[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-opendr-eu--opendr":3,"tool-opendr-eu--opendr":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",150037,2,"2026-04-10T23:33:47",[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":32,"last_commit_at":59,"category_tags":60,"status":17},4721,"markitdown","microsoft\u002Fmarkitdown","MarkItDown 是一款由微软 AutoGen 团队打造的轻量级 Python 工具，专为将各类文件高效转换为 Markdown 格式而设计。它支持 PDF、Word、Excel、PPT、图片（含 OCR）、音频（含语音转录）、HTML 乃至 YouTube 链接等多种格式的解析，能够精准提取文档中的标题、列表、表格和链接等关键结构信息。\n\n在人工智能应用日益普及的今天，大语言模型（LLM）虽擅长处理文本，却难以直接读取复杂的二进制办公文档。MarkItDown 恰好解决了这一痛点，它将非结构化或半结构化的文件转化为模型“原生理解”且 Token 效率极高的 Markdown 格式，成为连接本地文件与 AI 分析 pipeline 的理想桥梁。此外，它还提供了 MCP（模型上下文协议）服务器，可无缝集成到 Claude Desktop 等 LLM 应用中。\n\n这款工具特别适合开发者、数据科学家及 AI 研究人员使用，尤其是那些需要构建文档检索增强生成（RAG）系统、进行批量文本分析或希望让 AI 助手直接“阅读”本地文件的用户。虽然生成的内容也具备一定可读性，但其核心优势在于为机器",93400,"2026-04-06T19:52:38",[52,14],{"id":62,"github_repo":63,"name":64,"description_en":65,"description_zh":66,"ai_summary_zh":67,"readme_en":68,"readme_zh":69,"quickstart_zh":70,"use_case_zh":71,"hero_image_url":72,"owner_login":73,"owner_name":74,"owner_avatar_url":75,"owner_bio":76,"owner_company":77,"owner_location":77,"owner_email":77,"owner_twitter":77,"owner_website":78,"owner_url":79,"languages":80,"stars":119,"forks":120,"last_commit_at":121,"license":122,"difficulty_score":10,"env_os":123,"env_gpu":124,"env_ram":124,"env_deps":125,"category_tags":134,"github_topics":136,"view_count":139,"oss_zip_url":77,"oss_zip_packed_at":77,"status":17,"created_at":140,"updated_at":141,"faqs":142,"releases":171},1114,"opendr-eu\u002Fopendr","opendr","A modular, open and non-proprietary toolkit for core robotic functionalities by harnessing deep learning","OpenDR 是一个面向机器人开发的开源工具集，专注于通过深度学习技术增强机器人的感知与决策能力。它为开发者和研究人员提供了模块化的工具链，能够将深度学习框架（如 PyTorch、TensorFlow）与机器人操作系统（ROS）无缝衔接，帮助机器人实现更智能的人机交互、环境感知和自主决策。\n\n传统机器人开发在复杂场景下的适应性存在局限，而 OpenDR 通过集成深度学习模型，提升了机器人在动态环境中的实时感知能力，例如识别人类动作、理解场景语义，以及基于环境反馈优化行为策略。这种能力特别适用于医疗辅助、农业自动化和智能制造等需要高灵活性的场景。\n\nOpenDR 的 Python 接口设计简洁直观，适合快速开发与原型验证；同时提供 C API 满足高性能计算需求，适合对实时性要求严苛的工业应用。工具集内置丰富的 ROS 节点和 Webots 仿真支持，降低了机器人开发的技术门槛，即使是非专业开发者也能快速上手。\n\n其技术亮点包括：支持 ONNX 模型格式实现跨平台部署，兼容 OpenAI Gym 接口便于强化学习实验，以及模块化架构带来的灵活扩展性。目前可通过源码、pip 或 Docke","OpenDR 是一个面向机器人开发的开源工具集，专注于通过深度学习技术增强机器人的感知与决策能力。它为开发者和研究人员提供了模块化的工具链，能够将深度学习框架（如 PyTorch、TensorFlow）与机器人操作系统（ROS）无缝衔接，帮助机器人实现更智能的人机交互、环境感知和自主决策。\n\n传统机器人开发在复杂场景下的适应性存在局限，而 OpenDR 通过集成深度学习模型，提升了机器人在动态环境中的实时感知能力，例如识别人类动作、理解场景语义，以及基于环境反馈优化行为策略。这种能力特别适用于医疗辅助、农业自动化和智能制造等需要高灵活性的场景。\n\nOpenDR 的 Python 接口设计简洁直观，适合快速开发与原型验证；同时提供 C API 满足高性能计算需求，适合对实时性要求严苛的工业应用。工具集内置丰富的 ROS 节点和 Webots 仿真支持，降低了机器人开发的技术门槛，即使是非专业开发者也能快速上手。\n\n其技术亮点包括：支持 ONNX 模型格式实现跨平台部署，兼容 OpenAI Gym 接口便于强化学习实验，以及模块化架构带来的灵活扩展性。目前可通过源码、pip 或 Docker 三种方式安装，适配 CPU\u002FGPU 环境。对于希望将人工智能技术落地到机器人实体的研究人员和工程师而言，OpenDR 提供了从算法训练到实际部署的完整技术路径。","\u003Cdiv align=\"center\">\n\n\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fopendr-eu_opendr_readme_f360f25d2a86.png\" width=\"400px\">\n\n**A modular, open and non-proprietary toolkit for core robotic functionalities by harnessing deep learning**\n______________________________________________________________________\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fwww.opendr.eu\u002F\">Website\u003C\u002Fa> •\n  \u003Ca href=\"docs\u002Freference\u002Finstallation.md\">Installation\u003C\u002Fa> •\n  \u003Ca href=\"projects\u002Fpython\">Python Examples\u003C\u002Fa> •\n  \u003Ca href=\"projects\u002Fopendr_ws\">ROS1\u003C\u002Fa> •\n  \u003Ca href=\"projects\u002Fopendr_ws_2\">ROS2\u003C\u002Fa> •\n  \u003Ca href=\"projects\u002Fc_api\">C API\u003C\u002Fa> •\n  \u003Ca href=\"docs\u002Freference\u002Fcustomize.md\">Customization\u003C\u002Fa> •\n  \n  \u003Ca href=\"docs\u002Freference\u002Fissues.md\">Known Issues\u003C\u002Fa> •\n  \u003Ca href=\"TRC.md\">Toolkit Review Committee\u003C\u002Fa> •\n  \u003Ca href=\"CHANGELOG.md\">Changelog\u003C\u002Fa> •\n  \u003Ca href=\"LICENSE\">License\u003C\u002Fa>\n\u003C\u002Fp>\n\n[![License](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-Apache_2.0-blue.svg)](https:\u002F\u002Fopensource.org\u002Flicenses\u002FApache-2.0)\n[![Test Suite (master)](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Factions\u002Fworkflows\u002Ftests_suite.yml\u002Fbadge.svg)](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Factions\u002Fworkflows\u002Ftests_suite.yml)\n[![Test Packages](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Factions\u002Fworkflows\u002Ftest_packages.yml\u002Fbadge.svg)](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Factions\u002Fworkflows\u002Ftest_packages.yml)\n[![DOI](https:\u002F\u002Fzenodo.org\u002Fbadge\u002FDOI\u002F10.5281\u002Fzenodo.7540781.svg)](https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.7540781)\n\n\u003C\u002Fdiv>\n\n## About\n\nThe aim of [OpenDR Project](https:\u002F\u002Fopendr.eu) is to develop a **modular, open** and **non-proprietary toolkit** for core **robotic functionalities** by harnessing **deep learning** to provide advanced perception and cognition capabilities, meeting in this way the general requirements of robotics applications in the applications areas of healthcare, agri-food and agile production.\nOpenDR provides the means to link the **robotics applications to software libraries** (deep learning frameworks, e.g., [PyTorch](https:\u002F\u002Fpytorch.org\u002F) and [Tensorflow](https:\u002F\u002Fwww.tensorflow.org\u002F)) to the **operating environment ([ROS](https:\u002F\u002Fwww.ros.org\u002F))**.\nOpenDR focuses on the **AI and Cognition core technology** in order to provide tools that make robotic systems cognitive, giving them the ability to:\n1. interact with people and environments by developing deep learning methods for **human centric and environment active perception and cognition**,\n2. **learn and categorize** by developing deep learning **tools for training and inference in common robotics settings**, and\n3. **make decisions and derive knowledge** by developing deep learning tools for cognitive robot action and decision making.\n\nAs a result, the developed OpenDR toolkit will also enable cooperative human-robot interaction as well as the development of cognitive mechatronics where sensing and actuation are closely coupled with cognitive systems thus contributing to another two core technologies beyond AI and Cognition.\nOpenDR aims to develop, train, deploy and evaluate deep learning models that improve the technical capabilities of the core technologies beyond the current state of the art.\n\n\n## Where to start?\n\nYou can start by [installing](docs\u002Freference\u002Finstallation.md) the OpenDR toolkit.\nOpenDR can be installed in the following ways:\n1. By *cloning* this repository (CPU\u002FGPU support)\n2. Using *pip* (CPU\u002FGPU support only)\n3. Using *docker* (CPU\u002FGPU support)\n\n\n## What OpenDR provides?\n\nOpenDR provides an intuitive and easy to use **[Python interface](src\u002Fopendr)**, a **[C API](src\u002Fc_api) for performance critical application**, a wealth of **[usage examples and supporting tools](projects)**, as well as **ready-to-use [ROS nodes](projects\u002Fopendr_ws)**.\nOpenDR is built to support [Webots Open Source Robot Simulator](https:\u002F\u002Fcyberbotics.com\u002F), while it also extensively follows industry standards, such as [ONNX model format](https:\u002F\u002Fonnx.ai\u002F) and [OpenAI Gym Interface](https:\u002F\u002Fgym.openai.com\u002F).\n\n## How can I start using OpenDR?\n\nYou can find detailed documentation in OpenDR [wiki](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fwiki).\nThe main point of reference after installing the toolkit is the [tools index](docs\u002Freference\u002Findex.md).\nStarting from there, you can find detailed documentation for all the tools included in OpenDR.\n\n- If you are interested in ready-to-use ROS nodes, then you can directly jump to our [ROS1](projects\u002Fopendr_ws) and [ROS2](projects\u002Fopendr_ws_2) workspaces.\n- If you are interested for ready-to-use examples, then you can checkout the [projects](projects\u002Fpython) folder, which contains examples and tutorials for [perception](projects\u002Fpython\u002Fperception), [control](projects\u002Fpython\u002Fcontrol), [simulation](projects\u002Fpython\u002Fsimulation) and [hyperparameter tuning](projects\u002Fpython\u002Futils) tools.\n- If you want to explore our C API, then you explore the provided [C demos](projects\u002Fc_api).\n\n## How can I interface OpenDR?\n\nOpenDR is built upon Python.\nTherefore, the main OpenDR interface is written in Python and it is available through the [opendr](src\u002Fopendr) package.\nFurthermore, OpenDR provides [ROS1](projects\u002Fopendr_ws) and [ROS2](projects\u002Fopendr_ws_2) interfaces, as well as a [C interface](projects\u002Fc_api).\nNote that you can use as many tools as you wish at the same time, since there is no hardware limitation on the number of tools that can run at the same time.\nHowever, hardware limitations (e.g., GPU memory) might restrict the number of tools that can run at any given moment.\n\n## How to contribute\nPlease follow the instructions provided in the [wiki](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fwiki).\n\n## How to cite us\nIf you use OpenDR for your research, please cite the following paper that introduces OpenDR architecture and design:\n\u003Cpre>\n@inproceedings{opendr2022,\n  title={OpenDR: An Open Toolkit for Enabling High Performance, Low Footprint Deep Learning for Robotics},\n  author={Passalis, Nikolaos and Pedrazzi, Stefania and Babuska, Robert and Burgard, Wolfram and Dias, Daniel and Ferro, Francesco and Gabbouj, Moncef and Green, Ole and Iosifidis, Alexandros and Kayacan, Erdal and Kober, Jens and Michel, Olivier and Nikolaidis, Nikos and Nousi, Paraskevi and Pieters, Roel and Tzelepi, Maria and Valada, Abhinav and Tefas, Anastasios},\n    booktitle = {Proceedings of the 2022 IEEE\u002FRSJ International Conference on Intelligent Robots and Systems (to appear)},\n  year={2022}\n}\n\u003C\u002Fpre>\n\n\n\n## Acknowledgments\n*OpenDR project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 871449.*\n\u003Cdiv align=\"center\">\n\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fopendr-eu_opendr_readme_e4eb49b898c0.png\" height=\"70\"> \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fopendr-eu_opendr_readme_0a86ce2024fb.png\" height=\"70\">\n\u003C\u002Fdiv>\n","\u003Cdiv align=\"center\">\n\n\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fopendr-eu_opendr_readme_f360f25d2a86.png\" width=\"400px\">\n\n**一个通过深度学习实现核心机器人功能的模块化、开放且非专有的工具包**\n______________________________________________________________________\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fwww.opendr.eu\u002F\">官网\u003C\u002Fa> •\n  \u003Ca href=\"docs\u002Freference\u002Finstallation.md\">安装指南\u003C\u002Fa> •\n  \u003Ca href=\"projects\u002Fpython\">Python示例\u003C\u002Fa> •\n  \u003Ca href=\"projects\u002Fopendr_ws\">ROS1\u003C\u002Fa> •\n  \u003Ca href=\"projects\u002Fopendr_ws_2\">ROS2\u003C\u002Fa> •\n  \u003Ca href=\"projects\u002Fc_api\">C API\u003C\u002Fa> •\n  \u003Ca href=\"docs\u002Freference\u002Fcustomize.md\">自定义\u003C\u002Fa> •\n  \n  \u003Ca href=\"docs\u002Freference\u002Fissues.md\">已知问题\u003C\u002Fa> •\n  \u003Ca href=\"TRC.md\">工具包评审委员会\u003C\u002Fa> •\n  \u003Ca href=\"CHANGELOG.md\">更新日志\u003C\u002Fa> •\n  \u003Ca href=\"LICENSE\">许可证\u003C\u002Fa>\n\u003C\u002Fp>\n\n[![License](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-Apache_2.0-blue.svg)](https:\u002F\u002Fopensource.org\u002Flicenses\u002FApache-2.0)\n[![测试套件 (master)](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Factions\u002Fworkflows\u002Ftests_suite.yml\u002Fbadge.svg)](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Factions\u002Fworkflows\u002Ftests_suite.yml)\n[![测试包](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Factions\u002Fworkflows\u002Ftest_packages.yml\u002Fbadge.svg)](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Factions\u002Fworkflows\u002Ftest_packages.yml)\n[![DOI](https:\u002F\u002Fzenodo.org\u002Fbadge\u002FDOI\u002F10.5281\u002Fzenodo.7540781.svg)](https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.7540781)\n\n\u003C\u002Fdiv>\n\n## 关于项目\n\n[OpenDR项目](https:\u002F\u002Fopendr.eu)的目标是开发一个**模块化、开放**且**非专有工具包**（modular, open and non-proprietary toolkit），通过利用**深度学习**（deep learning）实现核心**机器人功能**（robotic functionalities），提供先进的感知和认知能力，从而满足医疗保健、农业食品和敏捷生产等应用领域对机器人技术的通用需求。\n\nOpenDR提供了将**机器人应用与软件库**（如深度学习框架 [PyTorch](https:\u002F\u002Fpytorch.org\u002F) 和 [Tensorflow](https:\u002F\u002Fwww.tensorflow.org\u002F)）连接到**操作系统环境**（[ROS](https:\u002F\u002Fwww.ros.org\u002F)）的方法。项目专注于**人工智能与认知核心科技**（AI and Cognition core technology），提供使机器人系统具备认知能力的工具，使其能够：\n1. 通过开发面向**以人为中心和环境主动感知与认知**（human centric and environment active perception and cognition）的深度学习方法实现人机交互；\n2. 通过开发适用于常见机器人场景的**训练与推理**（learn and categorize）深度学习工具实现学习与分类；\n3. 通过开发面向认知机器人动作与决策的深度学习工具实现决策与知识获取。\n\n因此，OpenDR工具包还将支持协作式人机交互以及认知机电一体化系统的开发，其中传感与执行器与认知系统紧密耦合，从而在人工智能与认知之外推动另外两项核心技术的发展。OpenDR旨在开发、训练、部署和评估深度学习模型，以提升核心技术的现有技术水平。\n\n## 如何开始？\n\n您可以通过[安装](docs\u002Freference\u002Finstallation.md)OpenDR工具包开始使用。OpenDR支持以下安装方式：\n1. *克隆*本仓库（支持CPU\u002FGPU）\n2. 使用*pip*（仅支持CPU\u002FGPU）\n3. 使用*docker*（支持CPU\u002FGPU）\n\n## OpenDR提供哪些功能？\n\nOpenDR提供直观易用的**[Python接口](src\u002Fopendr)**、**[性能敏感应用的C API](src\u002Fc_api)**、丰富的**[使用示例与支持工具](projects)**，以及**即用型 [ROS节点](projects\u002Fopendr_ws)**。OpenDR支持[Webots开源机器人模拟器](https:\u002F\u002Fcyberbotics.com\u002F)，并严格遵循行业标准，如[ONNX模型格式](https:\u002F\u002Fonnx.ai\u002F)和[OpenAI Gym接口](https:\u002F\u002Fgym.openai.com\u002F)。\n\n## 如何开始使用OpenDR？\n\n您可以在OpenDR [wiki](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fwiki)获取详细文档。安装工具包后的主要参考文档是[工具索引](docs\u002Freference\u002Findex.md)。通过该入口，您可以找到OpenDR所有工具的详细文档。\n\n- 若需要即用型ROS节点，可直接访问我们的[ROS1](projects\u002Fopendr_ws)和[ROS2](projects\u002Fopendr_ws_2)工作区；\n- 若需要即用型示例，可查看[projects](projects\u002Fpython)文件夹，其中包含[感知](projects\u002Fpython\u002Fperception)、[控制](projects\u002Fpython\u002Fcontrol)、[仿真](projects\u002Fpython\u002Fsimulation)和[超参数调优](projects\u002Fpython\u002Futils)工具的示例与教程；\n- 若需要探索C API，可查看提供的[C语言示例](projects\u002Fc_api)。\n\n## 如何与OpenDR交互？\n\nOpenDR基于Python构建，因此主接口为Python语言，通过[opendr](src\u002Fopendr)包提供。此外，OpenDR提供[ROS1](projects\u002Fopendr_ws)、[ROS2](projects\u002Fopendr_ws_2)接口和[C接口](projects\u002Fc_api)。您可同时使用任意数量的工具（无硬件限制），但硬件条件（如GPU内存）可能限制同时运行的工具数量。\n\n## 如何贡献代码\n\n请参考[wiki](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fwiki)中的贡献指南。\n\n## 如何引用本项目\n\n若在研究中使用OpenDR，请引用以下介绍OpenDR架构与设计的论文：\n\u003Cpre>\n@inproceedings{opendr2022,\n  title={OpenDR: An Open Toolkit for Enabling High Performance, Low Footprint Deep Learning for Robotics},\n  author={Passalis, Nikolaos and Pedrazzi, Stefania and Babuska, Robert and Burgard, Wolfram and Dias, Daniel and Ferro, Francesco and Gabbouj, Moncef and Green, Ole and Iosifidis, Alexandros and Kayacan, Erdal and Kober, Jens and Michel, Olivier and Nikolaidis, Nikos and Nousi, Paraskevi and Pieters, Roel and Tzelepi, Maria and Valada, Abhinav and Tefas, Anastasios},\n    booktitle = {Proceedings of the 2022 IEEE\u002FRSJ International Conference on Intelligent Robots and Systems (to appear)},\n  year={2022}\n}\n\u003C\u002Fpre>\n\n## 致谢\n*OpenDR项目获得欧盟地平线2020研究与创新计划（资助协议No 871449）资助。*\n\u003Cdiv align=\"center\">\n\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fopendr-eu_opendr_readme_e4eb49b898c0.png\" height=\"70\"> \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fopendr-eu_opendr_readme_0a86ce2024fb.png\" height=\"70\">\n\u003C\u002Fdiv>","# OpenDR 快速上手指南\n\n## 环境准备\n### 系统要求\n- 操作系统：Linux (推荐Ubuntu 18.04\u002F20.04)\n- 硬件：支持CPU\u002FGPU（CUDA 11.1+）\n- Python：3.6-3.9\n- 依赖工具：`python3-pip`, `git`, `docker`（可选）\n\n### 前置依赖安装\n```bash\nsudo apt-get update\nsudo apt-get install -y python3-pip git\n```\n\n## 安装步骤\n### 方式一：源码安装（推荐）\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr.git\ncd opendr\npip3 install -e .\n```\n\n### 方式二：PyPI 安装（国内加速版）\n```bash\npip3 install opendr -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple\n```\n\n### 方式三：Docker 镜像\n```bash\ndocker pull opendr\u002Fopendr:latest\n```\n\n## 基本使用\n### Python 快速示例\n```bash\n# 进入示例目录\ncd projects\u002Fpython\u002Fperception\n\n# 运行图像分类示例（需预装OpenCV）\npython3 example_classification.py\n```\n\n### ROS 节点启动（以ROS1为例）\n```bash\n# 构建工作空间\ncd projects\u002Fopendr_ws\ncatkin_make\n\n# 启动目标检测节点\nsource devel\u002Fsetup.bash\nroslaunch opendr_perception_example.launch\n```\n\n### 验证安装\n```python\n# Python交互式验证\npython3 -c \"import opendr; print(opendr.__version__)\"\n```\n\n> 📌 示例代码输出应显示当前版本号（如：0.2.0），表示安装成功\n\n## 附注\n- GPU加速需额外安装CUDA依赖\n- 完整文档请参考 [OpenDR Wiki](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fwiki)\n- 国内用户建议优先使用清华源加速PyPI安装","某医疗科技公司正在开发一款医院自主导航送药机器人，需要实现复杂动态环境下的实时避障与路径规划。\n\n### 没有 opendr 时\n- 需要手动整合 PyTorch\u002FTensorFlow 框架与 ROS 系统，环境感知模块开发耗时超过 3 个月\n- 基于传统 SLAM 的避障系统在夜间\u002F弱光环境下误判率高达 37%，导致频繁急停\n- 不同品牌传感器数据需要定制化驱动程序，多源异构数据融合耗用团队 40% 开发时间\n- 模型训练与部署需要维护两套代码库，算法迭代周期长达 2 周\n\n### 使用 opendr 后\n- 通过内置 ROS 节点直接调用预训练的环境感知模型（如 PointPillars 3D 目标检测），2 周内完成系统集成\n- 采用 opendr 提供的光照鲁棒性增强模块，夜间避障准确率提升至 98.5%\n- 使用统一的传感器抽象层接口，将多品牌激光雷达、RGB-D 相机的数据处理代码量减少 70%\n- 基于 ONNX 格式的模型转换工具链，实现训练（PyTorch）与推理（TensorRT）的无缝衔接，迭代周期缩短至 3 天\n\nopendr 通过提供开箱即用的深度学习机器人功能模块和标准化接口，使复杂场景下的机器人开发效率提升 5 倍以上，同时显著降低算法部署门槛。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fopendr-eu_opendr_0a86ce20.png","opendr-eu","OpenDR","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Fopendr-eu_4d791445.png","OpenDR H2020 Research Project",null,"https:\u002F\u002Fopendr.eu","https:\u002F\u002Fgithub.com\u002Fopendr-eu",[81,85,89,93,97,101,105,109,113,116],{"name":82,"color":83,"percentage":84},"Python","#3572A5",55.4,{"name":86,"color":87,"percentage":88},"Jupyter Notebook","#DA5B0B",35.6,{"name":90,"color":91,"percentage":92},"C++","#f34b7d",4.8,{"name":94,"color":95,"percentage":96},"EmberScript","#FFF4F3",2.2,{"name":98,"color":99,"percentage":100},"C","#555555",1.1,{"name":102,"color":103,"percentage":104},"Makefile","#427819",0.3,{"name":106,"color":107,"percentage":108},"Shell","#89e051",0.2,{"name":110,"color":111,"percentage":112},"CMake","#DA3434",0.1,{"name":114,"color":115,"percentage":112},"GLSL","#5686a5",{"name":117,"color":118,"percentage":112},"Cython","#fedf5b",725,104,"2026-04-08T04:40:45","Apache-2.0","Linux, macOS, Windows","未说明",{"notes":126,"python":127,"dependencies":128},"支持通过源码、pip或docker安装，推荐使用ROS1\u002FROS2环境。C API适用于性能敏感场景，GPU支持需自行配置CUDA环境。","3.8+",[129,130,131,132,133],"PyTorch","TensorFlow","ROS","ONNX","OpenAI Gym",[135,14],"其他",[137,138],"deep-learning","robotics",4,"2026-03-27T02:49:30.150509","2026-04-11T18:32:44.911704",[143,148,153,158,163,167],{"id":144,"question_zh":145,"answer_zh":146,"source_url":147},5023,"使用rgbd_hand_gesture_recognition.py时无法获取\u002Fopendr\u002Fgestures输出怎么办？","模型对距离和拍摄角度敏感，建议保持与训练数据相似的设置：正面视角、人物居中。可参考示例数据集：https:\u002F\u002Fdata.mendeley.com\u002Fdatasets\u002Fndrczc35bt\u002F1","https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fissues\u002F303",{"id":149,"question_zh":150,"answer_zh":151,"source_url":152},5024,"OpenDR不同算法的图像输入格式如何统一？","已通过添加opencv()方法实现格式转换，确保接口统一为NCHW和RGB格式。具体使用方法可参考更新后的示例代码。","https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fissues\u002F131",{"id":154,"question_zh":155,"answer_zh":156,"source_url":157},5025,"Docker镜像中导入YOLOv5DetectorLearner报错ModuleNotFoundError如何解决？","需安装GluonCV依赖。运行Docker时添加X11挂载参数：\nsudo docker run -it -v \u002Ftmp\u002F.X11-unix:\u002Ftmp\u002F.X11-unix -v \"$HOME\u002F.Xauthority:\u002Froot\u002F.Xauthority:ro\" --network host -e DISPLAY=unix$DISPLAY opendr\u002Fopendr-toolkit:cpu_nightly \u002Fbin\u002Fbash","https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fissues\u002F457",{"id":159,"question_zh":160,"answer_zh":161,"source_url":162},5026,"Nanodet C API编译时提示找不到torch\u002Fscript.h怎么办？","该问题已在PR #418 中修复，建议使用最新版本的Docker镜像。可通过创建带test release标签的PR验证编译日志。","https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fissues\u002F414",{"id":164,"question_zh":165,"answer_zh":166,"source_url":157},5027,"Docker中运行OpenDR可视化显示失败如何处理？","可尝试禁用可视化功能（设置show=False），或使用以下命令挂载X11：\ndocker run -it -v \u002Ftmp\u002F.X11-unix:\u002Ftmp\u002F.X11-unix -e DISPLAY=$DISPLAY opendr\u002Fopendr-toolkit:cpu_nightly",{"id":168,"question_zh":169,"answer_zh":170,"source_url":147},5028,"如何查看OpenDR支持的RGB手势识别示例？","文档已更新包含手势示例，访问：https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fblob\u002Fros1-fixes-no-output-rgbd-hand-gesture-recognition\u002Fdocs\u002Freference\u002Frgbd-hand-gesture-learner.md",[172,177,182,187,192,197,202],{"id":173,"version":174,"summary_zh":175,"released_at":176},203189,"v3.0.0","Released on Dec, 4th, 2023.\r\n\r\n- New Features:  \r\n  - Binary High Resolution Learner ([#402](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F402))\r\n  - ROS2 node for EfficientLPS ([#404](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F404))\r\n  - Fall and wave detection ROS nodes ([#423](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F423))\r\n  - Continual SLAM: Adds a new SLAM tool for Continual SLAM ([#424](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F424))\r\n  - Add RGB gesture recognition ([#436](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F436))\r\n  - FSeq2-NMS ([#442](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F442))\r\n  - Intent recognition tool ([#443](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F443))\r\n  - Robotti human detection simulation demo ([#451](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F451))\r\n  - Object Detection 2D Class Filtering ([#467](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F467))\r\n  - RL-based Learner for Active Face Recognition ([#473](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F473))\r\n  - YOLOv5s Inference Demo with Optimized Weights for Agricultural Use ([#476]([#](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F)476))\r\n  - Adaptive HR Pose Estimation ([#479](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F479))\r\n- Enhancements:\r\n  - Wave detection demo based on pose estimation ([#394](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F394))\r\n  - Facial expression recognition demo update ([#405](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F405))\r\n  - Object detection 2d camera demos ([#408](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F408))\r\n  - High Resolution Pose Estimation webcam demo ([#409](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F409))\r\n  - ROS nodes FPS performance measurements ([#419](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F419))\r\n  - Refactoring: pythonic joins in `test_clang_format.py`\u002F`test_cppcheck.py` ([#455](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F455))\r\n  - Test-tools improvement ([#456](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F456)) \r\n  - Adding prompt when transcribe with Whisper ([#462](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F462))\r\n \r\n- Bug Fixes:\r\n  - Fix package creator and sources ([#390](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F390))\r\n  - Lightweight OpenPose tool fixes and improvements ([#392](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F392))\r\n  - Fall Detection - alternative infer input ([#397](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F397))\r\n  - Yolov5 training bugfix ([#401](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F401))\r\n  - Fix the dependency conflict of geffnet installation ([#410](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F410))\r\n  - Fix bug in GEM ROS2 node ([#420](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F420))\r\n  - Fix link to nanodet documentation ([#421](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F421))\r\n  - EfficientLPS panoptic segmentation coloring bug ([#426](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F426))\r\n  - Bump flask from 1.1.2 to 2.3.2 ([#430](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F430))\r\n  - Fix tests on master branch ([#438](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F438))\r\n  - Added unzip installation as base ubuntu dependency and tool tests fixes ([#454](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F454))\r\n  - Active face recognition demo and bug fixes on Face Recognition ([#459](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F459))\r\n  - GPU installation fix ([#463](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F463))\r\n  - Fix ROS1 nodes argparse issue with .launch files ([#465](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F465))\r\n  - Minor fix on yolov5 webcam demo ([#466](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F466))\r\n  - Apply cuDNN init fix to all Object Detectors 2D ([#469](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F469))\r\n  - Updated test_suite_develop.yml based on latest test_suite.yml ([#471](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F471))\r\n  - Fix fmpgmapping ([#472](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F472))\r\n  - Synchronization and bugfixes ([#478](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F478))","2023-12-04T08:36:57",{"id":178,"version":179,"summary_zh":180,"released_at":181},203190,"v2.2.0","Released on July, 3rd, 2023.\r\n\r\n Dependency Updates ([#431](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F431)):\r\n    - Updated PyTorch version from 1.9.0 to 1.13.1.\r\n    - Updated Torchvision version from 0.10.0 to 0.14.1.\r\n    - Updated Torchaudio version from 0.9.0 to 0.13.1.\r\n    - Downgraded wheel version to 0.38.4 due to bug in recent version.","2023-07-03T13:30:56",{"id":183,"version":184,"summary_zh":185,"released_at":186},203191,"v2.1.0","Released on February, 22nd, 2023.\r\n\r\nNew Features:\r\n   - Added Efficient LiDAR Panoptic Segmentation ([#359](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F359)).\r\n   - Added Nanodet 2D Object Detection tool ([#352](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F352)).\r\n   - Added C API implementations of NanoDet 2D Object Detection tool ([#352](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F352)).\r\n   - Added C API implementations of forward pass of DETR 2D Object Detection tool ([#383](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F383)).\r\n   - Added C API implementations of forward pass of DeepSORT 2D Object Tracking tool ([#383](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F383)).\r\n   - Added C API implementations of forward pass of Lightweight OpenPose, Pose Estimator tool ([#383](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F383)).\r\n   - Added C API implementations of forward pass of X3D 2D Activity Recognition tool ([#383](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F383)).\r\n   - Added C API implementations of forward pass of Progressive Spatiotemporal GCN Skeleton-based Action Recognition tool ([#383](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F383)).\r\n   - Added Binary High Resolution Analysis tool ([#402](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F402)).\r\n   - Added Multi-Object-Search tool ([#363](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F363))\r\n\r\nEnhancements:\r\n   - Added support in C API for detection target structure and vector of detections ([#352](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F352))\r\n   - Added support in C API for tensor structure and vector of tensors ([#383](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F383))\r\n   - Added support in C API for json parser ([#383](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F383))","2023-02-22T08:09:47",{"id":188,"version":189,"summary_zh":190,"released_at":191},203192,"v2.0.0","Released on December, 31st, 2022.\r\n\r\nNew Features:\r\n   - Added YOLOv5 as an inference-only tool ([#360](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F360)).\r\n   - Added Continual Transformer Encoders ([#317](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F317)).\r\n   - Added Continual Spatio-Temporal Graph Convolutional Networks tool ([#370](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F370)).\r\n   - Added AmbiguityMeasure utility tool ([#361](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F361)).\r\n   - Added SiamRPN 2D tracking tool ([#367](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F367)).\r\n   - Added Facial Emotion Estimation tool ([#264](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F264)).\r\n   - Added High resolution pose estimation tool ([#356](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F356)).\r\n   - Added ROS2 nodes for all included tools ([#256](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F256)).\r\n   - Added missing ROS nodes and homogenized the interface across the tools ([#305](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fissues\u002F305)).\r\n\r\nBug Fixes:\r\n   - Fixed `BoundingBoxList`, `TrackingAnnotationList`, `BoundingBoxList3D` and `TrackingAnnotationList3D` confidence warnings ([#365](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F365)).\r\n   - Fixed undefined `image_id` and `segmentation` for COCO `BoundingBoxList` ([#365](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F365)).\r\n   - Fixed Continual X3D ONNX support ([#372](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F372)).\r\n   - Fixed several issues with ROS nodes and improved performance ([#305](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fissues\u002F305)).","2022-12-30T16:20:26",{"id":193,"version":194,"summary_zh":195,"released_at":196},203193,"v1.1.1","Released on June, 30th, 2022.\r\n\r\nFixes:\r\n    - Fix Efficient Panoptic Segmentation submodule commit ([#268](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F268)).\r\n    - Fix Face Recognition compilation error ([#267](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F267)).","2022-06-30T06:27:24",{"id":198,"version":199,"summary_zh":200,"released_at":201},203194,"v1.1","Released on June, 14th, 2022.\r\n\r\nNew Features:\r\n   - Added end-to-end planning tool (https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F223).\r\n   - Added seq2seq-nms module, along with other custom NMS implementations for 2D object detection.(https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F232).\r\n\r\nEnhancements:\r\n   - Added support for modular pip packages allowing tools to be installed separately (https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F201).\r\n   - Simplified the installation process for pip by including the appropriate post-installation scripts (https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F201).\r\n   - Improved the structure of the toolkit by moving io from utils to engine.helper (https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F201).\r\n   - Added support for post-install scripts and opendr dependencies in .ini files (https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F201).\r\n   - Updated toolkit to support CUDA 11.2 and improved GPU support (https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F215).\r\n   - Added a standalone pose-based fall detection tool (https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F237)\r\n\r\nBug Fixes:\r\n   - updated wheel building pipeline to include missing files and removed unnecessary dependencies (https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F200).\r\n   - panoptic_segmentation\u002Fefficient_ps: updated dataset preparation scripts to create correct validation ground truth (https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F221).\r\n   - panoptic_segmentation\u002Fefficient_ps: added specific configuration files for the provided pretrained models (https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F221).\r\n   - c_api\u002Fface_recognition: pass key by const reference in json_get_key_string() (https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F221).\r\n   - pose_estimation\u002Flightweight_open_pose: fixed height check on transformations.py according to original tool repo (https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F242).\r\n   - pose_estimation\u002Flightweight_open_pose: fixed two bugs where ONNX optimization failed on specific learner parameterization (https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F242).\r\n\r\nDependency Updates:\r\n   - heart anomaly detection: upgraded scikit-learn runtime dependency from 0.21.3 to 0.22 (https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F198).\r\n   - Relaxed all dependencies to allow future versions of non-critical tools to be used (https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fpull\u002F201).","2022-06-14T09:12:45",{"id":203,"version":204,"summary_zh":205,"released_at":206},203195,"v1.0","This is the first public version of OpenDR toolkit, which provides baseline deep learning tools for core robotic functionalities. The first version includes (among others):\r\n- an intuitive and easy-to-use **Python interface**\r\n- a wealth of **usage examples and supporting tools**\r\n- ready-to-use **ROS nodes**\r\n- a partial **C API**\r\n\r\nYou can find detailed **installation instructions** in [OpenDR repository](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fblob\u002Fmaster\u002Fdocs\u002Freference\u002Finstallation.md), while detailed documentation can be found in OpenDR [wiki](https:\u002F\u002Fgithub.com\u002Fopendr-eu\u002Fopendr\u002Fwiki).\r\n","2021-12-31T05:56:43"]