[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-Vincentqyw--image-matching-webui":3,"tool-Vincentqyw--image-matching-webui":64},[4,17,27,35,43,56],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":16},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。",162132,3,"2026-04-05T11:01:52",[13,14,15],"开发框架","图像","Agent","ready",{"id":18,"name":19,"github_repo":20,"description_zh":21,"stars":22,"difficulty_score":23,"last_commit_at":24,"category_tags":25,"status":16},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",138956,2,"2026-04-05T11:33:21",[13,15,26],"语言模型",{"id":28,"name":29,"github_repo":30,"description_zh":31,"stars":32,"difficulty_score":23,"last_commit_at":33,"category_tags":34,"status":16},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。",107662,"2026-04-03T11:11:01",[13,14,15],{"id":36,"name":37,"github_repo":38,"description_zh":39,"stars":40,"difficulty_score":23,"last_commit_at":41,"category_tags":42,"status":16},3704,"NextChat","ChatGPTNextWeb\u002FNextChat","NextChat 是一款轻量且极速的 AI 助手，旨在为用户提供流畅、跨平台的大模型交互体验。它完美解决了用户在多设备间切换时难以保持对话连续性，以及面对众多 AI 模型不知如何统一管理的痛点。无论是日常办公、学习辅助还是创意激发，NextChat 都能让用户随时随地通过网页、iOS、Android、Windows、MacOS 或 Linux 端无缝接入智能服务。\n\n这款工具非常适合普通用户、学生、职场人士以及需要私有化部署的企业团队使用。对于开发者而言，它也提供了便捷的自托管方案，支持一键部署到 Vercel 或 Zeabur 等平台。\n\nNextChat 的核心亮点在于其广泛的模型兼容性，原生支持 Claude、DeepSeek、GPT-4 及 Gemini Pro 等主流大模型，让用户在一个界面即可自由切换不同 AI 能力。此外，它还率先支持 MCP（Model Context Protocol）协议，增强了上下文处理能力。针对企业用户，NextChat 提供专业版解决方案，具备品牌定制、细粒度权限控制、内部知识库整合及安全审计等功能，满足公司对数据隐私和个性化管理的高标准要求。",87618,"2026-04-05T07:20:52",[13,26],{"id":44,"name":45,"github_repo":46,"description_zh":47,"stars":48,"difficulty_score":23,"last_commit_at":49,"category_tags":50,"status":16},2268,"ML-For-Beginners","microsoft\u002FML-For-Beginners","ML-For-Beginners 是由微软推出的一套系统化机器学习入门课程，旨在帮助零基础用户轻松掌握经典机器学习知识。这套课程将学习路径规划为 12 周，包含 26 节精炼课程和 52 道配套测验，内容涵盖从基础概念到实际应用的完整流程，有效解决了初学者面对庞大知识体系时无从下手、缺乏结构化指导的痛点。\n\n无论是希望转型的开发者、需要补充算法背景的研究人员，还是对人工智能充满好奇的普通爱好者，都能从中受益。课程不仅提供了清晰的理论讲解，还强调动手实践，让用户在循序渐进中建立扎实的技能基础。其独特的亮点在于强大的多语言支持，通过自动化机制提供了包括简体中文在内的 50 多种语言版本，极大地降低了全球不同背景用户的学习门槛。此外，项目采用开源协作模式，社区活跃且内容持续更新，确保学习者能获取前沿且准确的技术资讯。如果你正寻找一条清晰、友好且专业的机器学习入门之路，ML-For-Beginners 将是理想的起点。",84991,"2026-04-05T10:45:23",[14,51,52,53,15,54,26,13,55],"数据工具","视频","插件","其他","音频",{"id":57,"name":58,"github_repo":59,"description_zh":60,"stars":61,"difficulty_score":10,"last_commit_at":62,"category_tags":63,"status":16},3128,"ragflow","infiniflow\u002Fragflow","RAGFlow 是一款领先的开源检索增强生成（RAG）引擎，旨在为大语言模型构建更精准、可靠的上下文层。它巧妙地将前沿的 RAG 技术与智能体（Agent）能力相结合，不仅支持从各类文档中高效提取知识，还能让模型基于这些知识进行逻辑推理和任务执行。\n\n在大模型应用中，幻觉问题和知识滞后是常见痛点。RAGFlow 通过深度解析复杂文档结构（如表格、图表及混合排版），显著提升了信息检索的准确度，从而有效减少模型“胡编乱造”的现象，确保回答既有据可依又具备时效性。其内置的智能体机制更进一步，使系统不仅能回答问题，还能自主规划步骤解决复杂问题。\n\n这款工具特别适合开发者、企业技术团队以及 AI 研究人员使用。无论是希望快速搭建私有知识库问答系统，还是致力于探索大模型在垂直领域落地的创新者，都能从中受益。RAGFlow 提供了可视化的工作流编排界面和灵活的 API 接口，既降低了非算法背景用户的上手门槛，也满足了专业开发者对系统深度定制的需求。作为基于 Apache 2.0 协议开源的项目，它正成为连接通用大模型与行业专有知识之间的重要桥梁。",77062,"2026-04-04T04:44:48",[15,14,13,26,54],{"id":65,"github_repo":66,"name":67,"description_en":68,"description_zh":69,"ai_summary_zh":70,"readme_en":71,"readme_zh":72,"quickstart_zh":73,"use_case_zh":74,"hero_image_url":75,"owner_login":76,"owner_name":77,"owner_avatar_url":78,"owner_bio":79,"owner_company":80,"owner_location":81,"owner_email":82,"owner_twitter":83,"owner_website":81,"owner_url":84,"languages":85,"stars":112,"forks":113,"last_commit_at":114,"license":115,"difficulty_score":23,"env_os":116,"env_gpu":117,"env_ram":117,"env_deps":118,"category_tags":132,"github_topics":133,"view_count":10,"oss_zip_url":81,"oss_zip_packed_at":81,"status":16,"created_at":148,"updated_at":149,"faqs":150,"releases":175},882,"Vincentqyw\u002Fimage-matching-webui","image-matching-webui","🤗 image matching webui","image-matching-webui 是一个基于 Web 界面的图像匹配工具，能够帮助用户在两幅图像之间自动寻找并可视化关键点的对应关系。它通过集成多种前沿的图像匹配算法，让用户无需编写代码即可快速完成图像特征匹配，并直观地查看匹配结果。\n\n这个工具主要解决了传统图像匹配流程中需要手动编程、环境配置复杂、算法切换不便等问题。用户只需上传两张图片（支持本地文件或摄像头实时拍摄），从下拉菜单中选择一个匹配算法，系统便会自动提取并显示两幅图像之间的特征点匹配连线，大大降低了图像匹配任务的技术门槛。\n\nimage-matching-webui 非常适合计算机视觉领域的研究人员、算法工程师、学生以及相关技术爱好者使用。研究人员可以方便地对比不同算法的效果；开发者能快速验证和集成匹配模块；学生和初学者则能通过直观的界面理解图像匹配的基本概念。对于需要简单图像比对需求的普通用户，它也是一个易于上手的实用工具。\n\n其技术亮点在于集成了大量最新且高性能的匹配算法，例如 RIPE、RDD、OmniGlue、XFeat 等，覆盖了 2024-2025 年 CVPR、ICCV 等顶级会议的最新成果。工具基","image-matching-webui 是一个基于 Web 界面的图像匹配工具，能够帮助用户在两幅图像之间自动寻找并可视化关键点的对应关系。它通过集成多种前沿的图像匹配算法，让用户无需编写代码即可快速完成图像特征匹配，并直观地查看匹配结果。\n\n这个工具主要解决了传统图像匹配流程中需要手动编程、环境配置复杂、算法切换不便等问题。用户只需上传两张图片（支持本地文件或摄像头实时拍摄），从下拉菜单中选择一个匹配算法，系统便会自动提取并显示两幅图像之间的特征点匹配连线，大大降低了图像匹配任务的技术门槛。\n\nimage-matching-webui 非常适合计算机视觉领域的研究人员、算法工程师、学生以及相关技术爱好者使用。研究人员可以方便地对比不同算法的效果；开发者能快速验证和集成匹配模块；学生和初学者则能通过直观的界面理解图像匹配的基本概念。对于需要简单图像比对需求的普通用户，它也是一个易于上手的实用工具。\n\n其技术亮点在于集成了大量最新且高性能的匹配算法，例如 RIPE、RDD、OmniGlue、XFeat 等，覆盖了 2024-2025 年 CVPR、ICCV 等顶级会议的最新成果。工具基于 Gradio 构建，提供了友好的图形界面，并支持通过 Hugging Face Spaces 在线体验、PyPI 安装或 Docker 部署，兼顾了易用性与灵活性。","\u003C!-- [![Contributors][contributors-shield]][contributors-url]\n[![Forks][forks-shield]][forks-url]\n[![Stargazers][stars-shield]][stars-url]\n[![Issues][issues-shield]][issues-url] -->\n\n\u003Cp align=\"center\">\n  \u003Ch1 align=\"center\">\u003Cbr>\u003Cins>Image Matching WebUI\u003C\u002Fins>\n  \u003Cbr>Matching Keypoints between two images\u003C\u002Fh1>\n\u003C\u002Fp>\n\u003Cdiv align=\"center\">\n  \u003Ca target=\"_blank\" href=\"https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Factions\u002Fworkflows\u002Frelease.yml\">\u003Cimg src=\"https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Factions\u002Fworkflows\u002Frelease.yml\u002Fbadge.svg\" alt=\"PyPI Release\">\u003C\u002Fa>\n  \u003Ca target=\"_blank\" href='https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FRealcat\u002Fimage-matching-webui'>\u003Cimg src='https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F%F0%9F%A4%97%20Hugging%20Face-Spaces-blue'>\u003C\u002Fa>\n  \u003Ca target=\"_blank\" href=\"https:\u002F\u002Fpypi.org\u002Fproject\u002Fimcui\">\u003Cimg alt=\"PyPI - Version\" src=\"https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fv\u002Fimcui?style=flat&logo=pypi&label=imcui&link=https%3A%2F%2Fpypi.org%2Fproject%2Fimcui\">\u003C\u002Fa>\n  \u003Ca target=\"_blank\" href=\"https:\u002F\u002Fhub.docker.com\u002Fr\u002Fvincentqin\u002Fimage-matching-webui\">\u003Cimg alt=\"Docker Image Version\" src=\"https:\u002F\u002Fimg.shields.io\u002Fdocker\u002Fv\u002Fvincentqin\u002Fimage-matching-webui?sort=date&arch=amd64&logo=docker&label=imcui&link=https%3A%2F%2Fhub.docker.com%2Fr%2Fvincentqin%2Fimage-matching-webui\">\u003C\u002Fa>\n  \u003Ca target=\"_blank\" href=\"https:\u002F\u002Fpepy.tech\u002Fprojects\u002Fimcui\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FVincentqyw_image-matching-webui_readme_e579563a42fe.png\" alt=\"PyPI Downloads\">\u003C\u002Fa>\n  \u003Ca target=\"_blank\" href=\"https:\u002F\u002Fdeepwiki.com\u002FVincentqyw\u002Fimage-matching-webui\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDeepWiki-imcui-blue.svg\" alt=\"DeepWiki\">\u003C\u002Fa>\n\u003C\u002Fdiv>\n\n## Description\n\n`Image Matching WebUI (IMCUI)` efficiently matches image pairs using multiple famous image matching algorithms. The tool features a Graphical User Interface (GUI) designed using [gradio](https:\u002F\u002Fgradio.app\u002F). You can effortlessly select two images and a matching algorithm and obtain a precise matching result.\n**Note**: the images source can be either local images or webcam images.\n\nTry it on\n\u003Ca href='https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FRealcat\u002Fimage-matching-webui'>\u003Cimg src='https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F%F0%9F%A4%97%20Hugging%20Face-Spaces-blue'>\u003C\u002Fa>\n\u003Ca target=\"_blank\" href=\"https:\u002F\u002Flightning.ai\u002Frealcat\u002Fstudios\u002Fimage-matching-webui\">\u003Cimg src=\"https:\u002F\u002Fpl-bolts-doc-images.s3.us-east-2.amazonaws.com\u002Fapp-2\u002Fstudio-badge.svg\" alt=\"Open In Studio\"\u002F>\u003C\u002Fa>\n\nHere is a demo of the tool:\n\nhttps:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fassets\u002F18531182\u002F263534692-c3484d1b-cc00-4fdc-9b31-e5b7af07ecd9\n\nThe tool currently supports various popular image matching algorithms, namely:\n\n| Algorithm        | Supported | Conference\u002FJournal | Year | GitHub Link |\n|------------------|-----------|--------------------|------|-------------|\n| RIPE           | ✅ | ICCV    | 2025 | [Link](https:\u002F\u002Fgithub.com\u002Ffraunhoferhhi\u002FRIPE)  |\n| RDD            | ✅ | CVPR    | 2025 | [Link](https:\u002F\u002Fgithub.com\u002Fxtcpete\u002Frdd)  |\n| LiftFeat       | ✅ | ICRA    | 2025 | [Link](https:\u002F\u002Fgithub.com\u002Flyp-deeplearning\u002FLiftFeat) |\n| DaD            | ✅ | ARXIV   | 2025 | [Link](https:\u002F\u002Fgithub.com\u002FParskatt\u002Fdad) |\n| MINIMA         | ✅ | ARXIV   | 2024 | [Link](https:\u002F\u002Fgithub.com\u002FLSXI7\u002FMINIMA) |\n| XoFTR          | ✅ | CVPR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002FOnderT\u002FXoFTR) |\n| EfficientLoFTR | ✅ | CVPR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002Fzju3dv\u002FEfficientLoFTR) |\n| MASt3R         | ✅ | CVPR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002Fnaver\u002Fmast3r) |\n| DUSt3R         | ✅ | CVPR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002Fnaver\u002Fdust3r) |\n| OmniGlue       | ✅ | CVPR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fomniglue-onnx) |\n| XFeat          | ✅ | CVPR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002Fverlab\u002Faccelerated_features) |\n| RoMa           | ✅ | CVPR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002FVincentqyw\u002FRoMa) |\n| DeDoDe         | ✅ | 3DV     | 2024 | [Link](https:\u002F\u002Fgithub.com\u002FParskatt\u002FDeDoDe) |\n| Mickey         | ❌ | CVPR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002Fnianticlabs\u002Fmickey) |\n| GIM            | ✅ | ICLR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002Fxuelunshen\u002Fgim) |\n| ALIKED         | ✅ | ICCV    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002FShiaoming\u002FALIKED) |\n| LightGlue      | ✅ | ICCV    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002Fcvg\u002FLightGlue) |\n| DarkFeat       | ✅ | AAAI    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002FTHU-LYJ-Lab\u002FDarkFeat) |\n| SFD2           | ✅ | CVPR    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002Ffeixue94\u002Fsfd2) |\n| IMP            | ✅ | CVPR    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002Ffeixue94\u002Fimp-release) |\n| ASTR           | ❌ | CVPR    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002FASTR2023\u002FASTR) |\n| SEM            | ❌ | CVPR    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002FSEM2023\u002FSEM) |\n| DeepLSD        | ❌ | CVPR    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002Fcvg\u002FDeepLSD) |\n| GlueStick      | ✅ | ICCV    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002Fcvg\u002FGlueStick) |\n| ConvMatch      | ❌ | AAAI    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002FSuhZhang\u002FConvMatch) |\n| LoFTR          | ✅ | CVPR    | 2021 | [Link](https:\u002F\u002Fgithub.com\u002Fzju3dv\u002FLoFTR) |\n| SOLD2          | ✅ | CVPR    | 2021 | [Link](https:\u002F\u002Fgithub.com\u002Fcvg\u002FSOLD2) |\n| LineTR         | ❌ | RA-L    | 2021 | [Link](https:\u002F\u002Fgithub.com\u002Fyosungho\u002FLineTR) |\n| DKM            | ✅ | CVPR    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002FParskatt\u002FDKM) |\n| NCMNet         | ❌ | CVPR    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002Fxinliu29\u002FNCMNet) |\n| TopicFM        | ✅ | AAAI    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002FVincentqyw\u002FTopicFM) |\n| AspanFormer    | ✅ | ECCV    | 2022 | [Link](https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fml-aspanformer) |\n| LANet          | ✅ | ACCV    | 2022 | [Link](https:\u002F\u002Fgithub.com\u002Fwangch-g\u002Flanet) |\n| LISRD          | ❌ | ECCV    | 2022 | [Link](https:\u002F\u002Fgithub.com\u002Frpautrat\u002FLISRD) |\n| REKD           | ❌ | CVPR    | 2022 | [Link](https:\u002F\u002Fgithub.com\u002Fbluedream1121\u002FREKD) |\n| CoTR           | ✅ | ICCV    | 2021 | [Link](https:\u002F\u002Fgithub.com\u002Fubc-vision\u002FCOTR) |\n| ALIKE          | ✅ | TMM     | 2022 | [Link](https:\u002F\u002Fgithub.com\u002FShiaoming\u002FALIKE) |\n| RoRD           | ✅ | IROS    | 2021 | [Link](https:\u002F\u002Fgithub.com\u002FUditSinghParihar\u002FRoRD) |\n| SGMNet         | ✅ | ICCV    | 2021 | [Link](https:\u002F\u002Fgithub.com\u002Fvdvchen\u002FSGMNet) |\n| SuperPoint     | ✅ | CVPRW   | 2018 | [Link](https:\u002F\u002Fgithub.com\u002Fmagicleap\u002FSuperPointPretrainedNetwork) |\n| SuperGlue      | ✅ | CVPR    | 2020 | [Link](https:\u002F\u002Fgithub.com\u002Fmagicleap\u002FSuperGluePretrainedNetwork) |\n| D2Net          | ✅ | CVPR    | 2019 | [Link](https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fd2-net) |\n| R2D2           | ✅ | NeurIPS | 2019 | [Link](https:\u002F\u002Fgithub.com\u002Fnaver\u002Fr2d2) |\n| DISK           | ✅ | NeurIPS | 2020 | [Link](https:\u002F\u002Fgithub.com\u002Fcvlab-epfl\u002Fdisk) |\n| Key.Net        | ❌ | ICCV    | 2019 | [Link](https:\u002F\u002Fgithub.com\u002FaxelBarroso\u002FKey.Net) |\n| OANet          | ❌ | ICCV    | 2019 | [Link](https:\u002F\u002Fgithub.com\u002Fzjhthu\u002FOANet) |\n| SOSNet         | ✅ | CVPR    | 2019 | [Link](https:\u002F\u002Fgithub.com\u002Fscape-research\u002FSOSNet) |\n| HardNet        | ✅ | NeurIPS | 2017 | [Link](https:\u002F\u002Fgithub.com\u002FDagnyT\u002Fhardnet) |\n| SIFT           | ✅ | IJCV    | 2004 | [Link](https:\u002F\u002Fdocs.opencv.org\u002F4.x\u002Fda\u002Fdf5\u002Ftutorial_py_sift_intro.html) |\n\n\n## How to use\n\n### HuggingFace \u002F Lightning AI\n\nJust try it on \u003Ca href='https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FRealcat\u002Fimage-matching-webui'>\u003Cimg src='https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F%F0%9F%A4%97%20Hugging%20Face-Spaces-blue'>\u003C\u002Fa>\n\u003Ca target=\"_blank\" href=\"https:\u002F\u002Flightning.ai\u002Frealcat\u002Fstudios\u002Fimage-matching-webui\">\n  \u003Cimg src=\"https:\u002F\u002Fpl-bolts-doc-images.s3.us-east-2.amazonaws.com\u002Fapp-2\u002Fstudio-badge.svg\" alt=\"Open In Studio\"\u002F>\n\u003C\u002Fa>\n\nor deploy it locally following the instructions below.\n\n### Requirements\n\n- [Python 3.10+](https:\u002F\u002Fwww.python.org\u002Fdownloads\u002F)\n\n#### Install from pip [NEW]\n\nUpdate: now support install from [pip](https:\u002F\u002Fpypi.org\u002Fproject\u002Fimcui), just run:\n\n```bash\npip install imcui\n```\n\n#### Install from source\n\n``` bash\ngit clone --recursive https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui.git\ncd image-matching-webui\nconda env create -f environment.yaml\nconda activate imcui\npip install -e .\n```\n\nor using [docker](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fvincentqin\u002Fimage-matching-webui):\n\n```bash\ndocker pull vincentqin\u002Fimage-matching-webui:latest\n\n# Start the WebUI service\ndocker-compose up webui\n\n# Or run in the background\ndocker-compose up -d webui\n```\n\n\u003Cdetails>\n\u003Csummary>\u003Cstrong>More Docker Compose Commands\u003C\u002Fstrong> (click to expand)\u003C\u002Fsummary>\n\n```bash\n# Build and start the WebUI service\ndocker-compose up --build webui\n\n# Check the status of the WebUI service\ndocker-compose ps webui\n\n# View logs for the WebUI service\ndocker-compose logs webui\ndocker-compose logs -f webui  # Follow logs in real time\n\n# Stop the WebUI service\ndocker-compose stop webui\n\n# Restart the WebUI service\ndocker-compose restart webui\n\n# Remove the WebUI service container\ndocker-compose rm webui\n\n# Remove all containers\ndocker-compose down\n\n```\n\u003C\u002Fdetails>\n\n### Deploy to Railway\n\nDeploy to [Railway](https:\u002F\u002Frailway.app\u002F), setting up a `Custom Start Command` in `Deploy` section:\n\n``` bash\npython -m imcui.api.server\n```\n\n### Run demo\n``` bash\n# Using the package CLI (recommended)\nimcui\n\n# Or using the direct script\npython app.py\n```\nthen open http:\u002F\u002Flocalhost:7860 in your browser.\n\n![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FVincentqyw_image-matching-webui_readme_1562093fef88.jpg)\n\n### Command Line Interface\n\nThe `imcui` package provides a powerful command-line interface with various options:\n\n#### Basic Usage\n\n\u003Cdetails>\n\u003Csummary>\u003Cstrong>Basic Usage Commands\u003C\u002Fstrong> (click to expand)\u003C\u002Fsummary>\n\n```bash\n# Install the package\npip install imcui\n\n# Run with default settings\nimcui\n\n# Run with verbose output\nimcui --verbose\n\n# Run on a specific port\nimcui -p 7860\n\n# Run on a specific host\nimcui -s 127.0.0.1\n\n# Help\nimcui --help\n```\n\u003C\u002Fdetails>\n\n\n#### Command Line Options\n\n\u003Cdetails>\n\u003Csummary>\u003Cstrong>Basic Usage Commands\u003C\u002Fstrong> (click to expand)\u003C\u002Fsummary>\n\n| Option | Short | Default | Description |\n|--------|-------|---------|-------------|\n| `--server-name` | `-s` | `0.0.0.0` | Hostname or IP address to bind the server to |\n| `--server-port` | `-p` | `7860` | Port number to run the server on |\n| `--config` | `-c` | Auto-detected | Path to custom configuration YAML file |\n| `--example-data-root` | `-d` | `imcui\u002Fdatasets` | Root directory containing example datasets |\n| `--verbose` | `-v` | `False` | Enable verbose output for debugging |\n| `--version` | | | Show version information and exit |\n\n\u003C\u002Fdetails>\n\n\n### Add your own feature \u002F matcher\n\nI provide an example to add local feature in [imcui\u002Fhloc\u002Fextractors\u002Fexample.py](imcui\u002Fhloc\u002Fextractors\u002Fexample.py). Then add feature settings in `confs` in file [imcui\u002Fhloc\u002Fextract_features.py](imcui\u002Fhloc\u002Fextract_features.py). Last step is adding some settings to `matcher_zoo` in your configuration file.\n\n**Configuration file locations (in priority order):**\n1. Custom config file specified with `--config` parameter\n2. `config.yaml` in current directory\n3. `config\u002Fconfig.yaml` in current directory\n4. Package default config (`imcui\u002Fconfig\u002Fapp.yaml`)\n\n### Upload models\n\nIMCUI hosts all models on [Huggingface](https:\u002F\u002Fhuggingface.co\u002FRealcat\u002Fimcui_checkpoints).  You can upload your model to Huggingface and add it to the [Realcat\u002Fimcui_checkpoints](https:\u002F\u002Fhuggingface.co\u002FRealcat\u002Fimcui_checkpoints) repository.\n\n\n## Contributions welcome!\n\nExternal contributions are very much welcome. Please follow the [PEP8 style guidelines](https:\u002F\u002Fwww.python.org\u002Fdev\u002Fpeps\u002Fpep-0008\u002F) using a linter like flake8. This is a non-exhaustive list of features that might be valuable additions:\n\n- [x] support pip install command\n- [x] add [CPU CI](.github\u002Fworkflows\u002Fci.yml)\n- [x] add webcam support\n- [x] add [line feature matching](https:\u002F\u002Fgithub.com\u002FVincentqyw\u002FLineSegmentsDetection) algorithms\n- [x] example to add a new feature extractor \u002F matcher\n- [x] ransac to filter outliers\n- [ ] add [rotation images](https:\u002F\u002Fgithub.com\u002Fpidahbus\u002Fdeep-image-orientation-angle-detection) options before matching\n- [ ] support export matches to colmap ([#issue 6](https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fissues\u002F6))\n- [x] add config file to set default parameters\n- [x] dynamically load models and reduce GPU overload\n\nAdding local features \u002F matchers as submodules is very easy. For example, to add the [GlueStick](https:\u002F\u002Fgithub.com\u002Fcvg\u002FGlueStick):\n\n``` bash\ngit submodule add https:\u002F\u002Fgithub.com\u002Fcvg\u002FGlueStick.git imcui\u002Fthird_party\u002FGlueStick\n```\n\nIf remote submodule repositories are updated, don't forget to pull submodules with:\n\n``` bash\ngit submodule update --init --recursive  # init and download\ngit submodule update --remote  # update\n```\n\nIf you only want to update one submodule, use `git submodule update --remote imcui\u002Fthird_party\u002FGlueStick`.\n\nTo remove a submodule, follow these steps:\n\n\u003Cdetails>\n\u003Csummary>\u003Cstrong>More Remove Submodule Commands\u003C\u002Fstrong> (click to expand)\u003C\u002Fsummary>\n\n``` bash\ngit submodule deinit -f imcui\u002Fthird_party\u002FGlueStick\ngit rm -f imcui\u002Fthird_party\u002FGlueStick\nrm -rf .git\u002Fmodules\u002Fimcui\u002Fthird_party\u002FGlueStick\ngit add .gitmodules && \\\ngit commit -m \"Remove submodule imcui\u002Fthird_party\u002Fdust3r\"\n```\n\u003C\u002Fdetails>\n\n\nTo format code before committing, run:\n\n```bash\npre-commit run -a  # Auto-checks and fixes\n```\n\n## Contributors\n\n\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fgraphs\u002Fcontributors\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FVincentqyw_image-matching-webui_readme_b898eb4f9c1c.png\" \u002F>\n\u003C\u002Fa>\n\n## Resources\n- [Image Matching: Local Features & Beyond](https:\u002F\u002Fimage-matching-workshop.github.io)\n- [Long-term Visual Localization](https:\u002F\u002Fwww.visuallocalization.net)\n\n## Acknowledgement\n\nThis code is built based on [Hierarchical-Localization](https:\u002F\u002Fgithub.com\u002Fcvg\u002FHierarchical-Localization). We express our gratitude to the authors for their valuable source code.\n\n[contributors-shield]: https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fcontributors\u002FVincentqyw\u002Fimage-matching-webui.svg?style=for-the-badge\n[contributors-url]: https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fgraphs\u002Fcontributors\n[forks-shield]: https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fforks\u002FVincentqyw\u002Fimage-matching-webui.svg?style=for-the-badge\n[forks-url]: https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fnetwork\u002Fmembers\n[stars-shield]: https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FVincentqyw\u002Fimage-matching-webui.svg?style=for-the-badge\n[stars-url]: https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fstargazers\n[issues-shield]: https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues\u002FVincentqyw\u002Fimage-matching-webui.svg?style=for-the-badge\n[issues-url]: https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fissues\n","\u003C!-- [![Contributors][contributors-shield]][contributors-url]\n[![Forks][forks-shield]][forks-url]\n[![Stargazers][stars-shield]][stars-url]\n[![Issues][issues-shield]][issues-url] -->\n\n\u003Cp align=\"center\">\n  \u003Ch1 align=\"center\">\u003Cbr>\u003Cins>Image Matching WebUI\u003C\u002Fins>\n  \u003Cbr>两幅图像间的关键点匹配\u003C\u002Fh1>\n\u003C\u002Fp>\n\u003Cdiv align=\"center\">\n  \u003Ca target=\"_blank\" href=\"https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Factions\u002Fworkflows\u002Frelease.yml\">\u003Cimg src=\"https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Factions\u002Fworkflows\u002Frelease.yml\u002Fbadge.svg\" alt=\"PyPI Release\">\u003C\u002Fa>\n  \u003Ca target=\"_blank\" href='https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FRealcat\u002Fimage-matching-webui'>\u003Cimg src='https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F%F0%9F%A4%97%20Hugging%20Face-Spaces-blue'>\u003C\u002Fa>\n  \u003Ca target=\"_blank\" href=\"https:\u002F\u002Fpypi.org\u002Fproject\u002Fimcui\">\u003Cimg alt=\"PyPI - Version\" src=\"https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fv\u002Fimcui?style=flat&logo=pypi&label=imcui&link=https%3A%2F%2Fpypi.org%2Fproject%2Fimcui\">\u003C\u002Fa>\n  \u003Ca target=\"_blank\" href=\"https:\u002F\u002Fhub.docker.com\u002Fr\u002Fvincentqin\u002Fimage-matching-webui\">\u003Cimg alt=\"Docker Image Version\" src=\"https:\u002F\u002Fimg.shields.io\u002Fdocker\u002Fv\u002Fvincentqin\u002Fimage-matching-webui?sort=date&arch=amd64&logo=docker&label=imcui&link=https%3A%2F%2Fhub.docker.com%2Fr%2Fvincentqin%2Fimage-matching-webui\">\u003C\u002Fa>\n  \u003Ca target=\"_blank\" href=\"https:\u002F\u002Fpepy.tech\u002Fprojects\u002Fimcui\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FVincentqyw_image-matching-webui_readme_e579563a42fe.png\" alt=\"PyPI Downloads\">\u003C\u002Fa>\n  \u003Ca target=\"_blank\" href=\"https:\u002F\u002Fdeepwiki.com\u002FVincentqyw\u002Fimage-matching-webui\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDeepWiki-imcui-blue.svg\" alt=\"DeepWiki\">\u003C\u002Fa>\n\u003C\u002Fdiv>\n\n## 简介\n\n`Image Matching WebUI (IMCUI)` 是一款高效的图像匹配工具，集成了多种著名的图像匹配算法（Image Matching Algorithms）。该工具采用 [gradio](https:\u002F\u002Fgradio.app\u002F) 设计了图形用户界面（Graphical User Interface, GUI）。您可以轻松选择两幅图像和一个匹配算法，即可获得精确的匹配结果。\n**注意**：图像来源可以是本地图像或摄像头实时采集的图像。\n\n在线体验：\n\u003Ca href='https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FRealcat\u002Fimage-matching-webui'>\u003Cimg src='https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F%F0%9F%A4%97%20Hugging%20Face-Spaces-blue'>\u003C\u002Fa>\n\u003Ca target=\"_blank\" href=\"https:\u002F\u002Flightning.ai\u002Frealcat\u002Fstudios\u002Fimage-matching-webui\">\u003Cimg src=\"https:\u002F\u002Fpl-bolts-doc-images.s3.us-east-2.amazonaws.com\u002Fapp-2\u002Fstudio-badge.svg\" alt=\"Open In Studio\"\u002F>\u003C\u002Fa>\n\n工具演示视频：\n\nhttps:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fassets\u002F18531182\u002F263534692-c3484d1b-cc00-4fdc-9b31-e5b7af07ecd9\n\n该工具目前支持多种主流的图像匹配算法，具体如下：\n\n| 算法名称        | 是否支持 | 会议\u002F期刊 | 年份 | GitHub 链接 |\n|------------------|-----------|--------------------|------|-------------|\n| RIPE           | ✅ | ICCV    | 2025 | [Link](https:\u002F\u002Fgithub.com\u002Ffraunhoferhhi\u002FRIPE)  |\n| RDD            | ✅ | CVPR    | 2025 | [Link](https:\u002F\u002Fgithub.com\u002Fxtcpete\u002Frdd)  |\n| LiftFeat       | ✅ | ICRA    | 2025 | [Link](https:\u002F\u002Fgithub.com\u002Flyp-deeplearning\u002FLiftFeat) |\n| DaD            | ✅ | ARXIV   | 2025 | [Link](https:\u002F\u002Fgithub.com\u002FParskatt\u002Fdad) |\n| MINIMA         | ✅ | ARXIV   | 2024 | [Link](https:\u002F\u002Fgithub.com\u002FLSXI7\u002FMINIMA) |\n| XoFTR          | ✅ | CVPR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002FOnderT\u002FXoFTR) |\n| EfficientLoFTR | ✅ | CVPR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002Fzju3dv\u002FEfficientLoFTR) |\n| MASt3R         | ✅ | CVPR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002Fnaver\u002Fmast3r) |\n| DUSt3R         | ✅ | CVPR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002Fnaver\u002Fdust3r) |\n| OmniGlue       | ✅ | CVPR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fomniglue-onnx) |\n| XFeat          | ✅ | CVPR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002Fverlab\u002Faccelerated_features) |\n| RoMa           | ✅ | CVPR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002FVincentqyw\u002FRoMa) |\n| DeDoDe         | ✅ | 3DV     | 2024 | [Link](https:\u002F\u002Fgithub.com\u002FParskatt\u002FDeDoDe) |\n| Mickey         | ❌ | CVPR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002Fnianticlabs\u002Fmickey) |\n| GIM            | ✅ | ICLR    | 2024 | [Link](https:\u002F\u002Fgithub.com\u002Fxuelunshen\u002Fgim) |\n| ALIKED         | ✅ | ICCV    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002FShiaoming\u002FALIKED) |\n| LightGlue      | ✅ | ICCV    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002Fcvg\u002FLightGlue) |\n| DarkFeat       | ✅ | AAAI    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002FTHU-LYJ-Lab\u002FDarkFeat) |\n| SFD2           | ✅ | CVPR    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002Ffeixue94\u002Fsfd2) |\n| IMP            | ✅ | CVPR    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002Ffeixue94\u002Fimp-release) |\n| ASTR           | ❌ | CVPR    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002FASTR2023\u002FASTR) |\n| SEM            | ❌ | CVPR    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002FSEM2023\u002FSEM) |\n| DeepLSD        | ❌ | CVPR    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002Fcvg\u002FDeepLSD) |\n| GlueStick      | ✅ | ICCV    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002Fcvg\u002FGlueStick) |\n| ConvMatch      | ❌ | AAAI    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002FSuhZhang\u002FConvMatch) |\n| LoFTR          | ✅ | CVPR    | 2021 | [Link](https:\u002F\u002Fgithub.com\u002Fzju3dv\u002FLoFTR) |\n| SOLD2          | ✅ | CVPR    | 2021 | [Link](https:\u002F\u002Fgithub.com\u002Fcvg\u002FSOLD2) |\n| LineTR         | ❌ | RA-L    | 2021 | [Link](https:\u002F\u002Fgithub.com\u002Fyosungho\u002FLineTR) |\n| DKM            | ✅ | CVPR    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002FParskatt\u002FDKM) |\n| NCMNet         | ❌ | CVPR    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002Fxinliu29\u002FNCMNet) |\n| TopicFM        | ✅ | AAAI    | 2023 | [Link](https:\u002F\u002Fgithub.com\u002FVincentqyw\u002FTopicFM) |\n| AspanFormer    | ✅ | ECCV    | 2022 | [Link](https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fml-aspanformer) |\n| LANet          | ✅ | ACCV    | 2022 | [Link](https:\u002F\u002Fgithub.com\u002Fwangch-g\u002Flanet) |\n| LISRD          | ❌ | ECCV    | 2022 | [Link](https:\u002F\u002Fgithub.com\u002Frpautrat\u002FLISRD) |\n| REKD           | ❌ | CVPR    | 2022 | [Link](https:\u002F\u002Fgithub.com\u002Fbluedream1121\u002FREKD) |\n| CoTR           | ✅ | ICCV    | 2021 | [Link](https:\u002F\u002Fgithub.com\u002Fubc-vision\u002FCOTR) |\n| ALIKE          | ✅ | TMM     | 2022 | [Link](https:\u002F\u002Fgithub.com\u002FShiaoming\u002FALIKE) |\n| RoRD           | ✅ | IROS    | 2021 | [Link](https:\u002F\u002Fgithub.com\u002FUditSinghParihar\u002FRoRD) |\n| SGMNet         | ✅ | ICCV    | 2021 | [Link](https:\u002F\u002Fgithub.com\u002Fvdvchen\u002FSGMNet) |\n| SuperPoint     | ✅ | CVPRW   | 2018 | [Link](https:\u002F\u002Fgithub.com\u002Fmagicleap\u002FSuperPointPretrainedNetwork) |\n| SuperGlue      | ✅ | CVPR    | 2020 | [Link](https:\u002F\u002Fgithub.com\u002Fmagicleap\u002FSuperGluePretrainedNetwork) |\n| D2Net          | ✅ | CVPR    | 2019 | [Link](https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fd2-net) |\n| R2D2           | ✅ | NeurIPS | 2019 | [Link](https:\u002F\u002Fgithub.com\u002Fnaver\u002Fr2d2) |\n| DISK           | ✅ | NeurIPS | 2020 | [Link](https:\u002F\u002Fgithub.com\u002Fcvlab-epfl\u002Fdisk) |\n| Key.Net        | ❌ | ICCV    | 2019 | [Link](https:\u002F\u002Fgithub.com\u002FaxelBarroso\u002FKey.Net) |\n| OANet          | ❌ | ICCV    | 2019 | [Link](https:\u002F\u002Fgithub.com\u002Fzjhthu\u002FOANet) |\n| SOSNet         | ✅ | CVPR    | 2019 | [Link](https:\u002F\u002Fgithub.com\u002Fscape-research\u002FSOSNet) |\n| HardNet        | ✅ | NeurIPS | 2017 | [Link](https:\u002F\u002Fgithub.com\u002FDagnyT\u002Fhardnet) |\n| SIFT           | ✅ | IJCV    | 2004 | [Link](https:\u002F\u002Fdocs.opencv.org\u002F4.x\u002Fda\u002Fdf5\u002Ftutorial_py_sift_intro.html) |\n\n\n## 使用方法\n\n### HuggingFace \u002F Lightning AI\n\n直接在 \u003Ca href='https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FRealcat\u002Fimage-matching-webui'>\u003Cimg src='https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F%F0%9F%A4%97%20Hugging%20Face-Spaces-blue'>\u003C\u002Fa> 上试用\n\u003Ca target=\"_blank\" href=\"https:\u002F\u002Flightning.ai\u002Frealcat\u002Fstudios\u002Fimage-matching-webui\">\n  \u003Cimg src=\"https:\u002F\u002Fpl-bolts-doc-images.s3.us-east-2.amazonaws.com\u002Fapp-2\u002Fstudio-badge.svg\" alt=\"Open In Studio\"\u002F>\n\u003C\u002Fa>\n\n或按照以下说明在本地部署。\n\n### 环境要求\n\n- [Python 3.10+](https:\u002F\u002Fwww.python.org\u002Fdownloads\u002F)\n\n#### 通过 pip 安装 [新增]\n\n更新：现在支持从 [pip](https:\u002F\u002Fpypi.org\u002Fproject\u002Fimcui) 安装，只需运行：\n\n```bash\npip install imcui\n```\n\n#### 从源码安装\n\n``` bash\ngit clone --recursive https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui.git\ncd image-matching-webui\nconda env create -f environment.yaml\nconda activate imcui\npip install -e .\n```\n\n或使用 [docker](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fvincentqin\u002Fimage-matching-webui)：\n\n```bash\ndocker pull vincentqin\u002Fimage-matching-webui:latest\n\n# 启动 WebUI 服务\ndocker-compose up webui\n\n# 或在后台运行\ndocker-compose up -d webui\n```\n\n\u003Cdetails>\n\u003Csummary>\u003Cstrong>更多 Docker Compose 命令\u003C\u002Fstrong>（点击展开）\u003C\u002Fsummary>\n\n```bash\n# 构建并启动 WebUI 服务\ndocker-compose up --build webui\n\n# 检查 WebUI 服务状态\ndocker-compose ps webui\n\n# 查看 WebUI 服务日志\ndocker-compose logs webui\ndocker-compose logs -f webui  # 实时跟踪日志\n\n# 停止 WebUI 服务\ndocker-compose stop webui\n\n# 重启 WebUI 服务\ndocker-compose restart webui\n\n# 删除 WebUI 服务容器\ndocker-compose rm webui\n\n# 删除所有容器\ndocker-compose down\n\n```\n\u003C\u002Fdetails>\n\n### 部署到 Railway\n\n部署到 [Railway](https:\u002F\u002Frailway.app\u002F)，在 `Deploy` 部分设置 `Custom Start Command`：\n\n``` bash\npython -m imcui.api.server\n```\n\n### 运行演示\n``` bash\n# 使用包命令行界面（推荐）\nimcui\n\n# 或直接运行脚本\npython app.py\n```\n然后在浏览器中打开 http:\u002F\u002Flocalhost:7860。\n\n![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FVincentqyw_image-matching-webui_readme_1562093fef88.jpg)\n\n### 命令行界面\n\n`imcui` 包提供了强大的命令行界面，包含多种选项：\n\n#### 基本用法\n\n\u003Cdetails>\n\u003Csummary>\u003Cstrong>基本用法命令\u003C\u002Fstrong>（点击展开）\u003C\u002Fsummary>\n\n```bash\n# 安装包\npip install imcui\n\n# 使用默认设置运行\nimcui\n\n# 使用详细输出运行\nimcui --verbose\n\n# 在指定端口运行\nimcui -p 7860\n\n# 在指定主机运行\nimcui -s 127.0.0.1\n\n# 帮助\nimcui --help\n```\n\u003C\u002Fdetails>\n\n\n#### 命令行选项\n\n\u003Cdetails>\n\u003Csummary>\u003Cstrong>基本用法命令\u003C\u002Fstrong>（点击展开）\u003C\u002Fsummary>\n\n| 选项 | 简写 | 默认值 | 说明 |\n|--------|-------|---------|-------------|\n| `--server-name` | `-s` | `0.0.0.0` | 服务器绑定的主机名或 IP 地址 |\n| `--server-port` | `-p` | `7860` | 服务器运行的端口号 |\n| `--config` | `-c` | 自动检测 | 自定义配置 YAML 文件的路径 |\n| `--example-data-root` | `-d` | `imcui\u002Fdatasets` | 包含示例数据集的根目录 |\n| `--verbose` | `-v` | `False` | 启用详细输出以进行调试 |\n| `--version` | | | 显示版本信息并退出 |\n\n\u003C\u002Fdetails>\n\n\n### 添加你自己的特征（feature）\u002F匹配器（matcher）\n\n我提供了一个在 [imcui\u002Fhloc\u002Fextractors\u002Fexample.py](imcui\u002Fhloc\u002Fextractors\u002Fexample.py) 中添加局部特征（local feature）的示例。然后在文件 [imcui\u002Fhloc\u002Fextract_features.py](imcui\u002Fhloc\u002Fextract_features.py) 中的 `confs` 添加特征设置。最后一步是在你的配置文件中向 `matcher_zoo` 添加一些设置。\n\n**配置文件位置（按优先级排序）：**\n1. 使用 `--config` 参数指定的自定义配置文件\n2. 当前目录下的 `config.yaml`\n3. 当前目录下 `config\u002Fconfig.yaml`\n4. 包默认配置（`imcui\u002Fconfig\u002Fapp.yaml`）\n\n### 上传模型\n\nIMCUI 将所有模型托管在 [Huggingface](https:\u002F\u002Fhuggingface.co\u002FRealcat\u002Fimcui_checkpoints) 上。你可以将模型上传到 Huggingface 并添加到 [Realcat\u002Fimcui_checkpoints](https:\u002F\u002Fhuggingface.co\u002FRealcat\u002Fimcui_checkpoints) 仓库。\n\n\n## 欢迎贡献！\n\n非常欢迎外部贡献。请使用 flake8 等 linter 遵循 [PEP8 风格指南](https:\u002F\u002Fwww.python.org\u002Fdev\u002Fpeps\u002Fpep-0008\u002F)。以下是可能有价值的功能添加的非详尽列表：\n\n- [x] 支持 pip 安装命令\n- [x] 添加 [CPU CI](.github\u002Fworkflows\u002Fci.yml)\n- [x] 添加摄像头支持\n- [x] 添加 [线段特征匹配](https:\u002F\u002Fgithub.com\u002FVincentqyw\u002FLineSegmentsDetection) 算法\n- [x] 添加新特征提取器\u002F匹配器的示例\n- [x] 使用 RANSAC（Random Sample Consensus，随机抽样一致性）过滤异常值\n- [ ] 在匹配前添加 [旋转图像](https:\u002F\u002Fgithub.com\u002Fpidahbus\u002Fdeep-image-orientation-angle-detection) 选项\n- [ ] 支持将匹配结果导出到 COLMAP（[#issue 6](https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fissues\u002F6)）\n- [x] 添加配置文件以设置默认参数\n- [x] 动态加载模型并减少 GPU 负载\n\n将局部特征\u002F匹配器作为子模块（submodule）添加非常容易。例如，要添加 [GlueStick](https:\u002F\u002Fgithub.com\u002Fcvg\u002FGlueStick)：\n\n``` bash\ngit submodule add https:\u002F\u002Fgithub.com\u002Fcvg\u002FGlueStick.git imcui\u002Fthird_party\u002FGlueStick\n```\n\n如果远程子模块仓库有更新，别忘了使用以下命令拉取子模块：\n\n``` bash\ngit submodule update --init --recursive  # 初始化和下载\ngit submodule update --remote  # 更新\n```\n\n如果你只想更新一个子模块，使用 `git submodule update --remote imcui\u002Fthird_party\u002FGlueStick`。\n\n要删除子模块，请按照以下步骤操作：\n\n\u003Cdetails>\n\u003Csummary>\u003Cstrong>更多删除子模块命令\u003C\u002Fstrong>（点击展开）\u003C\u002Fsummary>\n\n``` bash\ngit submodule deinit -f imcui\u002Fthird_party\u002FGlueStick\ngit rm -f imcui\u002Fthird_party\u002FGlueStick\nrm -rf .git\u002Fmodules\u002Fimcui\u002Fthird_party\u002FGlueStick\ngit add .gitmodules && \\\ngit commit -m \"Remove submodule imcui\u002Fthird_party\u002Fdust3r\"\n```\n\u003C\u002Fdetails>\n\n\n在提交前格式化代码，运行：\n\n```bash\npre-commit run -a  # 自动检查和修复\n```\n\n## 贡献者\n\n\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fgraphs\u002Fcontributors\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FVincentqyw_image-matching-webui_readme_b898eb4f9c1c.png\" \u002F>\n\u003C\u002Fa>\n\n## 资源\n- [Image Matching: Local Features & Beyond](https:\u002F\u002Fimage-matching-workshop.github.io)\n- [Long-term Visual Localization](https:\u002F\u002Fwww.visuallocalization.net)\n\n## 致谢\n\n本代码基于 [Hierarchical-Localization](https:\u002F\u002Fgithub.com\u002Fcvg\u002FHierarchical-Localization) 构建。我们向原作者致以诚挚的感谢，感谢他们提供的宝贵源代码。\n\n[contributors-shield]: https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fcontributors\u002FVincentqyw\u002Fimage-matching-webui.svg?style=for-the-badge\n[contributors-url]: https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fgraphs\u002Fcontributors\n[forks-shield]: https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fforks\u002FVincentqyw\u002Fimage-matching-webui.svg?style=for-the-badge\n[forks-url]: https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fnetwork\u002Fmembers\n[stars-shield]: https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FVincentqyw\u002Fimage-matching-webui.svg?style=for-the-badge\n[stars-url]: https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fstargazers\n[issues-shield]: https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues\u002FVincentqyw\u002Fimage-matching-webui.svg?style=for-the-badge\n[issues-url]: https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fissues","# Image Matching WebUI 快速上手指南\n\n## 环境准备\n\n### 系统要求\n- **Python**: 3.10 或更高版本\n- **操作系统**: Linux \u002F macOS \u002F Windows（推荐 Linux）\n- **GPU**: 可选，支持 CUDA 加速\n\n### 前置依赖\n- [conda](https:\u002F\u002Fdocs.conda.io\u002Fen\u002Flatest\u002Fminiconda.html) 或 [miniconda](https:\u002F\u002Fdocs.conda.io\u002Fprojects\u002Fminiconda\u002Fen\u002Flatest\u002F)（推荐）\n- [Docker](https:\u002F\u002Fwww.docker.com\u002F)（可选，用于容器化部署）\n- [Git](https:\u002F\u002Fgit-scm.com\u002F)\n\n---\n\n## 安装步骤\n\n### 方式一：pip 安装（推荐，最简单）\n\n```bash\n# 使用清华镜像源加速（国内推荐）\npip install imcui -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple\n\n# 或使用官方源\npip install imcui\n```\n\n### 方式二：源码安装\n\n```bash\n# 克隆仓库（包含子模块）\ngit clone --recursive https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui.git\ncd image-matching-webui\n\n# 创建 conda 环境\nconda env create -f environment.yaml\nconda activate imcui\n\n# 安装包\npip install -e .\n```\n\n### 方式三：Docker 安装\n\n```bash\n# 拉取镜像（国内可使用 DaoCloud 等镜像加速）\ndocker pull vincentqin\u002Fimage-matching-webui:latest\n\n# 启动服务\ndocker-compose up webui\n\n# 后台运行\ndocker-compose up -d webui\n```\n\n---\n\n## 基本使用\n\n### 启动 WebUI\n\n```bash\n# 方式 1：使用 CLI 命令（推荐）\nimcui\n\n# 方式 2：直接运行脚本\npython app.py\n```\n\n服务启动后，在浏览器中打开：**http:\u002F\u002Flocalhost:7860**\n\n### 常用启动参数\n\n```bash\n# 指定端口\nimcui -p 8080\n\n# 指定主机地址\nimcui -s 127.0.0.1\n\n# 启用详细日志\nimcui --verbose\n\n# 查看帮助\nimcui --help\n```\n\n### 快速体验（无需本地安装）\n\n直接访问在线演示：\n- 🤗 [Hugging Face Spaces](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FRealcat\u002Fimage-matching-webui)\n- ⚡ [Lightning AI Studio](https:\u002F\u002Flightning.ai\u002Frealcat\u002Fstudios\u002Fimage-matching-webui)\n\n---\n\n## 使用流程\n\n1. **上传图片**：选择两张待匹配的图像（支持本地上传或摄像头拍摄）\n2. **选择算法**：从下拉菜单中选择特征提取器和匹配算法\n3. **运行匹配**：点击匹配按钮，查看关键点对应结果\n4. **导出结果**：支持可视化结果导出\n\n---\n\n## 支持的算法\n\n| 类型 | 代表算法 |\n|:---|:---|\n| **特征点+描述子** | SuperPoint, ALIKED, XFeat, DISK, R2D2, SIFT 等 |\n| **特征匹配器** | SuperGlue, LightGlue, LoFTR, EfficientLoFTR 等 |\n| **端到端匹配** | MASt3R, DUSt3R, RoMa, OmniGlue, DeDoDe 等 |\n| **线特征匹配** | SOLD2, GlueStick 等 |\n\n完整算法列表见项目 README。","某无人机测绘公司的算法工程师小李，需要为农业植保无人机开发一套视觉定位系统，用于在GPS信号弱的果园环境中实现精准悬停和航线跟踪。\n\n### 没有 image-matching-webui 时\n\n- 算法选型如同盲人摸象：LoFTR、SuperGlue、LightGlue等十几种匹配算法各有论文和官方代码，但环境依赖冲突严重，光是配好一个算法的运行环境就要折腾2-3天\n- 对比实验效率极低：想对比两种算法在果树冠层图像上的表现，需要分别写两套推理脚本，手动处理输入输出格式差异，单次对比周期长达一周\n- 现场调试举步维艰：带着笔记本去果园实地测试，发现算法对光照变化敏感，却无法快速验证其他算法是否更鲁棒，只能回公司后重新折腾环境\n- 团队协作沟通成本高：产品经理想看效果演示，小李只能截图论文里的结果，或者用命令行跑完后再用Photoshop拼对比图，反馈周期以天计算\n\n### 使用 image-matching-webui 后\n\n- 开箱即用的算法超市：浏览器打开即可同时调用MASt3R、XFeat、RoMa等20+种SOTA算法，无需关心底层依赖，30分钟内完成全量算法初筛\n- 拖拽式对比分析：上传两张果园航拍图，一键切换不同算法，匹配结果、关键点数量、推理耗时实时可视化，半天就能锁定最优候选方案\n- 现场即插即用验证：笔记本浏览器访问本地部署的WebUI，果园现场拍摄图像直接上传，5分钟验证算法在真实光照条件下的表现，快速迭代\n- 零门槛效果展示：产品经理、客户围坐一起，实时调整匹配阈值、对比不同算法结果，当场确认技术方案，决策周期从一周缩短到一小时\n\nimage-matching-webui 将原本需要数周的算法选型与验证流程压缩到数小时，让工程师把精力聚焦在业务优化而非环境配置上。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FVincentqyw_image-matching-webui_1562093f.jpg","Vincentqyw","Realcat","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002FVincentqyw_47167be0.png","    ⭐️Focusing on  Visual Localization, SfM and SLAM.","THU",null,"alpharealcat@gmail.com","AlphaRealcat","https:\u002F\u002Fgithub.com\u002FVincentqyw",[86,90,94,98,102,106,109],{"name":87,"color":88,"percentage":89},"Python","#3572A5",95.9,{"name":91,"color":92,"percentage":93},"C++","#f34b7d",2.9,{"name":95,"color":96,"percentage":97},"Jupyter Notebook","#DA5B0B",0.8,{"name":99,"color":100,"percentage":101},"Shell","#89e051",0.2,{"name":103,"color":104,"percentage":105},"Dockerfile","#384d54",0.1,{"name":107,"color":108,"percentage":105},"CMake","#DA3434",{"name":110,"color":111,"percentage":105},"Batchfile","#C1F12E",1236,118,"2026-04-01T03:52:02","Apache-2.0","Linux, macOS, Windows","未说明",{"notes":119,"python":120,"dependencies":121},"支持通过 pip 直接安装（pip install imcui）或从源码安装；提供 Docker 镜像部署方式；支持 HuggingFace Spaces 和 Lightning AI 在线体验；首次运行需从 HuggingFace 下载模型文件；支持 CPU 运行（已配置 CPU CI），但深度学习算法建议使用 GPU 加速；可通过 conda 环境管理依赖，具体依赖见 environment.yaml 文件","3.10+",[122,123,124,125,126,127,128,129,130,131],"gradio","torch","torchvision","opencv-python","numpy","scipy","huggingface_hub","pyyaml","tqdm","pillow",[14,13],[134,135,136,137,138,139,140,122,141,142,143,144,145,146,147],"feature-matching","image-matching","loftr","superglue","superpoint","keypoint-matching","visual-localization","lightglue","sift","topicfm","aspanformer","deep-learning","pose-estimation","kornia","2026-03-27T02:49:30.150509","2026-04-06T07:13:52.164430",[151,156,161,166,171],{"id":152,"question_zh":153,"answer_zh":154,"source_url":155},3816,"Windows 上运行报错，出现 OPENSSL_Applink 和 pycolmap 相关错误如何解决？","这是 pycolmap 安装问题导致的。解决方法：\n1. 使用 conda 安装 pycolmap：`conda install -c conda-forge pycolmap`\n2. 如果仍无法安装，请升级 PyTorch 版本到 2.4.0，该问题已在 PR #75 中修复","https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fissues\u002F21",{"id":157,"question_zh":158,"answer_zh":159,"source_url":160},3817,"如何在无网络环境的离线工作站部署运行？","有两种解决方案：\n1. **使用 Docker 镜像**：`docker pull vincentqin\u002Fimage-matching-webui`，适合无需修改源码的场景\n2. **使用离线版 Gradio**：将 requirements.txt 中的 `gradio` 替换为 `gradio-offline`，然后正常安装依赖即可运行\n\n如需修改源码（如适配卫星图像匹配），推荐使用第二种方案。","https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fissues\u002F133",{"id":162,"question_zh":163,"answer_zh":164,"source_url":165},3818,"出现 pydantic_core.ValidationError: event_id Field required 错误如何解决？","这是 Gradio 版本兼容性问题。解决方法：\n```bash\npip install gradio>=4.12.0\n```\n升级 Gradio 到 4.12.0 或更高版本即可修复。注意：使用 4.7.1 版本可能导致示例无法正常显示。","https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fissues\u002F20",{"id":167,"question_zh":168,"answer_zh":169,"source_url":170},3819,"如何验证单应性矩阵（homography matrix）的正确性？","验证单应性矩阵质量的方法：\n1. **重投影误差检查**：将源图像点通过 H 矩阵变换后，与目标图像对应点计算误差\n2. **对称性测试**：计算 H 和 H_inv，双向变换检查一致性\n3. **几何合理性**：检查变换后的图像是否出现不合理的拉伸或畸变\n\n参考实现：https:\u002F\u002Fstackoverflow.com\u002Fquestions\u002F14954220\u002Fhow-to-check-if-obtained-homography-matrix-is-good\n\n注意：对于水面、天空等弱纹理场景，特征匹配困难可能导致单应性矩阵不准确，建议增加重叠区域或采用多相机标定方案。","https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fissues\u002F69",{"id":172,"question_zh":173,"answer_zh":174,"source_url":160},3820,"Docker 部署时如何修改源码（如适配卫星图像处理）？","如果需要修改源码，建议不使用 Docker，而是采用本地部署：\n1. 克隆代码库到本地\n2. 根据需求修改源代码（如调整图像分块大小处理大尺寸卫星图像）\n3. 使用 `gradio-offline` 替换 `gradio` 以支持离线环境\n4. 运行 `python app.py --config config\u002Fconfig.yaml`\n\n对于 GPU 内存不足处理大尺寸图像（如 2K×2K），需要在源码中实现图像分块（tiling）策略，而非直接修改 Docker 镜像。",[176,181,186,191,196,201,206],{"id":177,"version":178,"summary_zh":179,"released_at":180},103334,"v0.0.7","## What's Changed\r\n* Fix gradio 6 compatibility by @fAIseh00d in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F151\r\n* feat: fix proper gpu support for docker compose by @ndeybach in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F147\r\n\r\n## New Contributors\r\n* @fAIseh00d made their first contribution in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F151\r\n* @ndeybach made their first contribution in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F147\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fcompare\u002Fv0.0.6...v0.0.7","2026-02-27T06:07:25",{"id":182,"version":183,"summary_zh":184,"released_at":185},103335,"v0.0.6","**Full Changelog**: https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fcompare\u002Fv0.0.6...v0.0.6\r\n\r\n## What's Changed\r\n* 🛠️ refactor: Modularize configuration management and improve model loading by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F145\r\n\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fcompare\u002Fv0.0.5...v0.0.6","2025-08-24T11:46:44",{"id":187,"version":188,"summary_zh":189,"released_at":190},103336,"v0.0.5","## What's Changed\r\n* 🚀 feat: Enhance Docker deployment with CI\u002FCD and compose support by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F140\r\n* 🔧 chore: Update CI workflow and unpin dependency versions by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F143\r\n\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fcompare\u002Fv0.0.4...v0.0.5","2025-08-23T16:46:30",{"id":192,"version":193,"summary_zh":194,"released_at":195},103337,"v0.0.4","## What's Changed\r\n* update: docker by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F122\r\n* update: cache model by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F123\r\n* add: dad detector with roma matcher by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F124\r\n* fix: roma inference by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F125\r\n* fix: rotation examples by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F127\r\n* add: liftfeat by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F130\r\n* update: download liftfeat models by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F131\r\n* add: rdd sparse and dense matcher by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F132\r\n* feat: Add RIPE feature extractor integration by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F136\r\n* fix: ci by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F138\r\n* update: version by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F139\r\n\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fcompare\u002Fv0.0.3...v0.0.4","2025-07-19T14:40:36",{"id":197,"version":198,"summary_zh":199,"released_at":200},103338,"v0.0.3","## What's Changed\r\n* update: enable gluestick by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F116\r\n* update: lib by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F117\r\n* update: compute H by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F120\r\n* update: server by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F121\r\n\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fcompare\u002Fv0.0.2...v0.0.3","2025-02-22T15:36:26",{"id":202,"version":203,"summary_zh":204,"released_at":205},103339,"v0.0.2","## What's Changed\r\n* update: checkpoints path by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F111\r\n* add: minima by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F112\r\n* add: force stop button by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F113\r\n\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fcompare\u002Fv0.0.1...v0.0.2","2025-01-06T00:09:05",{"id":207,"version":208,"summary_zh":209,"released_at":210},103340,"v0.0.1","## What's Changed\r\n* docker! by @karolmajek in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F25\r\n* Dev by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F27\r\n* FIX: RoMa cpu inference by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F29\r\n* update: docker by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F30\r\n* update: interface by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F31\r\n* Update: superpoint and run ransac by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F32\r\n* add SFD2 (CVPR 2023) and IMP (CVPR 2023) by @feixue94 in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F40\r\n* Dev by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F41\r\n* update: quiet wget by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F42\r\n* update: device by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F43\r\n* update: rename common -> ui by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F47\r\n* Add: mast3r by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F48\r\n* Dev by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F50\r\n* fix: force_resize image by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F52\r\n* FIX: model loading bug by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F60\r\n* Fix: force_resize by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F62\r\n* Fix: sfd2 by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F63\r\n* add: Efficient LoFTR by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F64\r\n* add: xfeat+lightglue by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F68\r\n* Dev: update env by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F74\r\n* Dev： update env by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F75\r\n* Dev: update dockerfile by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F76\r\n* add: api by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F77\r\n* fix: incorrect usage of `error_colormap` by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F79\r\n* add: matches color info in UI by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F80\r\n* update: api by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F81\r\n* update: api by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F82\r\n* update: python version by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F83\r\n* update: railway api by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F84\r\n* add: railway.toml by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F85\r\n* update: module path by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F86\r\n* update: module path by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F89\r\n* update: api and gradio -> 5.x by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F90\r\n* update: format code by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F93\r\n* update: ray serve api by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F94\r\n* add: xoftr by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F96\r\n* Added ALIKED+LightGlue by @Dawars in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F99\r\n* Format code and update ci by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F105\r\n* Update: ci by @agipro in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F106\r\n* update: v1.3 by @agipro in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F107\r\n* Fix: rel path by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F109\r\n* update: pkgs by @Vincentqyw in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F110\r\n\r\n## New Contributors\r\n* @karolmajek made their first contribution in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F25\r\n* @feixue94 made their first contribution in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F40\r\n* @Dawars made their first contribution in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F99\r\n* @agipro made their first contribution in https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fpull\u002F106\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FVincentqyw\u002Fimage-matching-webui\u002Fcommits\u002Fv0.0.1","2025-01-01T16:38:37"]