving: Methods, Datasets and Challenges [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2301.06262)], A Survey and Framework of Cooperative Perception: From Heterogeneous Singleton to Hierarchical Cooperation [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.10590)]\n- **(Talk)** Vehicle-to-Vehicle (V2V) Communication (Waabi CVPR 24 Tutorial on Self-Driving Cars) [[video](https:\u002F\u002Fyoutu.be\u002FyceuUthWz9s)], Vehicle-to-Vehicle (V2V) Communication (Waabi CVPR 23 Tutorial on Self-Driving Cars) [[video](https:\u002F\u002Fyoutu.be\u002FT-N51B8mZB8)], The Ultimate Solution for L4 Autonomous Driving [[video](https:\u002F\u002Fyoutu.be\u002FcyNxemm4Ujg)], When Vision Transformers Meet Cooperative Perception [[video](https:\u002F\u002Fyoutu.be\u002FrLAU4eqoOIU)], Scene Understanding beyond the Visible [[video](https:\u002F\u002Fyoutu.be\u002Foz0AnmJZCR4)], Robust Collaborative Perception against Communication Interruption [[video](https:\u002F\u002Fyoutu.be\u002F3cIWpMrsyeE)], Collaborative and Adversarial 3D Perception for Autonomous Driving [[video](https:\u002F\u002Fyoutu.be\u002FW-AONQMfGi0)], Vehicle-to-Vehicle Communication for Self-Driving [[video](https:\u002F\u002Fyoutu.be\u002FoikdOpmIoc4)], Adversarial Robustness for Self-Driving [[video](https:\u002F\u002Fyoutu.be\u002F8uBFXzyII5Y)], L4感知系统的终极形态：协同驾驶 [[video](https:\u002F\u002Fyoutu.be\u002FNvixMEDHht4)], CoBEVFlow-解决车-车\u002F路协同感知的时序异步问题 [[video](https:\u002F\u002Fyoutu.be\u002FIBTgalAjye8)], 新一代协作感知Where2comm减少通信带宽十万倍 [[video](https:\u002F\u002Fyoutu.be\u002Fi5coMk4hkuk)], 从任务相关到任务无关的多机器人协同感知 [[video](https:\u002F\u002Fcourse.zhidx.com\u002Fc\u002FMDlkZjcyZDgwZWI4ODBhOGQ4MzM=)], 协同自动驾驶：仿真与感知 [[video](https:\u002F\u002Fcourse.zhidx.com\u002Fc\u002FMmQ1YWUyMzM1M2I3YzVlZjE1NzM=)], 基于群体协作的超视距态势感知 [[video](https:\u002F\u002Fwww.koushare.com\u002Fvideo\u002Fvideodetail\u002F33015)]\n- **(Library)** V2Xverse: A Codebase for V2X-Based Collaborative End2End Autonomous Driving [[code](https:\u002F\u002Fgithub.com\u002FCollaborativePerception\u002FV2Xverse)] [[doc](https:\u002F\u002Fcollaborativeperception.github.io\u002FV2Xverse)], HEAL: An Extensible Framework for Open Heterogeneous Collaborative Perception [[code](https:\u002F\u002Fgithub.com\u002Fyifanlu0227\u002FHEAL)] [[doc](https:\u002F\u002Fhuggingface.co\u002Fyifanlu\u002FHEAL)], OpenCOOD: Open Cooperative Detection Framework for Autonomous Driving [[code](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002FOpenCOOD)] [[doc](https:\u002F\u002Fopencood.readthedocs.io\u002Fen\u002Flatest\u002Findex.html)], CoPerception: SDK for Collaborative Perception [[code](https:\u002F\u002Fgithub.com\u002Fcoperception\u002Fcoperception)] [[doc](https:\u002F\u002Fcoperception.readthedocs.io\u002Fen\u002Flatest)], OpenCDA: Simulation Tool Integrated with Prototype Cooperative Driving Automation [[code](https:\u002F\u002Fgithub.com\u002Fucla-mobility\u002FOpenCDA)] [[doc](https:\u002F\u002Fopencda-documentation.readthedocs.io\u002Fen\u002Flatest)]\n- **(Workshop)** Co-Intelligence@ECCV'24 [[web](https:\u002F\u002Fcoop-intelligence.github.io)], CoPerception@ICRA'23 [[web](https:\u002F\u002Fcoperception.github.io)], ScalableAD@ICRA'23 [[web](https:\u002F\u002Fsites.google.com\u002Fview\u002Ficra2023av\u002Fhome)]\n- **(Background)** Current Approaches and Future Directions for Point Cloud Object Detection in Intelligent Agents [[video](https:\u002F\u002Fyoutu.be\u002FxFFCQVwYeec)], 3D Object Detection for Autonomous Driving: A Review and New Outlooks [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.09474)], DACOM: Learning Delay-Aware Communication for Multi-Agent Reinforcement Learning [[video](https:\u002F\u002Fyoutu.be\u002FYBgW2oA_n3k)], A Survey of Multi-Agent Reinforcement Learning with Communication [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.08975)]\n\n### Typical Collaboration Modes:handshake::handshake::handshake:\n\n![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FLittle-Podi_Collaborative_Perception_readme_94c6af95b85a.png)\n\n### Possible Optimization Directions:fire::fire::fire:\n\n![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FLittle-Podi_Collaborative_Perception_readme_19d07569c559.png)\n\n\n\n## :bookmark:Method and Framework\n\nNote: {Related} denotes that it is not a pure collaborative perception paper but has related content.\n\n### Selected Preprint\n\n- **ACCO** (Is Discretization Fusion All You Need for Collaborative Perception?) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.13946)] [[code](https:\u002F\u002Fgithub.com\u002Fsidiangongyuan\u002FACCO)]\n- **AR2VP** (Dynamic V2X Autonomous Perception from Road-to-Vehicle Vision) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.19113)] [[code](https:\u002F\u002Fgithub.com\u002Ftjy1423317192\u002FAP2VP)]\n- **CPPC** (Point Cluster: A Compact Message Unit for Communication-Efficient Collaborative Perception) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=54XlM8Clkg)] [~~code~~]\n- **CP-FREEZER** (CP-FREEZER: Latency Attacks against Vehicular Cooperative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.01062)] [[code](https:\u002F\u002Fgithub.com\u002FWiSeR-Lab\u002FCP-FREEZER)]\n- **CMP** (CMP: Cooperative Motion Prediction with Multi-Agent Communication) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.17916)] [~~code~~]\n- **CoBEVFusion** (CoBEVFusion: Cooperative Perception with LiDAR-Camera Bird's-Eye View Fusion) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.06008)] [~~code~~]\n- **CoBEVGlue** (Self-Localized Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.12712)] [[code](https:\u002F\u002Fgithub.com\u002FVincentNi0107\u002FCoBEVGlue)]\n- **CoCMT** (CoCMT: Towards Communication-Efficient Corss-Modal Transformer For Collaborative Perception) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=S1NrbfMS7T)] [[code](https:\u002F\u002Fgithub.com\u002Ftaco-group\u002FCOCMT)]\n- **CoDiff** (CoDiff: Conditional Diffusion Model for Collaborative 3D Object Detection) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.14891)] [~~code~~]\n- **CoDriving** (Towards Collaborative Autonomous Driving: Simulation Platform and End-to-End System) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.09496)] [[code](https:\u002F\u002Fgithub.com\u002FCollaborativePerception\u002FV2Xverse)]\n- **CoDrivingLLM** (Towards Interactive and Learnable Cooperative Driving Automation: A Large Language Model-Driven Decision-making Framework) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.12812)] [[code](https:\u002F\u002Fgithub.com\u002FFanGShiYuu\u002FCoDrivingLLM)]\n- **CollaMamba** (CollaMamba: Efficient Collaborative Perception with Cross-Agent Spatial-Temporal State Space Model) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.07714)] [~~code~~]\n- **CoLMDriver** (CoLMDriver: LLM-Based Negotiation Benefits Cooperative Autonomous Driving) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.08683)] [[code](https:\u002F\u002Fgithub.com\u002Fcxliu0314\u002FCoLMDriver)]\n- **CoMamba** (CoMamba: Real-Time Cooperative Perception Unlocked with State Space Models) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.10699)] [~~code~~]\n- **CoPLOT** (Beyond BEV: Optimizing Point-Level Tokens for Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.19638)] [[code](https:\u002F\u002Fgithub.com\u002FCheeryLeeyy\u002FCoPLOT)]\n- **CP-Guard+** (CP-Guard+: A New Paradigm for Malicious Agent Detection and Defense in Collaborative Perception) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=9MNzHTSDgh)] [~~code~~]\n- **CTCE** (Leveraging Temporal Contexts to Enhance Vehicle-Infrastructure Cooperative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.10531)] [~~code~~]\n- **Debrief** (Talking Vehicles: Cooperative Driving via Natural Language) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=VYlfoA8I6A)] [~~code~~]\n- **DeepFleet** (DeepFleet: Multi-Agent Foundation Models for Mobile Robots) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.08574)] [~~code~~]\n- **DiffCP** (DiffCP: Ultra-Low Bit Collaborative Perception via Diffusion Model) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.19592)] [~~code~~]\n- **FadeLead** (Background Fades, Foreground Leads: Curriculum-Guided Background Pruning for Efficient Foreground-Centric Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.19250)] [~~code~~]\n- **Faster-HEAL** (Faster-HEAL: An Efficient and Privacy-Preserving Collaborative Perception Framework for Heterogeneous Autonomous Vehicles) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.07314)] [~~code~~]\n- **FlowAdapt** (Move What Matters: Parameter-Efficient Domain Adaptation via Optimal Transport Flow for Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.11565)] [~~code~~]\n- **HeatV2X** (HeatV2X: Scalable Heterogeneous Collaborative Perception via Efficient Alignment and Interaction) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.10211)] [[code](https:\u002F\u002Fgithub.com\u002Fhollowknight2167\u002FHeatV2X)]\n- **HyComm** (Communication-Efficient Multi-Agent 3D Detection via Hybrid Collaboration) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.07092)] [~~code~~]\n- **HyDRA** (HyDRA: Hybrid Domain-Aware Robust Architecture for Heterogeneous Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.23975)] [~~code~~]\n- **InSPE** (InSPE: Rapid Evaluation of Heterogeneous Multi-Modal Infrastructure Sensor Placement) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.08240)] [~~code~~]\n- **I2XTraj** (Knowledge-Informed Multi-Agent Trajectory Prediction at Signalized Intersections for Infrastructure-to-Everything) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.13461)] [~~code~~]\n- **JigsawComm** (JigsawComm: Joint Semantic Feature Encoding and Transmission for Communication-Efficient Cooperative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.17843)] [[code](https:\u002F\u002Fgithub.com\u002FWiSeR-Lab\u002FJigsawComm)]\n- **LCV2I** (LCV2I: Communication-Efficient and High-Performance Collaborative Perception Framework with Low-Resolution LiDAR) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.17039)] [~~code~~]\n- **LMMCoDrive** (LMMCoDrive: Cooperative Driving with Large Multimodal Model) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.11981)] [[code](https:\u002F\u002Fgithub.com\u002Fhenryhcliu\u002FLMMCoDrive)]\n- {Related} **MDG** (MDG: Masked Denoising Generation for Multi-Agent Behavior Modeling in Traffic Environments) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.17496)] [~~code~~]\n- **mmCooper** (mmCooper: A Multi-Agent Multi-Stage Communication-Efficient and Collaboration-Robust Cooperative Perception Framework) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.12263)] [~~code~~]\n- **MOT-CUP** (Collaborative Multi-Object Tracking with Conformal Uncertainty Propagation) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.14346)] [[code](https:\u002F\u002Fgithub.com\u002Fsusanbao\u002Fmot_cup)]\n- **OpenCOOD-Air** (OpenCOOD-Air: Prompting Heterogeneous Ground-Air Collaborative Perception with Spatial Conversion and Offset Prediction) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.13919)] [~~code~~]\n- **RefPtsFusion** (From Features to Reference Points: Lightweight and Adaptive Fusion for Cooperative Autonomous Driving) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.18757)] [~~code~~]\n- {Related} **RopeBEV** (RopeBEV: A Multi-Camera Roadside Perception Network in Bird's-Eye-View) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.11706)] [~~code~~]\n- **ParCon** (ParCon: Noise-Robust Collaborative Perception via Multi-Module Parallel Connection) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.11546)] [~~code~~]\n- **PragComm** (Pragmatic Communication in Multi-Agent Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.12694)] [[code](https:\u002F\u002Fgithub.com\u002FPhyllisH\u002FPragComm)]\n- **QPoint2Comm** (Send Less, Perceive More: Masked Quantized Point Cloud Communication for Loss-Tolerant Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.21667)] [~~code~~]\n- **QuantV2X** (QuantV2X: A Fully Quantized Multi-Agent System for Cooperative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.03704)] [[code](https:\u002F\u002Fgithub.com\u002Fucla-mobility\u002FQuantV2X)]\n- **QUEST** (QUEST: Query Stream for Vehicle-Infrastructure Cooperative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.01804)] [~~code~~]\n- **RCDN** (RCDN: Towards Robust Camera-Insensitivity Collaborative Perception via Dynamic Feature-Based 3D Neural Modeling) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.16868)] [~~code~~]\n- **RC-GeoCP** (RC-GeoCP: Geometric Consensus for Radar-Camera Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.00654)] [~~code~~]\n- **ReVQom** (Residual Vector Quantization For Communication-Efficient Multi-Agent Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.21464)] [~~code~~]\n- **RG-Attn** (RG-Attn: Radian Glue Attention for Multi-Modality Multi-Agent Cooperative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.16803)] [~~code~~]\n- **RiskMM** (Risk Map As Middleware: Towards Interpretable Cooperative End-to-End Autonomous Driving for Risk-Aware Planning) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.07686)] [~~code~~]\n- **RoCo-Sim** (RoCo-Sim: Enhancing Roadside Collaborative Perception through Foreground Simulation) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.10410)] [[code](https:\u002F\u002Fgithub.com\u002Fduyuwen-duen\u002FRoCo-Sim)]\n- **SafeCoop** (SafeCoop: Unravelling Full Stack Safety in Agentic Collaborative Driving) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.18123)] [[code](https:\u002F\u002Fgithub.com\u002Ftaco-group\u002FSafeCoop)]\n- **ShareVerse** (ShareVerse: Multi-Agent Consistent Video Generation for Shared World Modeling) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.02697)] [~~code~~]\n- **SiCP** (SiCP: Simultaneous Individual and Cooperative Perception for 3D Object Detection in Connected and Automated Vehicles) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.04822)] [[code](https:\u002F\u002Fgithub.com\u002FDarrenQu\u002FSiCP)]\n- **SparseAlign** (SparseAlign: A Fully Sparse Framework for Cooperative Object Detection) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.12982)] [~~code~~]\n- **Talking Vehicles** (Towards Natural Language Communication for Cooperative Autonomous Driving via Self-Play) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.18334)] [[code](https:\u002F\u002Fgithub.com\u002Fcuijiaxun\u002Ftalking-vehicles)]\n- **TOCOM-V2I** (Task-Oriented Communication for Vehicle-to-Infrastructure Cooperative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.20748)] [~~code~~]\n- {Related} **TYP** (Transfer Your Perspective: Controllable 3D Generation from Any Viewpoint in a Driving Scene) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.06682)] [~~code~~]\n- **UniMM-V2X** (UniMM-V2X: MoE-Enhanced Multi-Level Fusion for End-to-End Cooperative Autonomous Driving) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.09013)] [[code](https:\u002F\u002Fgithub.com\u002FSouig\u002FUniMM-V2X)]\n- **VIMI** (VIMI: Vehicle-Infrastructure Multi-View Intermediate Fusion for Camera-Based 3D Object Detection) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.10975)] [[code](https:\u002F\u002Fgithub.com\u002FBosszhe\u002FVIMI)]\n- **VLIF** (Is Intermediate Fusion All You Need for UAV-Based Collaborative Perception?) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.21774)] [[code](https:\u002F\u002Fgithub.com\u002Fuestchjw\u002FLIF)]\n- **V2V-GoT** (V2V-GoT: Vehicle-to-Vehicle Cooperative Autonomous Driving with Multimodal Large Language Models and Graph-of-Thoughts) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.18053)] [~~code~~]\n- **V2V-LLM** (V2V-LLM: Vehicle-to-Vehicle Cooperative Autonomous Driving with Multi-Modal Large Language Models) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.09980)] [[code](https:\u002F\u002Fgithub.com\u002Feddyhkchiu\u002FV2VLLM)]\n- **V2XPnP** (V2XPnP: Vehicle-to-Everything Spatio-Temporal Fusion for Multi-Agent Perception and Prediction) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.01812)] [[code](https:\u002F\u002Fgithub.com\u002FZewei-Zhou\u002FV2XPnP)]\n- **V2X-DGPE** (V2X-DGPE: Addressing Domain Gaps and Pose Errors for Robust Collaborative 3D Object Detection) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.02363)] [[code](https:\u002F\u002Fgithub.com\u002Fwangsch10\u002FV2X-DGPE)]\n- **V2X-DGW** (V2X-DGW: Domain Generalization for Multi-Agent Perception under Adverse Weather Conditions) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.11371)] [~~code~~]\n- **V2X-DSC** (V2X-DSC: Multi-Agent Collaborative Perception with Distributed Source Coding Guided Communication) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.00687)] [~~code~~]\n- **V2X-M2C** (V2X-M2C: Efficient Multi-Module Collaborative Perception with Two Connections) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.11546)] [~~code~~]\n- **V2X-PC** (V2X-PC: Vehicle-to-Everything Collaborative Perception via Point Cluster) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.16635)] [~~code~~]\n- **V2X-REALM** (V2X-REALM: Vision-Language Model-Based Robust End-to-End Cooperative Autonomous Driving with Adaptive Long-Tail Modeling) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.21041)] [~~code~~]\n- **V2X-RECT** (V2X-RECT: An Efficient V2X Trajectory Prediction Framework via Redundant Interaction Filtering and Tracking Error Correction) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.17941)] [~~code~~]\n- **V2X-ReaLO** (V2X-ReaLO: An Open Online Framework and Dataset for Cooperative Perception in Reality) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.10034)] [~~code~~]\n- **V2X-UniPool** (V2X-UniPool: Unifying Multimodal Perception and Knowledge Reasoning for Autonomous Driving) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.02580)] [[code](https:\u002F\u002Fgithub.com\u002Fsnowwhite1016\u002FV2X-UniPool)]\n- **V2X-VLM** (V2X-VLM: End-to-End V2X Cooperative Autonomous Driving Through Large Vision-Language Models) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.09251)] [~~code~~]\n\n### CVPR 2026\n\n- **CATNet** (CATNet: Collaborative Alignment and Transformation Network for Cooperative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.05255)] [~~code~~]\n- **CodeAlign** (Linking Modality Isolation in Heterogeneous Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.00609)] [[code](https:\u002F\u002Fgithub.com\u002Fcxliu0314\u002FCodeAlign)]\n- **CoLC** (CoLC: Communication-Efficient Collaborative Perception with LiDAR Completion) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.00682)] [[code](https:\u002F\u002Fgithub.com\u002FCatOneTwo\u002FCoLC)]\n- **CoopDiff** (CoopDiff: A Diffusion-Guided Approach for Cooperation under Corruptions) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.01688)] [~~code~~]\n- **MVIG** (Learning Mutual View Information Graph for Adaptive Adversarial Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.19596)] [[code](https:\u002F\u002Fgithub.com\u002Fyihangtao\u002FMVIG)]\n- **WhisperNet** (WhisperNet: A Scalable Solution for Bandwidth-Efficient Collaboration) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.01708)] [~~code~~]\n\n### ICLR 2026\n\n- **RDComm** (Rate-Distortion Optimized Communication for Collaborative Perception) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=920RxFvsMx)] [~~code~~]\n- **SiMO** (SiMO: Single-Modality-Operable Multimodal Collaborative Perception) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=h0iRgjTmVs)] [[code](https:\u002F\u002Fgithub.com\u002Fdempsey-wen\u002FSiMO)]\n\n### AAAI 2026\n\n- **InfoCom** (InfoCom: Kilobyte-Scale Communication-Efficient Collaborative Perception with Information Bottleneck) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2512.10305)] [[code](https:\u002F\u002Fgithub.com\u002Ffengxueguiren\u002FInfoCom)]\n- **SparseCoop** (SparseCoop: Cooperative Perception with Kinematic-Grounded Queries) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2512.06838)] [[code](https:\u002F\u002Fgithub.com\u002Fwang-jh18-SVM\u002FSparseCoop)]\n- **V2VLoc** (V2VLoc: Robust GNSS-Free Collaborative Perception via LiDAR Localization) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.14247)] [[code](https:\u002F\u002Fgithub.com\u002Fwklin214-glitch\u002FV2VLoc)]\n\n# ICRA 2026\n\n- **EIMC** (EIMC: Efficient Instance-aware Multi-Modal Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.02532)] [[code](https:\u002F\u002Fgithub.com\u002Fsidiangongyuan\u002FEIMC)]\n- **WaveComm** (WaveComm: Lightweight Communication for Collaborative Perception via Wavelet Feature Distillation) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.13365)] [[code](https:\u002F\u002Fgithub.com\u002Ferdemtbao\u002FWaveComm)]\n\n### WACV 2026\n\n- **FocalComm** (FocalComm: Hard Instance-Aware Multi-Agent Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2512.13982)] [[code](https:\u002F\u002Fgithub.com\u002Fscdrand23\u002FFocalComm)]\n\n### CVPR 2025\n\n- **CoGMP** (Generative Map Priors for Collaborative BEV Semantic Segmentation) [[paper](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2025\u002Fhtml\u002FFu_Generative_Map_Priors_for_Collaborative_BEV_Semantic_Segmentation_CVPR_2025_paper.html)] [~~code~~]\n- **CoSDH** (CoSDH: Communication-Efficient Collaborative Perception via Supply-Demand Awareness and Intermediate-Late Hybridization) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.03430)] [[code](https:\u002F\u002Fgithub.com\u002FXu2729\u002FCoSDH)]\n- **HeCoFuse** (HeCoFuse: Cross-Modal Complementary V2X Cooperative Perception with Heterogeneous Sensors) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.13677)] [[code](https:\u002F\u002Fgithub.com\u002FChuhengWei\u002FHeCoFuse)]\n- **LangCoop** (LangCoop: Collaborative Driving with Language) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.13406)] [[code](https:\u002F\u002Fgithub.com\u002Ftaco-group\u002FLangCoop)]\n- **PolyInter** (One is Plenty: A Polymorphic Feature Interpreter for Immutable Heterogeneous Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.16799)] [[code](https:\u002F\u002Fgithub.com\u002Fyuchen-xia\u002FPolyInter)]\n- **SparseAlign** (SparseAlign: A Fully Sparse Framework for Cooperative Object Detection) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.12982)] [~~code~~]\n- **TraF-Align** (TraF-Align: Trajectory-aware Feature Alignment for Asynchronous Multi-agent Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.19391)] [[code](https:\u002F\u002Fgithub.com\u002FzhyingS\u002FTraF-Align)]\n- **V2X-R** (V2X-R: Cooperative LiDAR-4D Radar Fusion for 3D Object Detection with Denoising Diffusion) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.08402)] [[code](https:\u002F\u002Fgithub.com\u002Fylwhxht\u002FV2X-R)]\n\n### NeurIPS 2025\n\n- **GenComm** (Pragmatic Heterogeneous Collaborative Perception via Generative Communication Mechanism) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.19618)] [[code](https:\u002F\u002Fgithub.com\u002Fjeffreychou777\u002FGenComm)]\n- **NegoCollab** (NegoCollab: A Common Representation Negotiation Approach for Heterogeneous Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.27647)] [[code](https:\u002F\u002Fgithub.com\u002Fscz023\u002FNegoCollab)]\n\n### ICCV 2025\n\n- **CoopTrack** (CoopTrack: Exploring End-to-End Learning for Efficient Cooperative Sequential Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.19239)] [[code](https:\u002F\u002Fgithub.com\u002Fzhongjiaru\u002FCoopTrack)]\n- **CoST** (CoST: Efficient Collaborative Perception From Unified Spatiotemporal Perspective) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.00359)] [[code](https:\u002F\u002Fgithub.com\u002Ftzhhhh123\u002FCoST)]\n- **INSTINCT** (INSTINCT: Instance-Level Interaction Architecture for Query-Based Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.23700)] [[code](https:\u002F\u002Fgithub.com\u002FCrazyShout\u002FINSTINCT)]\n- **MamV2XCalib** (MamV2XCalib: V2X-Based Target-Less Infrastructure Camera Calibration with State Space Model) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.23595)] [[code](https:\u002F\u002Fgithub.com\u002Fzhuyaoye\u002FMamV2XCalib)]\n- **SlimComm** (SlimComm: Doppler-Guided Sparse Queries for Bandwidth-Efficient Cooperative 3-D Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.13007)] [[code](https:\u002F\u002Fgithub.com\u002Ffzi-forschungszentrum-informatik\u002FSlimComm)]\n- **TurboTrain** (TurboTrain: Towards Efficient and Balanced Multi-Task Learning for Multi-Agent Perception and Prediction) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.04682)] [[code](https:\u002F\u002Fgithub.com\u002Fucla-mobility\u002FTurboTrain)]\n\n### ICLR 2025\n\n- **CPPC** (Point Cluster: A Compact Message Unit for Communication-Efficient Collaborative Perception) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=54XlM8Clkg)] [~~code~~]\n- **R&B-POP** (Learning 3D Perception from Others' Predictions) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=Ylk98vWQuQ)] [[code](https:\u002F\u002Fgithub.com\u002Fjinsuyoo\u002Frnb-pop)]\n- **STAMP** (STAMP: Scalable Task- And Model-Agnostic Collaborative Perception) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=8NdNniulYE)] [[code](https:\u002F\u002Fgithub.com\u002Ftaco-group\u002FSTAMP)]\n  \n### AAAI 2025\n\n- **CoPEFT** (CoPEFT: Fast Adaptation Framework for Multi-Agent Collaborative Perception with Parameter-Efficient Fine-Tuning) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.10705)] [[code](https:\u002F\u002Fgithub.com\u002Ffengxueguiren\u002FCoPEFT)]\n- **CP-Guard** (CP-Guard: Malicious Agent Detection and Defense in Collaborative Bird's Eye View Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.12000)] [~~code~~]\n- **DSRC** (DSRC: Learning Density-Insensitive and Semantic-Aware Collaborative Representation against Corruptions) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.10739)] [[code](https:\u002F\u002Fgithub.com\u002FTerry9a\u002FDSRC)]\n- **UniV2X** (End-to-End Autonomous Driving through V2X Cooperation) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.00717)] [[code](https:\u002F\u002Fgithub.com\u002FAIR-THU\u002FUniV2X)]\n\n### MM 2025\n\n- **How2Compress** (How2Compress: Scalable and Efficient Edge Video Analytics via Adaptive Granular Video Compression) [[paper](https:\u002F\u002Fwyhallenwu.github.io\u002Fassets\u002Fpdf\u002Fpaper_archive\u002Fhow2compress.pdf)] [[code](https:\u002F\u002Fgithub.com\u002Fwyhallenwu\u002Fhow2compress)]\n- **Selective Shift** (Selective Shift: Towards Personalized Domain Adaptation in Multi-Agent Collaborative Perception) [~~paper~~] [~~code~~]\n\n### ICRA 2025\n\n- **CoDynTrust** (CoDynTrust: Robust Asynchronous Collaborative Perception via Dynamic Feature Trust Modulus) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.08169)] [[code](https:\u002F\u002Fgithub.com\u002FCrazyShout\u002FCoDynTrust)]\n- **CoopDETR** (CoopDETR: A Unified Cooperative Perception Framework for 3D Detection via Object Query) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.19313)] [~~code~~]\n- **Co-MTP** (Co-MTP: A Cooperative Trajectory Prediction Framework with Multi-Temporal Fusion for Autonomous Driving) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.16589)] [[code](https:\u002F\u002Fgithub.com\u002Fxiaomiaozhang\u002FCo-MTP)]\n- **Direct-CP** (Direct-CP: Directed Collaborative Perception for Connected and Autonomous Vehicles via Proactive Attention) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.08840)] [~~code~~]\n- **V2X-DG** (V2X-DG: Domain Generalization for Vehicle-to-Everything Cooperative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.15435)] [~~code~~]\n\n### IROS 2025\n\n- **CooPre** (CooPre: Cooperative Pretraining for V2X Cooperative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.11241)] [[code](https:\u002F\u002Fgithub.com\u002Fucla-mobility\u002FCooPre)]\n- **CoPAD** (CoPAD : Multi-source Trajectory Fusion and Cooperative Trajectory Prediction with Anchor-Oriented Decoder in V2X Scenarios) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.15984)] [~~code~~]\n- **CRUISE** (CRUISE: Cooperative Reconstruction and Editing in V2X Scenarios using Gaussian Splatting) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.18473)] [[code](https:\u002F\u002Fgithub.com\u002FSainingZhang\u002FCRUISE)]\n\n### CVPR 2024\n\n- **CoHFF** (Collaborative Semantic Occupancy Prediction with Hybrid Feature Fusion in Connected Automated Vehicles) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.07635)] [~~code~~]\n- **CoopDet3D** (TUMTraf V2X Cooperative Perception Dataset) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.01316)] [[code](https:\u002F\u002Fgithub.com\u002Ftum-traffic-dataset\u002Fcoopdet3d)]\n- **CodeFilling** (Communication-Efficient Collaborative Perception via Information Filling with Codebook) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.04966)] [[code](https:\u002F\u002Fgithub.com\u002FPhyllisH\u002FCodeFilling)]\n- **ERMVP** (ERMVP: Communication-Efficient and Collaboration-Robust Multi-Vehicle Perception in Challenging Environments) [[paper](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2024\u002Fhtml\u002FZhang_ERMVP_Communication-Efficient_and_Collaboration-Robust_Multi-Vehicle_Perception_in_Challenging_Environments_CVPR_2024_paper.html)] [[code](https:\u002F\u002Fgithub.com\u002FTerry9a\u002FERMVP)]\n- **MRCNet** (Multi-Agent Collaborative Perception via Motion-Aware Robust Communication Network) [[paper](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2024\u002Fhtml\u002FHong_Multi-agent_Collaborative_Perception_via_Motion-aware_Robust_Communication_Network_CVPR_2024_paper.html)] [[code](https:\u002F\u002Fgithub.com\u002FIndigoChildren\u002Fcollaborative-perception-MRCNet)]\n\n### NeurIPS 2024\n\n- **V2X-Graph** (Learning Cooperative Trajectory Representations for Motion Forecasting) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.00371)] [[code](https:\u002F\u002Fgithub.com\u002FAIR-THU\u002FV2X-Graph)]\n\n### ECCV 2024\n\n- **Hetecooper** (Hetecooper: Feature Collaboration Graph for Heterogeneous Collaborative Perception) [[paper](https:\u002F\u002Feccv.ecva.net\u002Fvirtual\u002F2024\u002Fposter\u002F2467)] [~~code~~]\n- **Infra-Centric CP** (Rethinking the Role of Infrastructure in Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2410.11259)] [~~code~~]\n\n### ICLR 2024\n\n- **HEAL** (An Extensible Framework for Open Heterogeneous Collaborative Perception) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=KkrDUGIASk)] [[code](https:\u002F\u002Fgithub.com\u002Fyifanlu0227\u002FHEAL)]\n\n### AAAI 2024\n\n- **CMiMC** (What Makes Good Collaborative Views? Contrastive Mutual Information Maximization for Multi-Agent Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.10068)] [[code](https:\u002F\u002Fgithub.com\u002F77SWF\u002FCMiMC)]\n- **DI-V2X** (DI-V2X: Learning Domain-Invariant Representation for Vehicle-Infrastructure Collaborative 3D Object Detection) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.15742)] [[code](https:\u002F\u002Fgithub.com\u002FSerenos\u002FDI-V2X)]\n- **V2XFormer** (DeepAccident: A Motion and Accident Prediction Benchmark for V2X Autonomous Driving) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.01168)] [[code](https:\u002F\u002Fgithub.com\u002Ftianqi-wang1996\u002FDeepAccident)]\n\n### WACV 2024\n\n- **MACP** (MACP: Efficient Model Adaptation for Cooperative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.16870)] [[code](https:\u002F\u002Fgithub.com\u002FPurdueDigitalTwin\u002FMACP)]\n\n### ICRA 2024\n\n- **DMSTrack** (Probabilistic 3D Multi-Object Cooperative Tracking for Autonomous Driving via Differentiable Multi-Sensor Kalman Filter) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.14655)] [[code](https:\u002F\u002Fgithub.com\u002Feddyhkchiu\u002FDMSTrack)]\n- **FreeAlign** (Robust Collaborative Perception without External Localization and Clock Devices) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.02965)] [[code](https:\u002F\u002Fgithub.com\u002FMediaBrain-SJTU\u002FFreeAlign)]\n\n### CVPR 2023\n\n- {Related} **BEVHeight** (BEVHeight: A Robust Framework for Vision-Based Roadside 3D Object Detection) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.08498)] [[code](https:\u002F\u002Fgithub.com\u002FADLab-AutoDrive\u002FBEVHeight)]\n- **CoCa3D** (Collaboration Helps Camera Overtake LiDAR in 3D Detection) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.13560)] [[code](https:\u002F\u002Fgithub.com\u002FMediaBrain-SJTU\u002FCoCa3D)]\n- **FF-Tracking** (V2X-Seq: The Large-Scale Sequential Dataset for the Vehicle-Infrastructure Cooperative Perception and Forecasting) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.05938)] [[code](https:\u002F\u002Fgithub.com\u002FAIR-THU\u002FDAIR-V2X-Seq)]\n\n### NeurIPS 2023\n\n- **CoBEVFlow** (Robust Asynchronous Collaborative 3D Detection via Bird's Eye View Flow) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=UHIDdtxmVS)] [[code](https:\u002F\u002Fgithub.com\u002FMediaBrain-SJTU\u002FCoBEVFlow)]\n- **FFNet** (Flow-Based Feature Fusion for Vehicle-Infrastructure Cooperative 3D Object Detection) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=gsglrhvQxX)] [[code](https:\u002F\u002Fgithub.com\u002Fhaibao-yu\u002FFFNet-VIC3D)]\n- **How2comm** (How2comm: Communication-Efficient and Collaboration-Pragmatic Multi-Agent Perception) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=Dbaxm9ujq6)] [[code](https:\u002F\u002Fgithub.com\u002Fydk122024\u002FHow2comm)]\n\n### ICCV 2023\n\n- **CORE** (CORE: Cooperative Reconstruction for Multi-Agent Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.11514)] [[code](https:\u002F\u002Fgithub.com\u002Fzllxot\u002FCORE)]\n- **HM-ViT** (HM-ViT: Hetero-Modal Vehicle-to-Vehicle Cooperative Perception with Vision Transformer) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.10628)] [[code](https:\u002F\u002Fgithub.com\u002FXHwind\u002FHM-ViT)]\n- **ROBOSAC** (Among Us: Adversarially Robust Collaborative Perception by Consensus) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.09495)] [[code](https:\u002F\u002Fgithub.com\u002Fcoperception\u002FROBOSAC)]\n- **SCOPE** (Spatio-Temporal Domain Awareness for Multi-Agent Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.13929)] [[code](https:\u002F\u002Fgithub.com\u002Fstarfdu1418\u002FSCOPE)]\n- **TransIFF** (TransIFF: An Instance-Level Feature Fusion Framework for Vehicle-Infrastructure Cooperative 3D Detection with Transformers) [[paper](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2023\u002Fhtml\u002FChen_TransIFF_An_Instance-Level_Feature_Fusion_Framework_for_Vehicle-Infrastructure_Cooperative_3D_ICCV_2023_paper.html)] [~~code~~]\n- **UMC** (UMC: A Unified Bandwidth-Efficient and Multi-Resolution Based Collaborative Perception Framework) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.12400)] [[code](https:\u002F\u002Fgithub.com\u002Fispc-lab\u002FUMC)]\n\n### ICLR 2023\n\n- {Related} **CO3** (CO3: Cooperative Unsupervised 3D Representation Learning for Autonomous Driving) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=QUaDoIdgo0)] [[code](https:\u002F\u002Fgithub.com\u002FRunjian-Chen\u002FCO3)]\n\n### CoRL 2023\n\n- **BM2CP** {BM2CP: Efficient Collaborative Perception with LiDAR-Camera Modalities} [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=uJqxFjF1xWp)] [[code](https:\u002F\u002Fgithub.com\u002FbyzhaoAI\u002FBM2CP)]\n\n### MM 2023\n\n- **DUSA** (DUSA: Decoupled Unsupervised Sim2Real Adaptation for Vehicle-to-Everything Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.08117)] [[code](https:\u002F\u002Fgithub.com\u002Frefkxh\u002FDUSA)]\n- **FeaCo** (FeaCo: Reaching Robust Feature-Level Consensus in Noisy Pose Conditions) [[paper](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002Fabs\u002F10.1145\u002F3581783.3611880)] [[code](https:\u002F\u002Fgithub.com\u002Fjmgu0212\u002FFeaCo)]\n- **What2comm** (What2comm: Towards Communication-Efficient Collaborative Perception via Feature Decoupling) [[paper](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002Fabs\u002F10.1145\u002F3581783.3611699)] [~~code~~]\n\n### WACV 2023\n\n- **AdaFusion** (Adaptive Feature Fusion for Cooperative Perception Using LiDAR Point Clouds) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.00116)] [[code](https:\u002F\u002Fgithub.com\u002FDonghaoQiao\u002FAdaptive-Feature-Fusion-for-Cooperative-Perception)]\n\n### ICRA 2023\n\n- **CoAlign** (Robust Collaborative 3D Object Detection in Presence of Pose Errors) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.07214)] [[code](https:\u002F\u002Fgithub.com\u002Fyifanlu0227\u002FCoAlign)]\n- {Related} **DMGM** (Deep Masked Graph Matching for Correspondence Identification in Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.07555)] [[code](https:\u002F\u002Fgithub.com\u002Fgaopeng5\u002FDMGM)]\n- **Double-M Quantification** (Uncertainty Quantification of Collaborative Detection for Self-Driving) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.08162)] [[code](https:\u002F\u002Fgithub.com\u002Fcoperception\u002Fdouble-m-quantification)]\n- **MAMP** (Model-Agnostic Multi-Agent Perception Framework) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.13168)] [[code](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002Fmodel_anostic)]\n- **MATE** (Communication-Critical Planning via Multi-Agent Trajectory Exchange) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.06080)] [~~code~~]\n- **MPDA** (Bridging the Domain Gap for Multi-Agent Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2210.08451)] [[code](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002FMPDA)]\n- **WNT** (We Need to Talk: Identifying and Overcoming Communication-Critical Scenarios for Self-Driving) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.04352)] [~~code~~]\n\n### CVPR 2022\n\n- **Coopernaut** (COOPERNAUT: End-to-End Driving with Cooperative Perception for Networked Vehicles) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.02222)] [[code](https:\u002F\u002Fgithub.com\u002FUT-Austin-RPL\u002FCoopernaut)]\n- {Related} **LAV** (Learning from All Vehicles) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.11934)] [[code](https:\u002F\u002Fgithub.com\u002Fdotchen\u002FLAV)]\n- **TCLF** (DAIR-V2X: A Large-Scale Dataset for Vehicle-Infrastructure Cooperative 3D Object Detection) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.05575)] [[code](https:\u002F\u002Fgithub.com\u002FAIR-THU\u002FDAIR-V2X)]\n\n### NeurIPS 2022\n\n- **Where2comm** (Where2comm: Efficient Collaborative Perception via Spatial Confidence Maps) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=dLL4KXzKUpS)] [[code](https:\u002F\u002Fgithub.com\u002FMediaBrain-SJTU\u002Fwhere2comm)]\n\n### ECCV 2022\n\n- **SyncNet** (Latency-Aware Collaborative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2207.08560)] [[code](https:\u002F\u002Fgithub.com\u002FMediaBrain-SJTU\u002FSyncNet)]\n- **V2X-ViT** (V2X-ViT: Vehicle-to-Everything Cooperative Perception with Vision Transformer) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.10638)] [[code](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002Fv2x-vit)]\n\n### CoRL 2022\n\n- **CoBEVT** (CoBEVT: Cooperative Bird's Eye View Semantic Segmentation with Sparse Transformers) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=PAFEQQtDf8s)] [[code](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002FCoBEVT)]\n- **STAR** (Multi-Robot Scene Completion: Towards Task-Agnostic Collaborative Perception) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=hW0tcXOJas2)] [[code](https:\u002F\u002Fgithub.com\u002Fcoperception\u002Fstar)]\n\n### IJCAI 2022\n\n- **IA-RCP** (Robust Collaborative Perception against Communication Interruption) [[paper](https:\u002F\u002Flearn-to-race.org\u002Fworkshop-ai4ad-ijcai2022\u002Fpapers.html)] [~~code~~]\n\n### MM 2022\n\n- **CRCNet** (Complementarity-Enhanced and Redundancy-Minimized Collaboration Network for Multi-agent Perception) [[paper](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002Fabs\u002F10.1145\u002F3503161.3548197)] [~~code~~]\n\n### ICRA 2022\n\n- **AttFuse** (OPV2V: An Open Benchmark Dataset and Fusion Pipeline for Perception with Vehicle-to-Vehicle Communication) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.07644)] [[code](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002FOpenCOOD)]\n- **MP-Pose** (Multi-Robot Collaborative Perception with Graph Neural Networks) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.01760)] [~~code~~]\n\n### NeurIPS 2021\n\n- **DiscoNet** (Learning Distilled Collaboration Graph for Multi-Agent Perception) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=ZRcjSOmYraB)] [[code](https:\u002F\u002Fgithub.com\u002Fai4ce\u002FDiscoNet)]\n\n### ICCV 2021\n\n- **Adversarial V2V** (Adversarial Attacks On Multi-Agent Communication) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2101.06560)] [~~code~~]\n\n### IROS 2021\n\n- **MASH** (Overcoming Obstructions via Bandwidth-Limited Multi-Agent Spatial Handshaking) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2107.00771)] [[code](https:\u002F\u002Fgithub.com\u002Fyifanlu0227\u002FCoAlign)]\n\n### CVPR 2020\n\n- **When2com** (When2com: Multi-Agent Perception via Communication Graph Grouping) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2006.00176)] [[code](https:\u002F\u002Fgithub.com\u002FGT-RIPL\u002FMultiAgentPerception)]\n\n### ECCV 2020\n\n- **DSDNet** (DSDNet: Deep Structured Self-Driving Network) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2008.06041)] [~~code~~]\n- **V2VNet** (V2VNet: Vehicle-to-Vehicle Communication for Joint Perception and Prediction) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2008.07519)] [[code](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002FOpenCOOD)]\n\n### CoRL 2020\n\n- **Robust V2V** (Learning to Communicate and Correct Pose Errors) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.05289)] [[code](https:\u002F\u002Fgithub.com\u002Fyifanlu0227\u002FCoAlign)]\n\n### ICRA 2020\n\n- **Who2com** (Who2com: Collaborative Perception via Learnable Handshake Communication) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.09575)] [[code](https:\u002F\u002Fgithub.com\u002FGT-RIPL\u002FMultiAgentPerception)]\n- **MAIN** (Enhancing Multi-Robot Perception via Learned Data Association) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2107.00769)] [~~code~~]\n\n\n\n## :bookmark:Dataset and Simulator\n\nNote: {Real} denotes that the sensor data is obtained by real-world collection instead of simulation.\n\n### Selected Preprint\n\n- **Adver-City** (Adver-City: Open-Source Multi-Modal Dataset for Collaborative Perception Under Adverse Weather Conditions) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2410.06380)] [[code](https:\u002F\u002Fgithub.com\u002FQUARRG\u002FAdver-City)] [[project](https:\u002F\u002Flabs.cs.queensu.ca\u002Fquarrg\u002Fdatasets\u002Fadver-city)]\n- **AirV2X** (AirV2X: Unified Air-Ground Vehicle-to-Everything Collaboration) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.19283)] [[code](https:\u002F\u002Fgithub.com\u002Ftaco-group\u002FAirV2X-Perception)] [[project](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fxiangbog\u002FAirV2X-Perception)]\n- **CATS-V2V** (CATS-V2V: A Real-World Vehicle-to-Vehicle Cooperative Perception Dataset with Complex Adverse Traffic Scenarios) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.11168)] [~~code~~] [[project](https:\u002F\u002Fcats-v2v-dataset.github.io)]\n- {Real} **CoInfra** (CoInfra: A Large-Scale Cooperative Infrastructure Perception System and Dataset in Adverse Weather) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.02245)] [[code](https:\u002F\u002Fgithub.com\u002FNingMingHao\u002FCoInfra)] [~~project~~]\n- **CP-GuardBench** (CP-Guard+: A New Paradigm for Malicious Agent Detection and Defense in Collaborative Perception) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=9MNzHTSDgh)] [~~code~~] [~~project~~]\n- **Griffin** (Griffin: Aerial-Ground Cooperative Detection and Tracking Dataset and Benchmark) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.06983)] [[code](https:\u002F\u002Fgithub.com\u002Fwang-jh18-SVM\u002FGriffin)] [[project](https:\u002F\u002Fpan.baidu.com\u002Fs\u002F1NDgsuHB-QPRiROV73NRU5g)]\n- {Real} **InScope** (InScope: A New Real-world 3D Infrastructure-side Collaborative Perception Dataset for Open Traffic Scenarios) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.21581)] [[code](https:\u002F\u002Fgithub.com\u002Fxf-zh\u002FInScope)] [~~project~~]\n- **MobileVerse** (MobiVerse: Scaling Urban Mobility Simulation with Hybrid Lightweight Domain-Specific Generator and Large Language Models) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.21784)] [[code](https:\u002F\u002Fgithub.com\u002Fucla-mobility\u002FMobiVerse)] [~~project~~]\n- **Multi-V2X** (Multi-V2X: A Large Scale Multi-modal Multi-penetration-rate Dataset for Cooperative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.04980)] [[code](https:\u002F\u002Fgithub.com\u002FRadetzkyLi\u002FMulti-V2X)] [~~project~~]\n- **M3CAD** (M3CAD: Towards Generic Cooperative Autonomous Driving Benchmark) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.06746)] [[code](https:\u002F\u002Fgithub.com\u002Fzhumorui\u002FM3CAD)] [[project](https:\u002F\u002Fzhumorui.github.io\u002Fm3cad)]\n- **OPV2V-N** (RCDN: Towards Robust Camera-Insensitivity Collaborative Perception via Dynamic Feature-Based 3D Neural Modeling) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.16868)] [~~code~~] [~~project~~]\n- **TalkingVehiclesGym** (Towards Natural Language Communication for Cooperative Autonomous Driving via Self-Play) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.18334)] [[code](https:\u002F\u002Fgithub.com\u002Fcuijiaxun\u002Ftalking-vehicles)] [[project](https:\u002F\u002Ftalking-vehicles.github.io)]\n- **TruckV2X** (TruckV2X: A Truck-Centered Perception Dataset) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.09505)] [~~code~~] [[project](https:\u002F\u002Fxietenghu1.github.io\u002FTruckV2X)]\n- {Real} **UrbanV2X** (UrbanV2X: A Multisensory Vehicle-Infrastructure Dataset for Cooperative Navigation in Urban Areas) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2512.20224)] [[code](https:\u002F\u002Fgithub.com\u002Farclab-hku\u002FEvent_based_VO-VIO-SLAM)] [[project](https:\u002F\u002Fpolyu-taslab.github.io\u002FUrbanV2X)]\n- **V2V-QA** (V2V-LLM: Vehicle-to-Vehicle Cooperative Autonomous Driving with Multi-Modal Large Language Models) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.09980)] [[code](https:\u002F\u002Fgithub.com\u002Feddyhkchiu\u002FV2VLLM)] [[project](https:\u002F\u002Feddyhkchiu.github.io\u002Fv2vllm.github.io)]\n- {Real} **V2XPnP-Seq** (V2XPnP: Vehicle-to-Everything Spatio-Temporal Fusion for Multi-Agent Perception and Prediction) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.01812)] [[code](https:\u002F\u002Fgithub.com\u002FZewei-Zhou\u002FV2XPnP)] [[project](https:\u002F\u002Fmobility-lab.seas.ucla.edu\u002Fv2xpnp)]\n- {Real} **V2X-Radar** (V2X-Radar: A Multi-Modal Dataset with 4D Radar for Cooperative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.10962)] [[code](https:\u002F\u002Fgithub.com\u002Fyanglei18\u002FV2X-Radar)] [[project](http:\u002F\u002Fopenmpd.com\u002Fcolumn\u002FV2X-Radar)]\n- {Real} **V2X-Real** (V2X-Real: a Large-Scale Dataset for Vehicle-to-Everything Cooperative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.16034)] [~~code~~] [[project](https:\u002F\u002Fmobility-lab.seas.ucla.edu\u002Fv2x-real)]\n- {Real} **V2X-ReaLO** (V2X-ReaLO: An Open Online Framework and Dataset for Cooperative Perception in Reality) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.10034)] [~~code~~] [~~project~~]\n- **WHALES** (WHALES: A Multi-Agent Scheduling Dataset for Enhanced Cooperation in Autonomous Driving) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.13340)] [[code](https:\u002F\u002Fgithub.com\u002FchensiweiTHU\u002FWHALES)] [[project](https:\u002F\u002Fpan.baidu.com\u002Fs\u002F1dintX-d1T-m2uACqDlAM9A)]\n\n### CVPR 2025\n\n- **Mono3DVLT-V2X** (Mono3DVLT: Monocular-Video-Based 3D Visual Language Tracking) [[paper](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2025\u002Fhtml\u002FWei_Mono3DVLT_Monocular-Video-Based_3D_Visual_Language_Tracking_CVPR_2025_paper.html)] [~~code~~] [~~project~~]\n- **RCP-Bench** (RCP-Bench: Benchmarking Robustness for Collaborative Perception Under Diverse Corruptions) [[paper](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2025\u002Fhtml\u002FDu_RCP-Bench_Benchmarking_Robustness_for_Collaborative_Perception_Under_Diverse_Corruptions_CVPR_2025_paper.html)] [[code](https:\u002F\u002Fgithub.com\u002FLuckyDush\u002FRCP-Bench)] [~~project~~]\n- **V2X-R** (V2X-R: Cooperative LiDAR-4D Radar Fusion for 3D Object Detection with Denoising Diffusion) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.08402)] [[code](https:\u002F\u002Fgithub.com\u002Fylwhxht\u002FV2X-R)] [~~project~~]\n\n### NeurIPS 2025\n\n- {Real} **AGC-Drive** (AGC-Drive: A Large-Scale Dataset for Real-World Aerial-Ground Collaboration in Driving Scenarios) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=N07WGSPh9l)] [[code](https:\u002F\u002Fgithub.com\u002FPercepX\u002FAGC-Drive)] [[project](https:\u002F\u002Fagc-drive.github.io)]\n- **UrbanIng-V2X** (UrbanIng-V2X: A Large-Scale Multi-Vehicle, Multi-Infrastructure Dataset Across Multiple Intersections for Cooperative Perception) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=iSwIkUqyqf)] [[code](https:\u002F\u002Fgithub.com\u002Fthi-ad\u002FUrbanIng-V2X)] [[project](https:\u002F\u002Fpypi.org\u002Fproject\u002Furbaning)]\n\n### ICCV 2025\n\n- **CoPe-R** (SlimComm: Doppler-Guided Sparse Queries for Bandwidth-Efficient Cooperative 3-D Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.13007)] [[code](https:\u002F\u002Fgithub.com\u002Ffzi-forschungszentrum-informatik\u002FSlimComm)] [~~project~~]\n- {Real} **Mixed Signals** (Mixed Signals: A Diverse Point Cloud Dataset for Heterogeneous LiDAR V2X Collaboration) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.14156)] [[code](https:\u002F\u002Fgithub.com\u002Fchinitaberrio\u002FMixed-Signals)] [[project](https:\u002F\u002Fmixedsignalsdataset.cs.cornell.edu)]\n\n### CVPR 2024\n\n- {Real} **HoloVIC** (HoloVIC: Large-Scale Dataset and Benchmark for Multi-Sensor Holographic Intersection and Vehicle-Infrastructure Cooperative) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.02640)] [~~code~~] [[project](https:\u002F\u002Fholovic.net)]\n- {Real} **Open Mars Dataset** (Multiagent Multitraversal Multimodal Self-Driving: Open MARS Dataset) [[code](https:\u002F\u002Fgithub.com\u002Fai4ce\u002FMARS)] [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.09383)] [[project](https:\u002F\u002Fai4ce.github.io\u002FMARS)]\n- {Real} **RCooper** (RCooper: A Real-World Large-Scale Dataset for Roadside Cooperative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.10145)] [[code](https:\u002F\u002Fgithub.com\u002FAIR-THU\u002FDAIR-RCooper)] [[project](https:\u002F\u002Fwww.t3caic.com\u002Fqingzhen)]\n- {Real} **TUMTraf-V2X** (TUMTraf V2X Cooperative Perception Dataset) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.01316)] [[code](https:\u002F\u002Fgithub.com\u002Ftum-traffic-dataset\u002Ftum-traffic-dataset-dev-kit)] [[project](https:\u002F\u002Ftum-traffic-dataset.github.io\u002Ftumtraf-v2x)]\n\n### NeurIPS 2024\n\n- {Real} **DAIR-V2X-Traj** (Learning Cooperative Trajectory Representations for Motion Forecasting) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.00371)] [[code](https:\u002F\u002Fgithub.com\u002FAIR-THU\u002FV2X-Graph)] [[project](https:\u002F\u002Fthudair.baai.ac.cn\u002Findex)]\n\n### ECCV 2024\n\n- {Real} **H-V2X** (H-V2X: A Large Scale Highway Dataset for BEV Perception) [[paper](https:\u002F\u002Feccv2024.ecva.net\u002Fvirtual\u002F2024\u002Fposter\u002F126)] [~~code~~] [~~project~~]\n\n### ICLR 2024\n\n- **OPV2V-H** (An Extensible Framework for Open Heterogeneous Collaborative Perception) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=KkrDUGIASk)] [[code](https:\u002F\u002Fgithub.com\u002Fyifanlu0227\u002FHEAL)] [[project](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fyifanlu\u002FOPV2V-H)]\n\n### AAAI 2024\n\n- **DeepAccident** (DeepAccident: A Motion and Accident Prediction Benchmark for V2X Autonomous Driving) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.01168)] [[code](https:\u002F\u002Fgithub.com\u002Ftianqi-wang1996\u002FDeepAccident)] [[project](https:\u002F\u002Fdeepaccident.github.io)]\n\n### CVPR 2023\n\n- **CoPerception-UAV+** (Collaboration Helps Camera Overtake LiDAR in 3D Detection) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.13560)] [[code](https:\u002F\u002Fgithub.com\u002FMediaBrain-SJTU\u002FCoCa3D)] [[project](https:\u002F\u002Fsiheng-chen.github.io\u002Fdataset\u002FCoPerception+)]\n- **OPV2V+** (Collaboration Helps Camera Overtake LiDAR in 3D Detection) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.13560)] [[code](https:\u002F\u002Fgithub.com\u002FMediaBrain-SJTU\u002FCoCa3D)] [[project](https:\u002F\u002Fsiheng-chen.github.io\u002Fdataset\u002FCoPerception+)]\n- {Real} **V2V4Real** (V2V4Real: A Large-Scale Real-World Dataset for Vehicle-to-Vehicle Cooperative Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.07601)] [[code](https:\u002F\u002Fgithub.com\u002Fucla-mobility\u002FV2V4Real)] [[project](https:\u002F\u002Fmobility-lab.seas.ucla.edu\u002Fv2v4real)]\n- {Real} **DAIR-V2X-Seq** (V2X-Seq: The Large-Scale Sequential Dataset for the Vehicle-Infrastructure Cooperative Perception and Forecasting) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.05938)] [[code](https:\u002F\u002Fgithub.com\u002FAIR-THU\u002FDAIR-V2X-Seq)] [[project](https:\u002F\u002Fthudair.baai.ac.cn\u002Findex)]\n\n### NeurIPS 2023\n\n- **IRV2V** (Robust Asynchronous Collaborative 3D Detection via Bird's Eye View Flow) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=UHIDdtxmVS)] [~~code~~] [~~project~~]\n\n### ICCV 2023\n\n- **Roadside-Opt** (Optimizing the Placement of Roadside LiDARs for Autonomous Driving) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.07247)] [~~code~~] [~~project~~]\n\n### ICRA 2023\n\n- {Real} **DAIR-V2X-C Complemented** (Robust Collaborative 3D Object Detection in Presence of Pose Errors) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.07214)] [[code](https:\u002F\u002Fgithub.com\u002Fyifanlu0227\u002FCoAlign)] [[project](https:\u002F\u002Fsiheng-chen.github.io\u002Fdataset\u002Fdair-v2x-c-complemented)]\n- **RLS** (Analyzing Infrastructure LiDAR Placement with Realistic LiDAR Simulation Library) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.15975)] [[code](https:\u002F\u002Fgithub.com\u002FPJLab-ADG\u002FLiDARSimLib-and-Placement-Evaluation)] [~~project~~]\n- **V2XP-ASG** (V2XP-ASG: Generating Adversarial Scenes for Vehicle-to-Everything Perception) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.13679)] [[code](https:\u002F\u002Fgithub.com\u002FXHwind\u002FV2XP-ASG)] [~~project~~]\n\n### CVPR 2022\n\n- **AutoCastSim** (COOPERNAUT: End-to-End Driving with Cooperative Perception for Networked Vehicles) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.02222)] [[code](https:\u002F\u002Fgithub.com\u002Fhangqiu\u002FAutoCastSim)] [[project](https:\u002F\u002Futexas.app.box.com\u002Fv\u002Fcoopernaut-dataset)]\n- {Real} **DAIR-V2X** (DAIR-V2X: A Large-Scale Dataset for Vehicle-Infrastructure Cooperative 3D Object Detection) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.05575)] [[code](https:\u002F\u002Fgithub.com\u002FAIR-THU\u002FDAIR-V2X)] [[project](https:\u002F\u002Fthudair.baai.ac.cn\u002Findex)]\n\n### NeurIPS 2022\n\n- **CoPerception-UAV** (Where2comm: Efficient Collaborative Perception via Spatial Confidence Maps) [[paper&review](https:\u002F\u002Fopenreview.net\u002Fforum?id=dLL4KXzKUpS)] [[code](https:\u002F\u002Fgithub.com\u002FMediaBrain-SJTU\u002Fwhere2comm)] [[project](https:\u002F\u002Fsiheng-chen.github.io\u002Fdataset\u002Fcoperception-uav)]\n\n### ECCV 2022\n\n- **V2XSet** (V2X-ViT: Vehicle-to-Everything Cooperative Perception with Vision Transformer) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.10638)] [[code](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002Fv2x-vit)] [[project](https:\u002F\u002Fdrive.google.com\u002Fdrive\u002Ffolders\u002F1r5sPiBEvo8Xby-nMaWUTnJIPK6WhY1B6)]\n\n### ICRA 2022\n\n- **OPV2V** (OPV2V: An Open Benchmark Dataset and Fusion Pipeline for Perception with Vehicle-to-Vehicle Communication) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.07644)] [[code](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002FOpenCOOD)] [[project](https:\u002F\u002Fmobility-lab.seas.ucla.edu\u002Fopv2v)]\n\n### ACCV 2022\n\n- **DOLPHINS** (DOLPHINS: Dataset for Collaborative Perception Enabled Harmonious and Interconnected Self-Driving) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2207.07609)] [[code](https:\u002F\u002Fgithub.com\u002Fexplosion5\u002FDolphins)] [[project](https:\u002F\u002Fdolphins-dataset.net)]\n\n### ICCV 2021\n\n- **V2X-Sim** (V2X-Sim: Multi-Agent Collaborative Perception Dataset and Benchmark for Autonomous Driving) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2202.08449)] [[code](https:\u002F\u002Fgithub.com\u002Fai4ce\u002FV2X-Sim)] [[project](https:\u002F\u002Fai4ce.github.io\u002FV2X-Sim)]\n\n### CoRL 2017\n\n- **CARLA** (CARLA: An Open Urban Driving Simulator) [[paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F1711.03938)] [[code](https:\u002F\u002Fgithub.com\u002Fcarla-simulator\u002Fcarla)] [[project](https:\u002F\u002Fcarla.org)]\n","# 协作感知\n\n本仓库是对**V2I** \u002F **V2V** \u002F **V2X** 自动驾驶场景下**协作** \u002F **合作** \u002F **多智能体** 感知领域近期进展的论文摘要。论文按首字母顺序排列。\n\n### :link:跳转至：\n- ### [[方法与框架](https:\u002F\u002Fgithub.com\u002FLittle-Podi\u002FCollaborative_Perception#bookmarkmethod-and-framework)]\n- ### [[数据集和仿真器](https:\u002F\u002Fgithub.com\u002FLittle-Podi\u002FCollaborative_Perception#bookmarkdataset-and-simulator)]\n\n注：由于部分已发表的结果是在未明确训练和测试设置的情况下获得的，甚至使用了修改后的模型架构，因此很难在每个基准上公平地比较所有方法。事实上，许多研究都是在自己的设置下评估所有基线并报告结果。因此，不同论文之间可能会存在不一致之处。基于此，我在之前的更新中放弃了收集和复现所有基准的工作。如果你感兴趣，可以在[这个归档版本](https:\u002F\u002Fgithub.com\u002FLittle-Podi\u002FCollaborative_Perception\u002Ftree\u002F1be25908aea0a9f635ff4852b3a90729cf2b6aac)中找到大量相关结果。\n\n\n\n## :star2:推荐\n\n### 有用的学习资源:thumbsup::thumbsup::thumbsup:\n\n- **(论文)** 当自动驾驶汽车遇上V2X协同感知：我们离目标还有多远？[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.24927)]，端到端V2X协同自动驾驶竞赛中的研究挑战与进展[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.21610)]，自动驾驶车辆应与自然语言连接[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.01059)]，面向自动驾驶的协同感知数据集：综述[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.12696)]，面向网联自动驾驶的协同感知：挑战、可能解决方案及机遇[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.01544)]，用于自动驾驶的V2X协同感知：最新进展与挑战[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.03525)]，迈向车路协同自动驾驶：关于协同感知的综述[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.16714)]，自动驾驶中的协同感知：方法、数据集与挑战[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2301.06262)]，合作感知的综述与框架：从异构单体到层次化合作[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.10590)]\n- **(讲座)** 车对车通信（Waabi CVPR 24 自动驾驶教程）[[视频](https:\u002F\u002Fyoutu.be\u002FyceuUthWz9s)]，车对车通信（Waabi CVPR 23 自动驾驶教程）[[视频](https:\u002F\u002Fyoutu.be\u002FT-N51B8mZB8)]，L4自动驾驶的终极解决方案[[视频](https:\u002F\u002Fyoutu.be\u002FcyNxemm4Ujg)]，当视觉Transformer遇上协同感知[[视频](https:\u002F\u002Fyoutu.be\u002FrLAU4eqoOIU)]，超越可见范围的场景理解[[视频](https:\u002F\u002Fyoutu.be\u002Foz0AnmJZCR4)]，抗通信中断的鲁棒协同感知[[视频](https:\u002F\u002Fyoutu.be\u002F3cIWpMrsyeE)]，自动驾驶中的协同与对抗性3D感知[[视频](https:\u002F\u002Fyoutu.be\u002FW-AONQMfGi0)]，自动驾驶中的车对车通信[[视频](https:\u002F\u002Fyoutu.be\u002FoikdOpmIoc4)]，自动驾驶的对抗鲁棒性[[视频](https:\u002F\u002Fyoutu.be\u002F8uBFXzyII5Y)]，L4感知系统的终极形态：协同驾驶[[视频](https:\u002F\u002Fyoutu.be\u002FNvixMEDHht4)]，CoBEVFlow——解决车-车\u002F路协同感知的时序异步问题[[视频](https:\u002F\u002Fyoutu.be\u002FIBTgalAjye8)]，新一代协作感知Where2comm将通信带宽降低十万倍[[视频](https:\u002F\u002Fyoutu.be\u002Fi5coMk4hkuk)]，从任务相关到任务无关的多机器人协同感知[[视频](https:\u002F\u002Fcourse.zhidx.com\u002Fc\u002FMDlkZjcyZDgwZWI4ODBhOGQ4MzM=)]，协同自动驾驶：仿真与感知[[视频](https:\u002F\u002Fcourse.zhidx.com\u002Fc\u002FMmQ1YWUyMzM1M2I3YzVlZjE1NzM=)]，基于群体协作的超视距态势感知[[视频](https:\u002F\u002Fwww.koushare.com\u002Fvideo\u002Fvideodetail\u002F33015)]\n- **(库)** V2Xverse：基于V2X的端到端协同自动驾驶代码库[[代码](https:\u002F\u002Fgithub.com\u002FCollaborativePerception\u002FV2Xverse)] [[文档](https:\u002F\u002Fcollaborativeperception.github.io\u002FV2Xverse)]，HEAL：开放异构协同感知的可扩展框架[[代码](https:\u002F\u002Fgithub.com\u002Fyifanlu0227\u002FHEAL)] [[文档](https:\u002F\u002Fhuggingface.co\u002Fyifanlu\u002FHEAL)]，OpenCOOD：面向自动驾驶的开放合作检测框架[[代码](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002FOpenCOOD)] [[文档](https:\u002F\u002Fopencood.readthedocs.io\u002Fen\u002Flatest\u002Findex.html)]，CoPerception：协同感知SDK[[代码](https:\u002F\u002Fgithub.com\u002Fcoperception\u002Fcoperception)] [[文档](https:\u002F\u002Fcoperception.readthedocs.io\u002Fen\u002Flatest)]，OpenCDA：集成原型合作驾驶自动化功能的仿真工具[[代码](https:\u002F\u002Fgithub.com\u002Fucla-mobility\u002FOpenCDA)] [[文档](https:\u002F\u002Fopencda-documentation.readthedocs.io\u002Fen\u002Flatest)]\n- **(研讨会)** Co-Intelligence@ECCV'24 [[网站](https:\u002F\u002Fcoop-intelligence.github.io)]，CoPerception@ICRA'23 [[网站](https:\u002F\u002Fcoperception.github.io)]，ScalableAD@ICRA'23 [[网站](https:\u002F\u002Fsites.google.com\u002Fview\u002Ficra2023av\u002Fhome)]\n- **(背景知识)** 智能体中点云目标检测的当前方法与未来方向[[视频](https:\u002F\u002Fyoutu.be\u002FxFFCQVwYeec)]，自动驾驶中的3D目标检测：综述与新视角[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.09474)]，DACOM：为多智能体强化学习学习延迟感知通信[[视频](https:\u002F\u002Fyoutu.be\u002FYBgW2oA_n3k)]，带有通信的多智能体强化学习综述[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.08975)]\n\n### 典型的合作模式:handshake::handshake::handshake:\n\n![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FLittle-Podi_Collaborative_Perception_readme_94c6af95b85a.png)\n\n### 可能的优化方向:fire::fire::fire:\n\n![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FLittle-Podi_Collaborative_Perception_readme_19d07569c559.png)\n\n\n\n## :bookmark:方法与框架\n\n注：{Related} 表示该论文并非纯粹的协同感知论文，但包含相关内容。\n\n### 精选预印本\n\n- **ACCO**（协作感知只需离散化融合吗？）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.13946)] [[代码](https:\u002F\u002Fgithub.com\u002Fsidiangongyuan\u002FACCO)]\n- **AR2VP**（基于路侧到车端视觉的动态V2X自主感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.19113)] [[代码](https:\u002F\u002Fgithub.com\u002Ftjy1423317192\u002FAP2VP)]\n- **CPPC**（点簇：一种用于通信高效协作感知的紧凑消息单元）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=54XlM8Clkg)] [~~代码~~]\n- **CP-FREEZER**（CP-FREEZER：针对车载协作感知的时延攻击）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.01062)] [[代码](https:\u002F\u002Fgithub.com\u002FWiSeR-Lab\u002FCP-FREEZER)]\n- **CMP**（CMP：基于多智能体通信的协作运动预测）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.17916)] [~~代码~~]\n- **CoBEVFusion**（CoBEVFusion：激光雷达与摄像头鸟瞰图融合的协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.06008)] [~~代码~~]\n- **CoBEVGlue**（自定位协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.12712)] [[代码](https:\u002F\u002Fgithub.com\u002FVincentNi0107\u002FCoBEVGlue)]\n- **CoCMT**（CoCMT：迈向通信高效的跨模态Transformer协作感知）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=S1NrbfMS7T)] [[代码](https:\u002F\u002Fgithub.com\u002Ftaco-group\u002FCOCMT)]\n- **CoDiff**（CoDiff：用于协作3D目标检测的条件扩散模型）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.14891)] [~~代码~~]\n- **CoDriving**（迈向协作式自动驾驶：仿真平台与端到端系统）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.09496)] [[代码](https:\u002F\u002Fgithub.com\u002FCollaborativePerception\u002FV2Xverse)]\n- **CoDrivingLLM**（迈向交互式可学习的协作驾驶自动化：一个大语言模型驱动的决策框架）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.12812)] [[代码](https:\u002F\u002Fgithub.com\u002FFanGShiYuu\u002FCoDrivingLLM)]\n- **CollaMamba**（CollaMamba：基于跨智能体时空状态空间模型的高效协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.07714)] [~~代码~~]\n- **CoLMDriver**（CoLMDriver：基于LLM的协商助力协作式自动驾驶）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.08683)] [[代码](https:\u002F\u002Fgithub.com\u002Fcxliu0314\u002FCoLMDriver)]\n- **CoMamba**（CoMamba：利用状态空间模型解锁实时协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.10699)] [~~代码~~]\n- **CoPLOT**（超越BEV：优化点级标记以实现协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.19638)] [[代码](https:\u002F\u002Fgithub.com\u002FCheeryLeeyy\u002FCoPLOT)]\n- **CP-Guard+**（CP-Guard+：协作感知中恶意智能体检测与防御的新范式）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=9MNzHTSDgh)] [~~代码~~]\n- **CTCE**（利用时间上下文增强车路协同感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.10531)] [~~代码~~]\n- **Debrief**（对话车辆：通过自然语言实现协作驾驶）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=VYlfoA8I6A)] [~~代码~~]\n- **DeepFleet**（DeepFleet：面向移动机器人的多智能体基础模型）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.08574)] [~~代码~~]\n- **DiffCP**（DiffCP：基于扩散模型的超低比特协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.19592)] [~~代码~~]\n- **FadeLead**（背景淡化，前景主导：课程引导的背景剪枝以实现高效的以前景为中心的协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.19250)] [~~代码~~]\n- **Faster-HEAL**（Faster-HEAL：一种高效且隐私友好的异构自动驾驶协作感知框架）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.07314)] [~~代码~~]\n- **FlowAdapt**（移动关键信息：通过最优传输流实现参数高效的领域适应以支持协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.11565)] [~~代码~~]\n- **HeatV2X**（HeatV2X：通过高效对齐与交互实现可扩展的异构协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.10211)] [[代码](https:\u002F\u002Fgithub.com\u002Fhollowknight2167\u002FHeatV2X)]\n- **HyComm**（基于混合协作的通信高效多智能体3D检测）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.07092)] [~~代码~~]\n- **HyDRA**（HyDRA：面向异构协作感知的混合领域感知鲁棒架构）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.23975)] [~~代码~~]\n- **InSPE**（InSPE：快速评估异构多模态基础设施传感器布局）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.08240)] [~~代码~~]\n- **I2XTraj**（面向基础设施到万物的信号交叉口知识驱动多智能体轨迹预测）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.13461)] [~~代码~~]\n- **JigsawComm**（JigsawComm：面向通信高效协作感知的联合语义特征编码与传输）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.17843)] [[代码](https:\u002F\u002Fgithub.com\u002FWiSeR-Lab\u002FJigsawComm)]\n- **LCV2I**（LCV2I：基于低分辨率激光雷达的通信高效、高性能协作感知框架）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.17039)] [~~代码~~]\n- **LMMCoDrive**（LMMCoDrive：基于大型多模态模型的协作驾驶）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.11981)] [[代码](https:\u002F\u002Fgithub.com\u002Fhenryhcliu\u002FLMMCoDrive)]\n- {相关} **MDG**（MDG：用于交通环境中多智能体行为建模的掩码去噪生成）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.17496)] [~~代码~~]\n- **mmCooper**（mmCooper：一个多智能体多阶段、通信高效且协作鲁棒的协作感知框架）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.12263)] [~~代码~~]\n- **MOT-CUP**（具有保形不确定性传播的协作多目标跟踪）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.14346)] [[代码](https:\u002F\u002Fgithub.com\u002Fsusanbao\u002Fmot_cup)]\n- **OpenCOOD-Air**（OpenCOOD-Air：通过空间转换和偏移预测促进异构地面-空中协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.13919)] [~~代码~~]\n- **RefPtsFusion**（从特征到参考点：轻量级且适应性强的融合以支持协作式自动驾驶）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.18757)] [~~代码~~]\n- {相关} **RopeBEV**（RopeBEV：鸟瞰视角下的多摄像头路侧感知网络）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.11706)] [~~代码~~]\n- **ParCon**（ParCon：基于多模块并行连接的抗噪声协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.11546)] [~~代码~~]\n- **PragComm**（多智能体协作感知中的务实通信）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.12694)] [[代码](https:\u002F\u002Fgithub.com\u002FPhyllisH\u002FPragComm)]\n- **QPoint2Comm**（少发多知：用于容错协作感知的掩码量化点云通信）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.21667)] [~~代码~~]\n- **QuantV2X**（QuantV2X：一个完全量化的多智能体协作感知系统）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.03704)] [[代码](https:\u002F\u002Fgithub.com\u002Fucla-mobility\u002FQuantV2X)]\n- **QUEST**（QUEST：用于车路协同感知的查询流）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.01804)] [~~代码~~]\n- **RCDN**（RCDN：通过基于动态特征的3D神经建模实现对相机不敏感的稳健协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.16868)] [~~代码~~]\n- **RC-GeoCP**（RC-GeoCP：用于雷达-摄像头协作感知的几何一致性）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.00654)] [~~代码~~]\n- **ReVQom**（用于通信高效多智能体感知的残差向量量化）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.21464)] [~~代码~~]\n- **RG-Attn**（RG-Attn：用于多模态多智能体协作感知的弧度粘合注意力）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.16803)] [~~代码~~]\n- **RiskMM**（风险地图作为中间件：迈向可解释的协作式端到端自动驾驶，实现风险感知规划）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.07686)] [~~代码~~]\n- **RoCo-Sim**（RoCo-Sim：通过前景模拟增强路侧协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.10410)] [[代码](https:\u002F\u002Fgithub.com\u002Fduyuwen-duen\u002FRoCo-Sim)]\n- **SafeCoop**（SafeCoop：解开代理式协作驾驶中的全栈安全性）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.18123)] [[代码](https:\u002F\u002Fgithub.com\u002Ftaco-group\u002FSafeCoop)]\n- **ShareVerse**（ShareVerse：用于共享世界建模的多智能体一致视频生成）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.02697)] [~~代码~~]\n- **SiCP**（SiCP：面向网联与自动驾驶车辆3D目标检测的同时个体与协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.04822)] [[代码](https:\u002F\u002Fgithub.com\u002FDarrenQu\u002FSiCP)]\n- **SparseAlign**（SparseAlign：一个完全稀疏的协作目标检测框架）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.12982)] [~~代码~~]\n- **Talking Vehicles**（通过自我博弈实现自然语言通信，迈向协作式自动驾驶）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.18334)] [[代码](https:\u002F\u002Fgithub.com\u002Fcuijiaxun\u002Ftalking-vehicles)]\n- **TOCOM-V2I**（面向任务的车路协同感知通信）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.20748)] [~~代码~~]\n- {相关} **TYP**（转移你的视角：在驾驶场景中从任意视角可控地生成3D内容）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.06682)] [~~代码~~]\n- **UniMM-V2X**（UniMM-V2X：MoE增强的多层级融合，用于端到端协作式自动驾驶）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.09013)] [[代码](https:\u002F\u002Fgithub.com\u002FSouig\u002FUniMM-V2X)]\n- **VIMI**（VIMI：基于摄像头的3D目标检测中车路多视角中间融合）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.10975)] [[代码](https:\u002F\u002Fgithub.com\u002FBosszhe\u002FVIMI)]\n- **VLIF**（是否只需中间融合即可实现基于无人机的协作感知？）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.21774)] [[代码](https:\u002F\u002Fgithub.com\u002Fuestchjw\u002FLIF)]\n- **V2V-GoT**（V2V-GoT：基于多模态大语言模型和思维之图的车对车协作式自动驾驶）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.18053)] [~~代码~~]\n- **V2V-LLM**（V2V-LLM：基于多模态大语言模型的车对车协作式自动驾驶）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.09980)] [[代码](https:\u002F\u002Fgithub.com\u002Feddyhkchiu\u002FV2VLLM)]\n- **V2XPnP**（V2XPnP：面向多智能体感知与预测的车到万物时空融合）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.01812)] [[代码](https:\u002F\u002Fgithub.com\u002FZewei-Zhou\u002FV2XPnP)]\n- **V2X-DGPE**（V2X-DGPE：解决领域差距与姿态误差以实现稳健的协作3D目标检测）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.02363)] [[代码](https:\u002F\u002Fgithub.com\u002Fwangsch10\u002FV2X-DGPE)]\n- **V2X-DGW**（V2X-DGW：恶劣天气条件下多智能体感知的领域泛化）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.11371)] [~~代码~~]\n- **V2X-DSC**（V2X-DSC：基于分布式源编码指导通信的多智能体协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.00687)] [~~代码~~]\n- **V2X-M2C**（V2X-M2C：两种连接方式的高效多模块协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.11546)] [~~代码~~]\n- **V2X-PC**（V2X-PC：基于点簇的车到万物协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.16635)] [~~代码~~]\n- **V2X-REALM**（V2X-REALM：基于视觉-语言模型的稳健端到端协作式自动驾驶，具备自适应长尾建模能力）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.21041)] [~~代码~~]\n- **V2X-RECT**（V2X-RECT：通过冗余交互过滤与跟踪误差校正实现高效的V2X轨迹预测框架）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.17941)] [~~代码~~]\n- **V2X-ReaLO**（V2X-ReaLO：一个开放的在线框架与数据集，用于现实中的协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.10034)] [~~代码~~]\n- **V2X-UniPool**（V2X-UniPool：统一多模态感知与知识推理以支持自动驾驶）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.02580)] [[代码](https:\u002F\u002Fgithub.com\u002Fsnowwhite1016\u002FV2X-UniPool)]\n- **V2X-VLM**（V2X-VLM：通过大型视觉-语言模型实现端到端V2X协作式自动驾驶）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.09251)] [~~代码~~]\n\n### CVPR 2026\n\n- **CATNet**（CATNet：用于协作感知的协同对齐与变换网络）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.05255)] [~~代码~~]\n- **CodeAlign**（异构协作感知中的模态隔离链接）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.00609)] [[代码](https:\u002F\u002Fgithub.com\u002Fcxliu0314\u002FCodeAlign)]\n- **CoLC**（CoLC：基于激光雷达补全的通信高效协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.00682)] [[代码](https:\u002F\u002Fgithub.com\u002FCatOneTwo\u002FCoLC)]\n- **CoopDiff**（CoopDiff：一种受扩散模型引导的抗干扰协作方法）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.01688)] [~~代码~~]\n- **MVIG**（用于自适应对抗性协作感知的互视信息图学习）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.19596)] [[代码](https:\u002F\u002Fgithub.com\u002Fyihangtao\u002FMVIG)]\n- **WhisperNet**（WhisperNet：一种可扩展的带宽高效协作方案）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.01708)] [~~代码~~]\n\n### ICLR 2026\n\n- **RDComm**（面向协作感知的率失真优化通信）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=920RxFvsMx)] [~~代码~~]\n- **SiMO**（SiMO：单模态操作的多模态协作感知）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=h0iRgjTmVs)] [[代码](https:\u002F\u002Fgithub.com\u002Fdempsey-wen\u002FSiMO)]\n\n### AAAI 2026\n\n- **InfoCom**（InfoCom：基于信息瓶颈的千字节级通信高效协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2512.10305)] [[代码](https:\u002F\u002Fgithub.com\u002Ffengxueguiren\u002FInfoCom)]\n- **SparseCoop**（SparseCoop：基于运动学约束查询的协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2512.06838)] [[代码](https:\u002F\u002Fgithub.com\u002Fwang-jh18-SVM\u002FSparseCoop)]\n- **V2VLoc**（V2VLoc：通过激光雷达定位实现的鲁棒无GNSS协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.14247)] [[代码](https:\u002F\u002Fgithub.com\u002Fwklin214-glitch\u002FV2VLoc)]\n\n# ICRA 2026\n\n- **EIMC**（EIMC：高效的实例感知多模态协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.02532)] [[代码](https:\u002F\u002Fgithub.com\u002Fsidiangongyuan\u002FEIMC)]\n- **WaveComm**（WaveComm：基于小波特征蒸馏的协作感知轻量级通信）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.13365)] [[代码](https:\u002F\u002Fgithub.com\u002Ferdemtbao\u002FWaveComm)]\n\n### WACV 2026\n\n- **FocalComm**（FocalComm：硬实例感知的多智能体感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2512.13982)] [[代码](https:\u002F\u002Fgithub.com\u002Fscdrand23\u002FFocalComm)]\n\n### CVPR 2025\n\n- **CoGMP**（用于协作BEV语义分割的生成式地图先验）[[论文](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2025\u002Fhtml\u002FFu_Generative_Map_Priors_for_Collaborative_BEV_Semantic_Segmentation_CVPR_2025_paper.html)] [~~代码~~]\n- **CoSDH**（CoSDH：基于供需感知与中后期混合的通信高效协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.03430)] [[代码](https:\u002F\u002Fgithub.com\u002FXu2729\u002FCoSDH)]\n- **HeCoFuse**（HeCoFuse：基于异构传感器的跨模态互补V2X协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.13677)] [[代码](https:\u002F\u002Fgithub.com\u002FChuhengWei\u002FHeCoFuse)]\n- **LangCoop**（LangCoop：基于语言的协作驾驶）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.13406)] [[代码](https:\u002F\u002Fgithub.com\u002Ftaco-group\u002FLangCoop)]\n- **PolyInter**（一即足够：用于不可变异构协作感知的多态特征解释器）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.16799)] [[代码](https:\u002F\u002Fgithub.com\u002Fyuchen-xia\u002FPolyInter)]\n- **SparseAlign**（SparseAlign：用于协作目标检测的全稀疏框架）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.12982)] [~~代码~~]\n- **TraF-Align**（TraF-Align：面向异步多智能体感知的轨迹感知特征对齐）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.19391)] [[代码](https:\u002F\u002Fgithub.com\u002FzhyingS\u002FTraF-Align)]\n- **V2X-R**（V2X-R：基于去噪扩散的3D目标检测用协作激光雷达-4D雷达融合）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.08402)] [[代码](https:\u002F\u002Fgithub.com\u002Fylwhxht\u002FV2X-R)]\n\n### NeurIPS 2025\n\n- **GenComm**（基于生成式通信机制的务实异构协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.19618)] [[代码](https:\u002F\u002Fgithub.com\u002Fjeffreychou777\u002FGenComm)]\n- **NegoCollab**（NegoCollab：一种用于异构协作感知的共同表示协商方法）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.27647)] [[代码](https:\u002F\u002Fgithub.com\u002Fscz023\u002FNegoCollab)]\n\n### ICCV 2025\n\n- **CoopTrack**（CoopTrack：探索端到端学习以实现高效的协作序列感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.19239)] [[代码](https:\u002F\u002Fgithub.com\u002Fzhongjiaru\u002FCoopTrack)]\n- **CoST**（CoST：从统一时空视角出发的高效协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.00359)] [[代码](https:\u002F\u002Fgithub.com\u002Ftzhhhh123\u002FCoST)]\n- **INSTINCT**（INSTINCT：基于查询的协作感知的实例级交互架构）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.23700)] [[代码](https:\u002F\u002Fgithub.com\u002FCrazyShout\u002FINSTINCT)]\n- **MamV2XCalib**（MamV2XCalib：基于V2X的无靶标基础设施相机标定，采用状态空间模型）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.23595)] [[代码](https:\u002F\u002Fgithub.com\u002Fzhuyaoye\u002FMamV2XCalib)]\n- **SlimComm**（SlimComm：基于多普勒引导的稀疏查询，用于带宽高效的3D协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.13007)] [[代码](https:\u002F\u002Fgithub.com\u002Ffzi-forschungszentrum-informatik\u002FSlimComm)]\n- **TurboTrain**（TurboTrain：迈向多智能体感知与预测的高效均衡多任务学习）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.04682)] [[代码](https:\u002F\u002Fgithub.com\u002Fucla-mobility\u002FTurboTrain)]\n\n### ICLR 2025\n\n- **CPPC**（点簇：用于通信高效协作感知的紧凑消息单元）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=54XlM8Clkg)] [~~代码~~]\n- **R&B-POP**（从他人的预测中学习3D感知）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=Ylk98vWQuQ)] [[代码](https:\u002F\u002Fgithub.com\u002Fjinsuyoo\u002Frnb-pop)]\n- **STAMP**（STAMP：可扩展的任务与模型无关的协作感知）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=8NdNniulYE)] [[代码](https:\u002F\u002Fgithub.com\u002Ftaco-group\u002FSTAMP)]\n\n### AAAI 2025\n\n- **CoPEFT**（CoPEFT：基于参数高效微调的多智能体协同感知快速适应框架）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.10705)] [[代码](https:\u002F\u002Fgithub.com\u002Ffengxueguiren\u002FCoPEFT)]\n- **CP-Guard**（CP-Guard：协同鸟瞰感知中的恶意智能体检测与防御）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.12000)] [~~代码~~]\n- **DSRC**（DSRC：学习抗损坏的密度无关语义感知协同表示）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.10739)] [[代码](https:\u002F\u002Fgithub.com\u002FTerry9a\u002FDSRC)]\n- **UniV2X**（通过V2X协作实现端到端自动驾驶）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.00717)] [[代码](https:\u002F\u002Fgithub.com\u002FAIR-THU\u002FUniV2X)]\n\n### MM 2025\n\n- **How2Compress**（How2Compress：通过自适应粒度视频压缩实现可扩展且高效的边缘视频分析）[[论文](https:\u002F\u002Fwyhallenwu.github.io\u002Fassets\u002Fpdf\u002Fpaper_archive\u002Fhow2compress.pdf)] [[代码](https:\u002F\u002Fgithub.com\u002Fwyhallenwu\u002Fhow2compress)]\n- **Selective Shift**（Selective Shift：迈向多智能体协同感知中的个性化领域适应）[~~论文~~] [~~代码~~]\n\n### ICRA 2025\n\n- **CoDynTrust**（CoDynTrust：基于动态特征信任模量的鲁棒异步协同感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.08169)] [[代码](https:\u002F\u002Fgithub.com\u002FCrazyShout\u002FCoDynTrust)]\n- **CoopDETR**（CoopDETR：一种基于对象查询的统一3D目标检测协同感知框架）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.19313)] [~~代码~~]\n- **Co-MTP**（Co-MTP：面向自动驾驶的多时间尺度融合协同轨迹预测框架）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.16589)] [[代码](https:\u002F\u002Fgithub.com\u002Fxiaomiaozhang\u002FCo-MTP)]\n- **Direct-CP**（Direct-CP：通过主动注意力实现联网自动驾驶车辆的定向协同感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.08840)] [~~代码~~]\n- **V2X-DG**（V2X-DG：面向车联网协同感知的领域泛化）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.15435)] [~~代码~~]\n\n### IROS 2025\n\n- **CooPre**（CooPre：面向V2X协同感知的协同预训练）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.11241)] [[代码](https:\u002F\u002Fgithub.com\u002Fucla-mobility\u002FCooPre)]\n- **CoPAD**（CoPAD：V2X场景下基于锚点导向解码器的多源轨迹融合与协同轨迹预测）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.15984)] [~~代码~~]\n- **CRUISE**（CRUISE：利用高斯泼溅技术在V2X场景中进行协同重建与编辑）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.18473)] [[代码](https:\u002F\u002Fgithub.com\u002FSainingZhang\u002FCRUISE)]\n\n### CVPR 2024\n\n- **CoHFF**（联网自动驾驶车辆中基于混合特征融合的协同语义占用预测）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.07635)] [~~代码~~]\n- **CoopDet3D**（TUMTraf V2X协同感知数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.01316)] [[代码](https:\u002F\u002Fgithub.com\u002Ftum-traffic-dataset\u002Fcoopdet3d)]\n- **CodeFilling**（通过码本信息填充实现通信高效的协同感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.04966)] [[代码](https:\u002F\u002Fgithub.com\u002FPhyllisH\u002FCodeFilling)]\n- **ERMVP**（ERMVP：在复杂环境中实现通信高效且协作鲁棒的多车感知）[[论文](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2024\u002Fhtml\u002FZhang_ERMVP_Communication-Efficient_and_Collaboration-Robust_Multi-Vehicle_Perception_in_Challenging_Environments_CVPR_2024_paper.html)] [[代码](https:\u002F\u002Fgithub.com\u002FTerry9a\u002FERMVP)]\n- **MRCNet**（基于运动感知鲁棒通信网络的多智能体协同感知）[[论文](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2024\u002Fhtml\u002FHong_Multi-agent_Collaborative_Perception_via_Motion-aware_Robust_Communication_Network_CVPR_2024_paper.html)] [[代码](https:\u002F\u002Fgithub.com\u002FIndigoChildren\u002Fcollaborative-perception-MRCNet)]\n\n### NeurIPS 2024\n\n- **V2X-Graph**（用于运动预测的协同轨迹表示学习）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.00371)] [[代码](https:\u002F\u002Fgithub.com\u002FAIR-THU\u002FV2X-Graph)]\n\n### ECCV 2024\n\n- **Hetecooper**（Hetecooper：面向异构协同感知的特征协作图）[[论文](https:\u002F\u002Feccv.ecva.net\u002Fvirtual\u002F2024\u002Fposter\u002F2467)] [~~代码~~]\n- **Infra-Centric CP**（重新思考基础设施在协同感知中的作用）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2410.11259)] [~~代码~~]\n\n### ICLR 2024\n\n- **HEAL**（一个可扩展的开放异构协同感知框架）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=KkrDUGIASk)] [[代码](https:\u002F\u002Fgithub.com\u002Fyifanlu0227\u002FHEAL)]\n\n### AAAI 2024\n\n- **CMiMC**（什么构成了良好的协同视图？基于对比互信息最大化的多智能体感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.10068)] [[代码](https:\u002F\u002Fgithub.com\u002F77SWF\u002FCMiMC)]\n- **DI-V2X**（DI-V2X：学习车路协同3D目标检测的领域不变表示）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.15742)] [[代码](https:\u002F\u002Fgithub.com\u002FSerenos\u002FDI-V2X)]\n- **V2XFormer**（DeepAccident：面向V2X自动驾驶的运动与事故预测基准）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.01168)] [[代码](https:\u002F\u002Fgithub.com\u002Ftianqi-wang1996\u002FDeepAccident)]\n\n### WACV 2024\n\n- **MACP**（MACP：面向协同感知的高效模型适应）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.16870)] [[代码](https:\u002F\u002Fgithub.com\u002FPurdueDigitalTwin\u002FMACP)]\n\n### ICRA 2024\n\n- **DMSTrack**（基于可微分多传感器卡尔曼滤波器的自动驾驶用概率3D多目标协同跟踪）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.14655)] [[代码](https:\u002F\u002Fgithub.com\u002Feddyhkchiu\u002FDMSTrack)]\n- **FreeAlign**（无需外部定位和时钟设备的鲁棒协同感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.02965)] [[代码](https:\u002F\u002Fgithub.com\u002FMediaBrain-SJTU\u002FFreeAlign)]\n\n### CVPR 2023\n\n- {相关} **BEVHeight**（BEVHeight：基于视觉的路边3D目标检测鲁棒框架）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.08498)] [[代码](https:\u002F\u002Fgithub.com\u002FADLab-AutoDrive\u002FBEVHeight)]\n- **CoCa3D**（协作助力摄像头在3D目标检测中超越LiDAR）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.13560)] [[代码](https:\u002F\u002Fgithub.com\u002FMediaBrain-SJTU\u002FCoCa3D)]\n- **FF-Tracking**（V2X-Seq：面向车路协同感知与预测的大规模序列数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.05938)] [[代码](https:\u002F\u002Fgithub.com\u002FAIR-THU\u002FDAIR-V2X-Seq)]\n\n### NeurIPS 2023\n\n- **CoBEVFlow**（基于鸟瞰图流的鲁棒异步协同3D检测）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=UHIDdtxmVS)] [[代码](https:\u002F\u002Fgithub.com\u002FMediaBrain-SJTU\u002FCoBEVFlow)]\n- **FFNet**（基于流的特征融合用于车路协同3D目标检测）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=gsglrhvQxX)] [[代码](https:\u002F\u002Fgithub.com\u002Fhaibao-yu\u002FFFNet-VIC3D)]\n- **How2comm**（How2comm：通信高效且协作实用的多智能体感知）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=Dbaxm9ujq6)] [[代码](https:\u002F\u002Fgithub.com\u002Fydk122024\u002FHow2comm)]\n\n### ICCV 2023\n\n- **CORE**（CORE：面向多智能体感知的协同重建）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.11514)] [[代码](https:\u002F\u002Fgithub.com\u002Fzllxot\u002FCORE)]\n- **HM-ViT**（HM-ViT：基于视觉Transformer的异构模态车车协同感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.10628)] [[代码](https:\u002F\u002Fgithub.com\u002FXHwind\u002FHM-ViT)]\n- **ROBOSAC**（Among Us：通过共识实现对抗性鲁棒的协同感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.09495)] [[代码](https:\u002F\u002Fgithub.com\u002Fcoperception\u002FROBOSAC)]\n- **SCOPE**（面向多智能体协同感知的时空域感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.13929)] [[代码](https:\u002F\u002Fgithub.com\u002Fstarfdu1418\u002FSCOPE)]\n- **TransIFF**（TransIFF：基于Transformer的车路协同3D检测实例级特征融合框架）[[论文](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2023\u002Fhtml\u002FChen_TransIFF_An_Instance-Level_Feature_Fusion_Framework_for_Vehicle-Infrastructure_Cooperative_3D_ICCV_2023_paper.html)] [~~代码~~]\n- **UMC**（UMC：统一的带宽高效多分辨率协同感知框架）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.12400)] [[代码](https:\u002F\u002Fgithub.com\u002Fispc-lab\u002FUMC)]\n\n### ICLR 2023\n\n- {相关} **CO3**（CO3：面向自动驾驶的协同无监督3D表征学习）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=QUaDoIdgo0)] [[代码](https:\u002F\u002Fgithub.com\u002FRunjian-Chen\u002FCO3)]\n\n### CoRL 2023\n\n- **BM2CP** {BM2CP：基于激光雷达与摄像头模态的高效协同感知} [[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=uJqxFjF1xWp)] [[代码](https:\u002F\u002Fgithub.com\u002FbyzhaoAI\u002FBM2CP)]\n\n### MM 2023\n\n- **DUSA**（DUSA：解耦的无监督Sim2Real适应用于车联网协同感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.08117)] [[代码](https:\u002F\u002Fgithub.com\u002Frefkxh\u002FDUSA)]\n- **FeaCo**（FeaCo：在噪声姿态条件下实现鲁棒的特征级一致性）[[论文](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002Fabs\u002F10.1145\u002F3581783.3611880)] [[代码](https:\u002F\u002Fgithub.com\u002Fjmgu0212\u002FFeaCo)]\n- **What2comm**（What2comm：通过特征解耦实现通信高效的协同感知）[[论文](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002Fabs\u002F10.1145\u002F3581783.3611699)] [~~代码~~]\n\n### WACV 2023\n\n- **AdaFusion**（用于协同感知的自适应特征融合，利用激光雷达点云）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.00116)] [[代码](https:\u002F\u002Fgithub.com\u002FDonghaoQiao\u002FAdaptive-Feature-Fusion-for-Cooperative-Perception)]\n\n### ICRA 2023\n\n- **CoAlign**（在姿态误差存在时的鲁棒协同3D目标检测）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.07214)] [[代码](https:\u002F\u002Fgithub.com\u002Fyifanlu0227\u002FCoAlign)]\n- {相关} **DMGM**（用于协同感知中对应关系识别的深度掩码图匹配）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.07555)] [[代码](https:\u002F\u002Fgithub.com\u002Fgaopeng5\u002FDMGM)]\n- **Double-M Quantification**（面向自动驾驶的协同检测不确定性量化）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.08162)] [[代码](https:\u002F\u002Fgithub.com\u002Fcoperception\u002Fdouble-m-quantification)]\n- **MAMP**（模型无关的多智能体感知框架）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.13168)] [[代码](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002Fmodel_anostic)]\n- **MATE**（通过多智能体轨迹交换进行通信关键型规划）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.06080)] [~~代码~~]\n- **MPDA**（弥合多智能体感知的领域差距）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2210.08451)] [[代码](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002FMPDA)]\n- **WNT**（我们需要谈谈：识别并克服自动驾驶中的通信关键场景）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.04352)] [~~代码~~]\n\n### CVPR 2022\n\n- **Coopernaut**（COOPERNAUT：面向联网车辆的协同感知端到端驾驶）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.02222)] [[代码](https:\u002F\u002Fgithub.com\u002FUT-Austin-RPL\u002FCoopernaut)]\n- {相关} **LAV**（向所有车辆学习）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.11934)] [[代码](https:\u002F\u002Fgithub.com\u002Fdotchen\u002FLAV)]\n- **TCLF**（DAIR-V2X：用于车路协同3D目标检测的大规模数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.05575)] [[代码](https:\u002F\u002Fgithub.com\u002FAIR-THU\u002FDAIR-V2X)]\n\n### NeurIPS 2022\n\n- **Where2comm**（Where2comm：通过空间置信度图实现高效协同感知）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=dLL4KXzKUpS)] [[代码](https:\u002F\u002Fgithub.com\u002FMediaBrain-SJTU\u002Fwhere2comm)]\n\n### ECCV 2022\n\n- **SyncNet**（延迟感知的协同感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2207.08560)] [[代码](https:\u002F\u002Fgithub.com\u002FMediaBrain-SJTU\u002FSyncNet)]\n- **V2X-ViT**（V2X-ViT：基于视觉Transformer的车车协同感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.10638)] [[代码](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002Fv2x-vit)]\n\n### CoRL 2022\n\n- **CoBEVT**（CoBEVT：基于稀疏Transformer的协同鸟瞰图语义分割）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=PAFEQQtDf8s)] [[代码](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002FCoBEVT)]\n- **STAR**（多机器人场景补全：迈向任务无关的协同感知）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=hW0tcXOJas2)] [[代码](https:\u002F\u002Fgithub.com\u002Fcoperception\u002Fstar)]\n\n### IJCAI 2022\n\n- **IA-RCP**（抗通信中断的鲁棒协同感知）[[论文](https:\u002F\u002Flearn-to-race.org\u002Fworkshop-ai4ad-ijcai2022\u002Fpapers.html)] [~~代码~~]\n\n### MM 2022\n\n- **CRCNet**（面向多智能体感知的互补增强与冗余最小化协作网络）[[论文](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002Fabs\u002F10.1145\u002F3503161.3548197)] [~~代码~~]\n\n### ICRA 2022\n\n- **AttFuse**（OPV2V：一个用于车车通信感知的开放基准数据集及融合管道）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.07644)] [[代码](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002FOpenCOOD)]\n- **MP-Pose**（基于图神经网络的多机器人协同感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.01760)] [~~代码~~]\n\n### NeurIPS 2021\n\n- **DiscoNet**（用于多智能体感知的蒸馏协作图学习）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=ZRcjSOmYraB)] [[代码](https:\u002F\u002Fgithub.com\u002Fai4ce\u002FDiscoNet)]\n\n### ICCV 2021\n\n- **Adversarial V2V**（多智能体通信中的对抗攻击）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2101.06560)] [~~代码~~]\n\n### IROS 2021\n\n- **MASH**（通过带宽受限的多智能体空间握手克服障碍）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2107.00771)] [[代码](https:\u002F\u002Fgithub.com\u002Fyifanlu0227\u002FCoAlign)]\n\n### CVPR 2020\n\n- **When2com**（When2com：基于通信图分组的多智能体感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2006.00176)] [[代码](https:\u002F\u002Fgithub.com\u002FGT-RIPL\u002FMultiAgentPerception)]\n\n### ECCV 2020\n\n- **DSDNet**（DSDNet：深度结构化自动驾驶网络）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2008.06041)] [~~代码~~]\n- **V2VNet**（V2VNet：用于联合感知与预测的车车通信网络）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2008.07519)] [[代码](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002FOpenCOOD)]\n\n### CoRL 2020\n\n- **Robust V2V**（学习通信并纠正姿态误差）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.05289)] [[代码](https:\u002F\u002Fgithub.com\u002Fyifanlu0227\u002FCoAlign)]\n\n### ICRA 2020\n\n- **Who2com**（Who2com：基于可学习握手通信的协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.09575)] [[代码](https:\u002F\u002Fgithub.com\u002FGT-RIPL\u002FMultiAgentPerception)]\n- **MAIN**（通过学习的数据关联增强多机器人感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2107.00769)] [~~代码~~]\n\n\n\n## :bookmark:数据集与仿真平台\n\n注：{Real}表示传感器数据来源于真实世界采集，而非仿真。\n\n### 精选预印本\n\n- **Adver-City**（Adver-City：面向恶劣天气下协作感知的开源多模态数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2410.06380)] [[代码](https:\u002F\u002Fgithub.com\u002FQUARRG\u002FAdver-City)] [[项目](https:\u002F\u002Flabs.cs.queensu.ca\u002Fquarrg\u002Fdatasets\u002Fadver-city)]\n- **AirV2X**（AirV2X：空地车辆到万物协同统一）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.19283)] [[代码](https:\u002F\u002Fgithub.com\u002Ftaco-group\u002FAirV2X-Perception)] [[项目](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fxiangbog\u002FAirV2X-Perception)]\n- **CATS-V2V**（CATS-V2V：包含复杂恶劣交通场景的真实车车协同感知数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.11168)] [~~代码~~] [[项目](https:\u002F\u002Fcats-v2v-dataset.github.io)]\n- {Real} **CoInfra**（CoInfra：恶劣天气下的大规模协同基础设施感知系统及数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.02245)] [[代码](https:\u002F\u002Fgithub.com\u002FNingMingHao\u002FCoInfra)] [~~项目~~]\n- **CP-GuardBench**（CP-Guard+：协作感知中恶意代理检测与防御的新范式）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=9MNzHTSDgh)] [~~代码~~] [~~项目~~]\n- **Griffin**（Griffin：空地协同探测与跟踪数据集及基准测试）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.06983)] [[代码](https:\u002F\u002Fgithub.com\u002Fwang-jh18-SVM\u002FGriffin)] [[项目](https:\u002F\u002Fpan.baidu.com\u002Fs\u002F1NDgsuHB-QPRiROV73NRU5g)]\n- {Real} **InScope**（InScope：面向开放交通场景的新型真实3D基础设施侧协作感知数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.21581)] [[代码](https:\u002F\u002Fgithub.com\u002Fxf-zh\u002FInScope)] [~~项目~~]\n- **MobileVerse**（MobiVerse：利用混合轻量级领域专用生成器和大语言模型扩展城市出行仿真）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.21784)] [[代码](https:\u002F\u002Fgithub.com\u002Fucla-mobility\u002FMobiVerse)] [~~项目~~]\n- **Multi-V2X**（Multi-V2X：面向协作感知的大规模多模态、多渗透率数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.04980)] [[代码](https:\u002F\u002Fgithub.com\u002FRadetzkyLi\u002FMulti-V2X)] [~~项目~~]\n- **M3CAD**（M3CAD：迈向通用协作自动驾驶基准测试）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.06746)] [[代码](https:\u002F\u002Fgithub.com\u002Fzhumorui\u002FM3CAD)] [[项目](https:\u002F\u002Fzhumorui.github.io\u002Fm3cad)]\n- **OPV2V-N**（RCDN：基于动态特征的3D神经建模，实现鲁棒的相机不敏感协作感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.16868)] [~~代码~~] [~~项目~~]\n- **TalkingVehiclesGym**（通过自我博弈实现协作自动驾驶的自然语言通信）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.18334)] [[代码](https:\u002F\u002Fgithub.com\u002Fcuijiaxun\u002Ftalking-vehicles)] [[项目](https:\u002F\u002Ftalking-vehicles.github.io)]\n- **TruckV2X**（TruckV2X：以卡车为中心的感知数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.09505)] [~~代码~~] [[项目](https:\u002F\u002Fxietenghu1.github.io\u002FTruckV2X)]\n- {Real} **UrbanV2X**（UrbanV2X：面向城市区域协同导航的多传感器车路协同数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2512.20224)] [[代码](https:\u002F\u002Fgithub.com\u002Farclab-hku\u002FEvent_based_VO-VIO-SLAM)] [[项目](https:\u002F\u002Fpolyu-taslab.github.io\u002FUrbanV2X)]\n- **V2V-QA**（V2V-LLM：基于多模态大语言模型的车车协同自动驾驶）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.09980)] [[代码](https:\u002F\u002Fgithub.com\u002Feddyhkchiu\u002FV2VLLM)] [[项目](https:\u002F\u002Feddyhkchiu.github.io\u002Fv2vllm.github.io)]\n- {Real} **V2XPnP-Seq**（V2XPnP：面向多智能体感知与预测的车到万物时空融合）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.01812)] [[代码](https:\u002F\u002Fgithub.com\u002FZewei-Zhou\u002FV2XPnP)] [[项目](https:\u002F\u002Fmobility-lab.seas.ucla.edu\u002Fv2xpnp)]\n- {Real} **V2X-Radar**（V2X-Radar：包含4D雷达的多模态协同感知数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.10962)] [[代码](https:\u002F\u002Fgithub.com\u002Fyanglei18\u002FV2X-Radar)] [[项目](http:\u002F\u002Fopenmpd.com\u002Fcolumn\u002FV2X-Radar)]\n- {Real} **V2X-Real**（V2X-Real：面向车到万物协同感知的大规模数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.16034)] [~~代码~~] [[项目](https:\u002F\u002Fmobility-lab.seas.ucla.edu\u002Fv2x-real)]\n- {Real} **V2X-ReaLO**（V2X-ReaLO：现实环境中协作感知的开放式在线框架与数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.10034)] [~~代码~~] [~~项目~~]\n- **WHALES**（WHALES：用于提升自动驾驶中协作效率的多智能体调度数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.13340)] [[代码](https:\u002F\u002Fgithub.com\u002FchensiweiTHU\u002FWHALES)] [[项目](https:\u002F\u002Fpan.baidu.com\u002Fs\u002F1dintX-d1T-m2uACqDlAM9A)]\n\n### CVPR 2025\n\n- **Mono3DVLT-V2X**（Mono3DVLT：基于单目视频的三维视觉语言跟踪）[[论文](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2025\u002Fhtml\u002FWei_Mono3DVLT_Monocular-Video-Based_3D_Visual_Language_Tracking_CVPR_2025_paper.html)] [~~代码~~] [~~项目~~]\n- **RCP-Bench**（RCP-Bench：针对多种干扰下的协同感知鲁棒性基准测试）[[论文](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2025\u002Fhtml\u002FDu_RCP-Bench_Benchmarking_Robustness_for_Collaborative_Perception_Under_Diverse_Corruptions_CVPR_2025_paper.html)] [[代码](https:\u002F\u002Fgithub.com\u002FLuckyDush\u002FRCP-Bench)] [~~项目~~]\n- **V2X-R**（V2X-R：基于去噪扩散的协作式激光雷达-4D雷达融合3D目标检测）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.08402)] [[代码](https:\u002F\u002Fgithub.com\u002Fylwhxht\u002FV2X-R)] [~~项目~~]\n\n### NeurIPS 2025\n\n- {真实} **AGC-Drive**（AGC-Drive：面向驾驶场景中空地协同的真实世界大规模数据集）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=N07WGSPh9l)] [[代码](https:\u002F\u002Fgithub.com\u002FPercepX\u002FAGC-Drive)] [[项目](https:\u002F\u002Fagc-drive.github.io)]\n- **UrbanIng-V2X**（UrbanIng-V2X：跨多个交叉口的大规模多车辆、多基础设施协同感知数据集）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=iSwIkUqyqf)] [[代码](https:\u002F\u002Fgithub.com\u002Fthi-ad\u002FUrbanIng-V2X)] [[项目](https:\u002F\u002Fpypi.org\u002Fproject\u002Furbaning)]\n\n### ICCV 2025\n\n- **CoPe-R**（SlimComm：基于多普勒引导的稀疏查询，用于带宽高效的三维协同感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.13007)] [[代码](https:\u002F\u002Fgithub.com\u002Ffzi-forschungszentrum-informatik\u002FSlimComm)] [~~项目~~]\n- {真实} **Mixed Signals**（Mixed Signals：面向异构激光雷达V2X协同的多样化点云数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.14156)] [[代码](https:\u002F\u002Fgithub.com\u002Fchinitaberrio\u002FMixed-Signals)] [[项目](https:\u002F\u002Fmixedsignalsdataset.cs.cornell.edu)]\n\n### CVPR 2024\n\n- {真实} **HoloVIC**（HoloVIC：多传感器全息交叉口与车路协同的大规模数据集及基准测试）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.02640)] [~~代码~~] [[项目](https:\u002F\u002Fholovic.net)]\n- {真实} **Open Mars Dataset**（多智能体、多路径、多模态自动驾驶：Open MARS数据集）[[代码](https:\u002F\u002Fgithub.com\u002Fai4ce\u002FMARS)] [[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.09383)] [[项目](https:\u002F\u002Fai4ce.github.io\u002FMARS)]\n- {真实} **RCooper**（RCooper：面向路边协同感知的真实世界大规模数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.10145)] [[代码](https:\u002F\u002Fgithub.com\u002FAIR-THU\u002FDAIR-RCooper)] [[项目](https:\u002F\u002Fwww.t3caic.com\u002Fqingzhen)]\n- {真实} **TUMTraf-V2X**（TUMTraf V2X协同感知数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.01316)] [[代码](https:\u002F\u002Fgithub.com\u002Ftum-traffic-dataset\u002Ftum-traffic-dataset-dev-kit)] [[项目](https:\u002F\u002Ftum-traffic-dataset.github.io\u002Ftumtraf-v2x)]\n\n### NeurIPS 2024\n\n- {真实} **DAIR-V2X-Traj**（学习用于运动预测的协同轨迹表示）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.00371)] [[代码](https:\u002F\u002Fgithub.com\u002FAIR-THU\u002FV2X-Graph)] [[项目](https:\u002F\u002Fthudair.baai.ac.cn\u002Findex)]\n\n### ECCV 2024\n\n- {真实} **H-V2X**（H-V2X：面向BEV感知的大规模高速公路数据集）[[论文](https:\u002F\u002Feccv2024.ecva.net\u002Fvirtual\u002F2024\u002Fposter\u002F126)] [~~代码~~] [~~项目~~]\n\n### ICLR 2024\n\n- **OPV2V-H**（开放异构协同感知的可扩展框架）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=KkrDUGIASk)] [[代码](https:\u002F\u002Fgithub.com\u002Fyifanlu0227\u002FHEAL)] [[项目](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fyifanlu\u002FOPV2V-H)]\n\n### AAAI 2024\n\n- **DeepAccident**（DeepAccident：面向V2X自动驾驶的运动与事故预测基准测试）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.01168)] [[代码](https:\u002F\u002Fgithub.com\u002Ftianqi-wang1996\u002FDeepAccident)] [[项目](https:\u002F\u002Fdeepaccident.github.io)]\n\n### CVPR 2023\n\n- **CoPerception-UAV+**（协作助力摄像头在3D检测中超越激光雷达）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.13560)] [[代码](https:\u002F\u002Fgithub.com\u002FMediaBrain-SJTU\u002FCoCa3D)] [[项目](https:\u002F\u002Fsiheng-chen.github.io\u002Fdataset\u002FCoPerception+)]\n- **OPV2V+**（协作助力摄像头在3D检测中超越激光雷达）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.13560)] [[代码](https:\u002F\u002Fgithub.com\u002FMediaBrain-SJTU\u002FCoCa3D)] [[项目](https:\u002F\u002Fsiheng-chen.github.io\u002Fdataset\u002FCoPerception+)]\n- {真实} **V2V4Real**（V2V4Real：面向车车协同感知的大规模真实世界数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.07601)] [[代码](https:\u002F\u002Fgithub.com\u002Fucla-mobility\u002FV2V4Real)] [[项目](https:\u002F\u002Fmobility-lab.seas.ucla.edu\u002Fv2v4real)]\n- {真实} **DAIR-V2X-Seq**（V2X-Seq：面向车路协同感知与预测的大规模序列数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.05938)] [[代码](https:\u002F\u002Fgithub.com\u002FAIR-THU\u002FDAIR-V2X-Seq)] [[项目](https:\u002F\u002Fthudair.baai.ac.cn\u002Findex)]\n\n### NeurIPS 2023\n\n- **IRV2V**（通过鸟瞰视图流实现鲁棒的异步协同3D检测）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=UHIDdtxmVS)] [~~代码~~] [~~项目~~]\n\n### ICCV 2023\n\n- **Roadside-Opt**（优化自动驾驶中路边激光雷达的部署位置）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.07247)] [~~代码~~] [~~项目~~]\n\n### ICRA 2023\n\n- {真实} **DAIR-V2X-C Complemented**（在姿态误差存在时实现鲁棒的协同3D物体检测）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.07214)] [[代码](https:\u002F\u002Fgithub.com\u002Fyifanlu0227\u002FCoAlign)] [[项目](https:\u002F\u002Fsiheng-chen.github.io\u002Fdataset\u002Fdair-v2x-c-complemented)]\n- **RLS**（利用真实的激光雷达仿真库分析基础设施激光雷达的部署）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.15975)] [[代码](https:\u002F\u002Fgithub.com\u002FPJLab-ADG\u002FLiDARSimLib-and-Placement-Evaluation)] [~~项目~~]\n- **V2XP-ASG**（V2XP-ASG：为车到万物感知生成对抗场景）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.13679)] [[代码](https:\u002F\u002Fgithub.com\u002FXHwind\u002FV2XP-ASG)] [~~项目~~]\n\n### CVPR 2022\n\n- **AutoCastSim**（COOPERNAUT：面向联网车辆的端到端协同感知自动驾驶）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.02222)] [[代码](https:\u002F\u002Fgithub.com\u002Fhangqiu\u002FAutoCastSim)] [[项目](https:\u002F\u002Futexas.app.box.com\u002Fv\u002Fcoopernaut-dataset)]\n- {真实} **DAIR-V2X**（DAIR-V2X：面向车路协同3D目标检测的大规模数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.05575)] [[代码](https:\u002F\u002Fgithub.com\u002FAIR-THU\u002FDAIR-V2X)] [[项目](https:\u002F\u002Fthudair.baai.ac.cn\u002Findex)]\n\n### NeurIPS 2022\n\n- **CoPerception-UAV**（Where2comm：通过空间置信度图实现高效协同感知）[[论文&评审](https:\u002F\u002Fopenreview.net\u002Fforum?id=dLL4KXzKUpS)] [[代码](https:\u002F\u002Fgithub.com\u002FMediaBrain-SJTU\u002Fwhere2comm)] [[项目](https:\u002F\u002Fsiheng-chen.github.io\u002Fdataset\u002Fcoperception-uav)]\n\n### ECCV 2022\n\n- **V2XSet**（V2X-ViT：基于视觉Transformer的车路协同感知）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.10638)] [[代码](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002Fv2x-vit)] [[项目](https:\u002F\u002Fdrive.google.com\u002Fdrive\u002Ffolders\u002F1r5sPiBEvo8Xby-nMaWUTnJIPK6WhY1B6)]\n\n### ICRA 2022\n\n- **OPV2V**（OPV2V：面向车车通信感知的开放基准数据集与融合流水线）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.07644)] [[代码](https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002FOpenCOOD)] [[项目](https:\u002F\u002Fmobility-lab.seas.ucla.edu\u002Fopv2v)]\n\n### ACCV 2022\n\n- **DOLPHINS**（DOLPHINS：用于实现和谐互联自动驾驶的协同感知数据集）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2207.07609)] [[代码](https:\u002F\u002Fgithub.com\u002Fexplosion5\u002FDolphins)] [[项目](https:\u002F\u002Fdolphins-dataset.net)]\n\n### ICCV 2021\n\n- **V2X-Sim**（V2X-Sim：面向自动驾驶的多智能体协同感知数据集与基准测试平台）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F2202.08449)] [[代码](https:\u002F\u002Fgithub.com\u002Fai4ce\u002FV2X-Sim)] [[项目](https:\u002F\u002Fai4ce.github.io\u002FV2X-Sim)]\n\n### CoRL 2017\n\n- **CARLA**（CARLA：开源城市驾驶模拟器）[[论文](https:\u002F\u002Farxiv.org\u002Fabs\u002F1711.03938)] [[代码](https:\u002F\u002Fgithub.com\u002Fcarla-simulator\u002Fcarla)] [[项目](https:\u002F\u002Fcarla.org)]","# Collaborative_Perception 快速上手指南\n\n**Collaborative_Perception** 并非一个单一的可执行软件包，而是一个专注于 **车路协同（V2X）**、**车车协同（V2V）** 及 **多智能体感知** 领域的论文汇总与资源索引库。它整理了最新的研究方法、数据集、仿真器及相关代码库。\n\n本指南将指导开发者如何利用该资源库快速定位所需工具，并基于推荐的框架（如 OpenCOOD 或 V2Xverse）搭建开发环境。\n\n## 1. 环境准备\n\n由于该仓库主要作为资源索引，实际开发需依赖其推荐的具体框架（如 **OpenCOOD**, **HEAL**, **V2Xverse** 等）。以下以社区最广泛使用的 **OpenCOOD** 框架为例说明通用环境要求。\n\n### 系统要求\n- **操作系统**: Ubuntu 18.04 \u002F 20.04 \u002F 22.04 (推荐)\n- **GPU**: NVIDIA GPU (显存建议 ≥ 8GB)，支持 CUDA 11.x 或更高版本\n- **Python**: 3.8 - 3.10\n\n### 前置依赖\n在克隆具体框架代码前，请确保系统已安装以下基础工具：\n```bash\nsudo apt-get update\nsudo apt-get install -y git vim wget curl build-essential cmake\n```\n\n> **注意**：国内开发者建议在后续步骤中配置清华源或阿里源以加速依赖下载。\n\n## 2. 安装步骤\n\n由于 `Collaborative_Perception` 本身是论文列表，请选择列表中 **Method and Framework** 部分的一个具体项目进行安装。此处以 **OpenCOOD** (Open Cooperative Detection Framework) 为例。\n\n### 步骤 1: 克隆项目代码\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002FDerrickXuNu\u002FOpenCOOD.git\ncd OpenCOOD\n```\n\n### 步骤 2: 创建虚拟环境\n推荐使用 `conda` 管理环境：\n```bash\nconda create -n opencood python=3.8\nconda activate opencood\n```\n\n### 步骤 3: 安装 PyTorch 及依赖\n**国内加速方案**：使用清华大学镜像源安装 PyTorch（根据你的 CUDA 版本调整，以下为 CUDA 11.8 示例）：\n```bash\npip install torch torchvision torchaudio --index-url https:\u002F\u002Fdownload.pytorch.org\u002Fwhl\u002Fcu118\n```\n\n安装项目其余依赖：\n```bash\n# 配置 pip 国内镜像（可选，加速下载）\npip config set global.index-url https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple\n\n# 安装 requirements\npip install -r requirements.txt\n```\n\n### 步骤 4: 编译扩展模块\n部分协同感知算法需要编译 CUDA 扩展：\n```bash\npython setup.py develop\n```\n\n## 3. 基本使用\n\n以下展示如何运行一个基础的协同感知训练示例（以 OpenCOOD 中的 V2V 场景为例）。\n\n### 数据准备\n你需要先下载协同感知数据集（如 OPV2V, DAIR-V2X 等）。请在 `data_root` 目录下按照项目文档结构放置数据。\n*注：具体数据集下载链接可在 `Collaborative_Perception` 仓库的 **Dataset and Simulator** 章节查找。*\n\n### 运行训练\n使用配置文件启动训练。以下命令演示使用单卡训练一个基础的协同感知模型：\n\n```bash\ncd opencood\u002Ftools\npython train.py --model f_cooper --save_dir .\u002Foutput --config_file ..\u002Fopencood\u002Fopencood\u002Fdata_utils\u002Fyaml\u002Fopv2v_f_cooper.yaml\n```\n\n### 运行推理\u002F可视化\n训练完成后，可使用以下命令进行可视化测试：\n```bash\npython vis.py --model f_cooper --save_dir .\u002Foutput --config_file ..\u002Fopencood\u002Fopencood\u002Fdata_utils\u002Fyaml\u002Fopv2v_f_cooper.yaml --load_pth .\u002Foutput\u002Flatest.pth\n```\n\n### 探索更多模型\n回到 **Collaborative_Perception** 主页面，查看 **Selected Preprint** 或 **Method and Framework** 列表，点击对应论文的 `[code]` 链接，即可获取如 **CoBEVFusion**, **Where2comm**, **V2Xverse** 等其他先进算法的实现代码，并参照其各自的 README 进行复用。","某自动驾驶车队在复杂城市路口进行 L4 级路测时，面临多车协同感知的算法选型与集成难题。\n\n### 没有 Collaborative_Perception 时\n- 研发团队需在海量学术论文中盲目摸索，难以快速定位适用于 V2V（车对车）或 V2I（车对路）场景的最新协同感知框架。\n- 由于缺乏统一的数据集和模拟器索引，复现基准模型耗时耗力，且常因训练设置不透明导致结果无法公平对比。\n- 面对通信带宽受限或时序异步等实际工程痛点，找不到经过验证的解决方案（如 Where2comm 或 CoBEVFlow），只能重复造轮子。\n- 团队成员对协同感知的理论边界认知模糊，缺乏系统的综述资源和学习路径，导致技术路线决策缓慢。\n\n### 使用 Collaborative_Perception 后\n- 开发者可直接查阅按字母排序的论文库，迅速锁定针对特定场景（如遮挡消除）的最优方法论与代码实现。\n- 利用整理好的数据集与模拟器资源清单，大幅缩短环境搭建周期，并参考归档版本中的历史数据避免复现陷阱。\n- 通过推荐的视频教程和专题报告，快速掌握解决通信中断、带宽优化及异构协作等关键问题的前沿技术。\n- 依托丰富的综述文献与学习资源，团队能清晰构建从理论到落地的知识体系，高效制定技术演进路线。\n\nCollaborative_Perception 通过一站式聚合前沿成果与实战资源，将自动驾驶协同感知技术的研发探索周期从数月缩短至数周。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FLittle-Podi_Collaborative_Perception_9ce85255.png","Little-Podi","Shenyuan Gao","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002FLittle-Podi_e114da8a.jpg","Ph.D. student at HKUST since 22fall.","NVIDIA Research · HKUST","Hong Kong SAR","sygao@connect.ust.hk","ShenyuanGao","https:\u002F\u002Fbit.ly\u002Fsygao_scholar","https:\u002F\u002Fgithub.com\u002FLittle-Podi",null,592,64,"2026-04-10T07:34:03",5,"","未说明",{"notes":90,"python":88,"dependencies":91},"该仓库是一个论文摘要列表（Paper Digest），汇集了协同感知领域的研究论文、数据集、模拟器和相关代码库链接，本身不是一个可直接运行的单一软件工具。因此，README 中未包含具体的操作系统、GPU、内存、Python 版本或依赖库的安装需求。用户若需运行其中列出的具体算法（如 OpenCOOD, V2Xverse 等），需前往对应的子项目仓库查阅其独立的环境配置说明。",[],[15,13],[94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109],"autonomous-driving","collaborative-perception","cooperative-perception","v2v","multi-agent-perception","v2i","v2x","awesome-list","paper-list","multi-agent-system","cvpr","eccv","iccv","neurips","corl","iclr","2026-03-27T02:49:30.150509","2026-04-13T16:15:25.902109",[113,118,123,128,133,138,143],{"id":114,"question_zh":115,"answer_zh":116,"source_url":117},31739,"为什么在复现 When2Com 和 Who2Com 时，两者的指标非常接近，甚至有时 When2Com 不如 Who2Com？","这可能是因为实验设置（如 MRMPS）中存在噪声，导致 When2Com 学到了“捷径”，即仅依靠自车特征就能收敛。而 Who2Com 通过显式的拼接提供了这种捷径，从而表现出合作效果。建议尝试在 When2Com 中添加自车特征拼接来观察变化。此外，这两种方法在握手阶段的消息生成方式较为粗暴，可以尝试更先进的通信方法。根本原因可能在于这些方法的设计倾向于与高相似度的智能体通信，但在协作系统中，智能体通常更需要来自他人的互补信息而非相似信息。","https:\u002F\u002Fgithub.com\u002FLittle-Podi\u002FCollaborative_Perception\u002Fissues\u002F2",{"id":119,"question_zh":120,"answer_zh":121,"source_url":122},31740,"Where2Comm 为什么不直接传输位姿和边界框等纯文本信息以降低通信成本，而是要传输特征图？","直接传输位姿和边界框属于“后期协作”（late collaboration）模式，虽然能显著降低通信成本，但并非性能与带宽权衡的最优解。例如，如果所有车辆对某个物体的观测都不完整，单独都无法检测到该物体，那么共享的边界框中就不会包含它，导致系统无法检测该物体。而共享信息量更大但通信开销较高的特征图，可以将部分观测聚集起来，恢复出完整的物体检测。当前研究的目标正是推动这种性能与带宽的平衡。","https:\u002F\u002Fgithub.com\u002FLittle-Podi\u002FCollaborative_Perception\u002Fissues\u002F3",{"id":124,"question_zh":125,"answer_zh":126,"source_url":127},31741,"Where2Comm 已经通过选择空间稀疏区域降低了通信成本，是否还需要使用压缩比？压缩会导致通道信息丢失吗？","即使 Where2Comm 通过阈值（如 0.01）过滤了大部分特征图实现了空间稀疏，压缩仍然是可选的优化手段，但是否必须取决于具体的带宽限制需求。关于通道丢失问题，压缩确实可能带来信息损失，需要在通信效率和感知性能之间进行权衡。对于 V2X-VIT 中显示的 0x 压缩率，通常意味着默认未启用压缩或使用了无损\u002F特定编码。如果想进一步降低通信成本，可以尝试在训练中直接将压缩比调整为 32，但需重新评估模型性能下降的程度。","https:\u002F\u002Fgithub.com\u002FLittle-Podi\u002FCollaborative_Perception\u002Fissues\u002F4",{"id":129,"question_zh":130,"answer_zh":131,"source_url":132},31742,"如何将 CoPerception-UAV 数据集转换为 OpenCDA 支持的格式？","维护者表示对该非广泛使用的数据集没有具体的转换方案。建议直接联系该数据集的作者，或在他们的代码仓库中寻找详细的转换脚本和说明文档。","https:\u002F\u002Fgithub.com\u002FLittle-Podi\u002FCollaborative_Perception\u002Fissues\u002F5",{"id":134,"question_zh":135,"answer_zh":136,"source_url":137},31743,"如何向该仓库推荐或添加新的协同感知论文和代码资源？","用户可以通过提交 Issue 的方式推荐新的工作。在 Issue 中提供论文的标题、发表会议\u002F期刊、论文链接（如 arXiv）以及代码仓库链接。维护者在确认后会将其添加到仓库的列表中以造福社区。例如，已有用户成功添加了 CoPEFT (AAAI 2025), INSTINCT (ICCV 2025), CooPre (IROS 2025) 等工作。","https:\u002F\u002Fgithub.com\u002FLittle-Podi\u002FCollaborative_Perception\u002Fissues\u002F8",{"id":139,"question_zh":140,"answer_zh":141,"source_url":142},31744,"有哪些推荐的协同感知仿真框架、课程或综述材料可以辅助学习？","推荐以下资源：1. 课程讲座：'Cooperative Driving Automation -- Simulation and Perception'；2. 仿真框架：OpenCDA (支持全栈协同驾驶自动化，包括感知、定位、规划和控制)；3. 综述论文：关于不同协作模式的详细讨论可参考 arXiv:2301.06262。这些材料已被收录在该仓库中供社区参考。","https:\u002F\u002Fgithub.com\u002FLittle-Podi\u002FCollaborative_Perception\u002Fissues\u002F1",{"id":144,"question_zh":145,"answer_zh":146,"source_url":147},31745,"SlimmCom 项目是否有基于雷达的数据集可用？","是的，SlimmCom 项目已更新并发布了基于雷达的数据集，名为 CoPe-R。数据集生成代码及相应的 README 文档已在该项目的代码仓库中提供，相关代码也将陆续上传。","https:\u002F\u002Fgithub.com\u002FLittle-Podi\u002FCollaborative_Perception\u002Fissues\u002F11",[]]