【Linux下的摄像头背景伪造】’Linux-Fake-Background-Webcam - Faking your webcam background under GNU/Linux.' https://github.com/fangfufu/Linux-Fake-Background-Webcam
FaceXLib：pytorch人脸相关开发库 github.com/xinntao/facexlib
ncnn-android-scrfd：安卓平台ncnn库应用人脸检测演示项目 github.com/nihui/ncnn-android-scrfd
'轻量级cpu版实时口罩检测' https://github.com/sbdcv/sbd_mask
LaneDet：PyTorch车道检测工具箱 github.com/Turoad/lanedet
真人自拍生成二次元头像 https://mp.weixin.qq.com/s/Qr4PlxjiQPPoxRDskeDIUw https://github.com/znxlwm/UGATIT-pytorch
BeautyGAN: Instance-level Facial Makeup Transfer with Deep Generative Adversarial Network https://github.com/Honlan/BeautyGAN
最新的卡通风效果 https://github.com/SystemErrorWang/White-box-Cartoonization
新版动画滤镜，看首页上的样品照片还挺不错的 TachibanaYoshino / AnimeGANv2 https://github.com/TachibanaYoshino/AnimeGANv2
人像卡通化 (Photo to Cartoon)探索 Few-shot Knowledge Transfer for Fine-grained Cartoon Face Generation https://github.com/minivision-ai/photo2cartoon
SwapNet：用PyTorch复现的虚拟试衣，提供Docker支持 https://github.com/andrewjong/SwapNet
【唇色/发色编辑器】 https://github.com/zllrunning/face-makeup.PyTorch
Pose Animator。该项目可识别出人类的表情变化与肢体动作，并将其映射到 2D 矢量图型上，让它跟着你的节拍动起来 https://github.com/yemount/pose-animator/
【用脸玩的贪吃蛇】 https://github.com/paruby/snake-face
【(OpenCV/Numpy)用手势控制赛车游戏】’ChoudharyChanchal - using openCV and Numpy libraries of Python to control a racing game with the steering wheel' https://github.com/ChoudharyChanchal/game_control
【做手势控制游戏(OpenCV)】’Gesture Gaming using OpenCV' https://github.com/gnana1997/Gesture-Gaming-using-OpenCV
跟神经网络玩“石头、剪子、布” https://github.com/victorqribeiro/jokenpo
【(Python)做动作，打游戏】 https://github.com/pateldigant/gesture-gaming-python
https://github.com/techwithtim/Snake-Game + usrevent https://github.com/yas-sim/human-pose-estimation-2d-demo
sense：面向动作识别的神经网络推理引擎 https:// github.com/TwentyBN/sense
MoveNet：基于TensorFlow.js的下一代姿态检测库 https://blog.tensorflow.org/2021/05/next-generation-pose-detection-with-movenet-and-tensorflowjs.html
ncnn_nanodet_hand：安卓平台上的ncnn手部检测/手势检测 github.com/FeiGeChuanShu/ncnn_nanodet_hand
Shuwa Gesture Toolkit：手语/手势识别工具包 github.com/google/shuwa
Nano发送rtsp流
ABR Control：Python机械臂控制库 github.com/abr/abr_control
github.com/silverbulletmdc/showdata
《Learning To Count Everything》(CVPR 2021) github.com/cvlab-stonybrook/LearningToCountEverything
PyTorch Remote Sensing (torchrs)：遥感任务数据和PyTorch模型实现 github.com/isaaccorley/torchrs
github.com/nadermx/backgroundremover
让聊天更专注：用深度显著先验减少视觉干扰 deep-saliency-prior.github.io/
Portal：用最快的方式加载和可视化深度学习模型(用TypeScript, React, Flask构建) github.com/datature/portal
Sign Language-to-Speech with DeepStack's Custom API：基于DeepStack定制API开发的手语转语音应用 github.com/SteveKola/Sign-Language-to-Speech-with-DeepStack-Custom-API
Lite.AI：各种最新AI模型的用户友好C++库，可实现目标检测、人脸检测、画风迁移、人脸对齐、人脸识别、分割、着色、人脸属性分析、图像分类、抠图等 github.com/DefTruth/lite.ai
着装人体相关文献列表，包括人体数字化、虚拟试穿等相关方向 github.com/weihaox/awesome-clothed-human
Cherry on Py：支持人脸检测、视频目标检测、视频摘要的云计算平台 github.com/PicardParis/cherry-on-py
AnyoneNet: Synchronized Speech and Talking Head Generation for arbitrary person
https://arxiv.org/abs/2108.04325
FLAME-in-NeRF : Neural control of Radiance Fields for Free View Face Animation
https://arxiv.org/abs/2108.04913
CompreFace，人脸识别与检测系统，不懂机器学习技术亦可使用。 支持快速轻松集成到任意系统上，提供一系列可用于人脸识别检测、年龄性别识别、特征点检测等 REST API，并支持 Docker 部署。 github.com/exadel-inc/CompreFace
ONNX高效车道检测实例
github.com/ibaiGorordo/onnx-Ultra-Fast-Lane-Detection-Inference
fastgradio - a python library to quickly build and share gradio interfaces of your trained fastai models #Visual
github.com/aliabd/fastgradio
github.com/TangleSpace/hydralit
Keypoint Communities github.com/DuncanZauss/Keypoint_Communities
XManager: 机器学习实验管理框架 github.com/deepmind/xmanager
A Universal PyTorch Source-Build Docker Template：PyTorch/CUDA/cuDNN项目模板 github.com/veritas9872/PyTorch-Universal-Docker-Template
Pose2Carton：基于3D视觉用姿态控制卡通人物的教学用API
github.com/yuzhenbo/pose2carton
基于Fastapi + MLflow + streamlit的机器学习数字识别部署实例 github.com/zademn/mnist-mlops-learning
PSGAN-NCNN - PSGAN running with ncnn 妆容迁移/仿妆Imitation Makeup/Makeup Transfer github.com/EdVince/PSGAN-NCNN
invoice。可通过 API 接口，快速识别发票代码、发票号码、开票日期、校验码、税后金额等内容。 GitHub：github.com/guanshuicheng/invoice
一款开箱即用的 C++ AI 模型工具箱：Lite.AI.ToolKit，涵盖目标检测、人脸检测、人脸识别、语义分割、抠图等领域。 里面包括了 70+ 流行开源模型，如最新的 RVM、YOLOX、YoloV5、DeepLabV3、ArcFace 等模型，对用户友好，简单易用。 GitHub：github.com/DefTruth/lite.ai.toolkit
《PP-PicoDet: A Better Real-Time Object Detector on Mobile Devices》 github.com/PaddlePaddle/PaddleDetection
github.com/mle-infrastructure/mle-monitor
YOLOP-opencv-dnn - 用OpenCV部署全景驾驶感知网络YOLOP，可同时处理交通目标检测、可驾驶区域分割、车道线检测，三项视觉感知任务，包含C++和Python两种版本的程序实现 github.com/hpc203/YOLOP-opencv-dnn
Yolov5/Deepsort的C++实现 github.com/RichardoMrMu/yolov5-deepsort-tensorrt
CVZone：基于OpenCV和Mediapipe库的计算机视觉包 github.com/cvzone/cvzone
CoreML-Models - Converted CoreML Models Zoo. github.com/john-rocky/CoreML-Models
awesome-demos：用Gradio开发的机器学习Demo集锦 github.com/gradio-app/awesome-demos
XManager: 机器学习实验管理框架 github.com/deepmind/xmanager
txtai：用Python, Hugging Face Transformers, Sentence Transformers and FastAPI方便构建AI语义搜索应用 github.com/neuml/txtai
Audio2Face：音频驱动人脸生成 github.com/FACEGOOD/Audio2Face
Neptune：机器学习实验管理利器 https://neptune.ai/product/experiment-tracking
Yolov5 for Fire Detection：基于Yolov5的图片/视频火焰检测 github.com/spacewalk01/Yolov5-Fire-Detection
github.com/hpc203/virtual_try_on_use_deep_learning
Clothes Segmentation using U2NET：U2NET服饰分割 github.com/levindabhi/cloth-segmentation
VTuber Python Unity Tutorial：用Python和Unity实现的VTuber github.com/mmmmmm44/VTuber-Python-Unity
github.com/mpc001/Visual_Speech_Recognition_for_Multiple_Languages
github.com/deepdoctection/deepdoctection
使用深度学习算法实现虚拟试衣，结合了人体姿态估计、人体分割、几何匹配和 GAN，四种模型，只依赖 opencv 库就能运行。 GitHub：github.com/hpc203/virtual_try_on_use_deep_learning
'Android BlazePose demo by ncnn' github.com/FeiGeChuanShu/ncnn_Android_BlazePose
TheVrMLEyeToolbox：眼动跟踪工具包 github.com/SummerSigh/TheVrMLEyeToolbox
github.com/inspurer/WorkAttendanceSystem
github.com/Baiyuetribe/paper2gui
Lite.AI.ToolKit：开箱即用的C++ AI工具箱，包括70+流行的开源模型，如最新的RVM, YOLOX, YOLOP, YOLOR, YoloV5, DeepLabV3, ArcFace等模型，支持ONNXRuntime/NCNN/MNN/TNN github.com/DefTruth/lite.ai.toolkit
SysMocap：用于3D虚拟角色动画制作的实时动作捕捉系统 github.com/xianfei/SysMocap
【Image-Processing-Node-Editor：用节点编辑器执行图像/视频处理的应用】'Image-Processing-Node-Editor - An application that performs image processing with the node editor' by KazuhitoTakahashi GitHub: github.com/Kazuhito00/Image-Processing-Node-Editor
'ONNX-HybridNets-Multitask-Road-Detection - Python scripts for performing road segemtnation and car detection using the HybridNets multitask model in ONNX.' by Ibai Gorordo GitHub: github.com/ibaiGorordo/ONNX-HybridNets-Multitask-Road-Detection
Paper2GUI，一款面向普通人的 AI 桌面 APP 工具箱，开箱即用，非常方便。 已支持 15+ AI 模型，内容涵盖语音合成、视频补帧、视频超分、目标检测、图片风格化、图片 OCR 识别等领域，兼容 Windows、Mac、Linux 等主流操作系统。 GitHub：github.com/Baiyuetribe/paper2gui
github.com/DennisLiu1993/Fastest_Image_Pattern_Matching
【Mage：开源数据管理平台，可用来清洗数据，为训练AI/ML模型做准备】'Mage - Mage is an open-source data management platform that helps you clean data and prepare it for training AI/ML models.' GitHub: github.com/mage-ai/mage-ai
【计算机视觉最佳实践、代码示例和相关文档】'Computer Vision - Best Practices, code samples, and documentation for Computer Vision.' by Microsoft GitHub: github.com/microsoft/computervision-recipes
GitHub 上一套开源的环视物体检测方案：BEVFormer，实现了在鸟瞰图视角下的特征优化，可应用于汽车自动驾驶场景。 GitHub：github.com/zhiqi-li/BEVFormer
【envd：面向数据科学/机器学习工程团队的开发环境，内置 Jupiter/VSCode】’envd - Development Environment for data science and AI/ML' by TensorChord GitHub: github.com/tensorchord/envd
【freemocap：3D人体骨架检测】'freemocap - Free like Freedom' by jonmatthis GitHub: https:// github.com/jonmatthis/freemocap
【Movenet.Pytorch：用PyTorch重写的MoveNet人体关键点检测】’Movenet.Pytorch - A Pytorch implementation of MoveNet from Google. Include training code and pre-trained model.' by Mr.Fire GitHub: github.com/fire717/movenet.pytorch
【min(DALL·E)：DALL·E Mini的最小化实现】’min(DALL·E) - a minimal implementation of DALL·E Mini’ by kuprel GitHub: github.com/kuprel/min-dalle
【FastDeploy：简单易用的推理部署工具箱。覆盖业界主流优质预训练模型并提供开箱即用的开发体验，包括图像分类、目标检测、图像分割、人脸检测、人体关键点识别、文字识别等多任务，满足开发者多场景，多硬件、多平台的快速部署需求】'FastDeploy - Easy-to-use Deep Learning Model Deployment Toolkit' by PaddlePaddle GitHub: github.com/PaddlePaddle/FastDeploy
【chaiNNer：基于流程图/节点的图像处理 GUI，旨在使链式图像处理任务(特别是由神经网络完成的任务)变得简单、直观且可定制】'chaiNNer - A flowchart/node-based image processing GUI aimed at making chaining image processing tasks (especially those done by neural networks) easy, intuitive, and customizable.' by Joey Ballentine GitHub: github.com/joeyballentine/chaiNNer
【CheekyKeys：用表请控制键盘[喵喵]】’CheekyKeys - Use Python, OpenCV, and MediaPipe to control a keyboard with facial gestures' by everythingishacked GitHub: github.com/everythingishacked/CheekyKeys
'ONNX HRNET 2D Human Pose Estimation (WIP) - Python scripts for performing 2D human pose estimation using the HRNET family models in ONNX.' by Ibai Gorordo GitHub: github.com/ibaiGorordo/ONNX-HRNET-Human-Pose-Estimation
GitHub 上一个开源的大模型深度学习系统：ColossalAI，提供了一系列开发工具，让分布式 AI 模型训练流程变得更加简单。 另外，该项目近期已无缝支持 Hugging Face，开发者仅需添加几行代码，便可实现大模型的低成本训练和微调。 GitHub：github.com/hpcaitech/ColossalAI 目前，Colossal-AI 技术解决方案已成功在自动驾驶、云计算、零售、医药、芯片等行业知名厂商落地应用，并广受开发者好评。 项目提供了完整的中文教程
【Poisson Image Editing：基于雅可比方法的快速泊松图像编辑算法，可利用多核CPU或GPU处理高分辨率图像输入】’Poisson Image Editing - A Parallel Implementation - A fast poisson image editing implementation that can utilize multi-core CPU or GPU to handle a high-resolution image input.' by Jiayi Weng GitHub: github.com/Trinkle23897/Fast-Poisson-Image-Editing
【DeepFit：健身人体姿态跟踪】’DeepFit - tracks your body movements throughout a workout' by Naman Arora GitHub: github.com/namanarora42/DeepFit
【GymLytics：健身姿态视觉分析】’GymLytics - Visual Analytics of different exercises for humans' by Akshay Bahadur GitHub: github.com/akshaybahadur21/GymLytics
【Robustar：鲁棒视觉分类交互式工具箱】'Robustar - Interactive Toolbox for Robust Vision Classification' by Haohan Wang GitHub: github.com/HaohanWang/Robustar
【PyMLpipe：用于简化机器学习模型监控和部署的Python库】’PyMLpipe - a Python library for ease Machine Learning Model monitoring and Deployment’ by neelindresh GitHub: github.com/neelindresh/pymlpipe
【EasyVtuber：Vtuber换脸直播工具】’EasyVtuber - tha3, but run 40fps on 3080 with virtural webcam support' by yuzhongliang22 GitHub: github.com/yuyuyzl/EasyVtuber
'Yolov7 + StrongSORT with OSNet - Real-time multi-camera multi-object tracker using YOLOv7 and StrongSORT with OSNet' by Mike GitHub: github.com/mikel-brostrom/Yolov7_StrongSORT_OSNet
'yolov7-object-tracking - YOLOv7 Object Tracking Using PyTorch, OpenCV and Sort Tracking' by Muhammad Rizwan Munawar GitHub: github.com/RizwanMunawar/yolov7-object-tracking
'yolov7-mask-ncnn - c++ version of yolov7-mask with ncnn' by FeiGeChuanShu GitHub: github.com/FeiGeChuanShu/yolov7-mask-ncnn
'Yolov7 Flask - A Beautiful Flask Web API for Yolov7 (and custom) models' by Michael.Wang GitHub: github.com/Michael-OvO/Yolov7-Flask
姿势参考神器 Pose Search，调整骨架姿势就能搜出一堆相应的参考图片，精细到每个关节，简直是画图辅助神器 ▶ 试玩地址：x6ud.github.io/pose-search ▶ 伟人开发者：@x6udpngx 这位大佬的动物头部参考神器，也同样神奇又好用 ▶ 动物头骨版：x6ud.github.io
'nonebot-plugin-drawer - 基于文心大模型的AI机器人画画插件。' by CrazyBoyM GitHub: github.com/CrazyBoyM/nonebot-plugin-drawer
【你画它补：基于Stable Diffusion的图画自动补全Demo，随意涂鸦或上传图片，给出完善方向的提示，程序会自动替你“补画”】“Diffuse The Rest - a Hugging Face Space by huggingface”
【PyDeepFakeDet：Deepfake换脸检测库】’PyDeepFakeDet - integrated and scalable tool for Deepfake detection.' by Junke, Wang GitHub: github.com/wangjk666/PyDeepFakeDet
【OpenHandMocap：视觉手部动作捕捉库】’OpenHandMocap' by JinTian GitHub: github.com/jinfagang/OpenHandMocap
[CV]《Adaptively-Realistic Image Generation from Stroke and Sketch with Diffusion Model》S Cheng, Y Chen, W Chiu, H Lee, H Tseng [National Chiao Tung University & Snap Inc & Meta] (2022) https://arxiv.org/abs/2208.12675
[CV]《Unsupervised Scene Sketch to Photo Synthesis》J Wang, S Jeon, S X. Yu, X Zhang, H Arora, Y Lou [UC Berkeley & Amazon] (2022) https://arxiv.org/abs/2209.02834
【基于Docker带Web界面方便部署的Stable Diffusion】’Stable Diffusion WebUI Docker - Easy Docker setup for Stable Diffusion with user-friendly UI' by AbdBarho GitHub: github.com/AbdBarho/stable-diffusion-webui-docker
【stablediffusion-infinity：基于Stable Diffusion的图片扩展补全】’stablediffusion-infinity - Outpainting with Stable Diffusion on an infinite canvas' by Lnyan GitHub: github.com/lkwq007/stablediffusion-infinity
【基于Stable Diffusion的提示图像补全】’stable-diffusion-prompt-inpainting - This project helps you do prompt-based inpainting without having to paint the mask - using Stable Diffusion and Clipseg' by amrrs GitHub: github.com/amrrs/stable-diffusion-prompt-inpainting
【LAVIS - 一站式语言-视觉智能库】'LAVIS - A One-stop Library for Language-Vision Intelligence' by Salesforce GitHub: github.com/salesforce/LAVIS
https://github.com/Baiyuetribe/paper2gui
自制桌面级小机器人。它具备 USB 通信显示画面功能以及 6 个自由度，支持手势识别和人体关键点检测。这里有配套的全部开发资料和 SDK，让你也可以制作出一个这样有趣的机器人 https://github.com/peng-zhihui/ElectronBot
RxImg is an image processing tool based on reactive data streams, allowing image processing pipeline to be built on a low-code graphical user interfaceConfig files https://github.com/rximg/rximg
'COCO Viewer - Simple COCO Viewer in Tkinter' by Taras Savchyn GitHub: github.com/trsvchn/coco-viewer
【EVA Video Analytics System：探索性视频分析系统】'EVA Video Analytics System - Exploratory Video Analytics System' by Georgia Tech Database Group GitHub: github.com/georgia-tech-db/eva
非专业人员可以“闭眼入”的三款现代实战型机器学习开发库：
- 表格: XGBoost/LightGBM github.com/dmlc/xgboost github.com/microsoft/LightGBM
- 文本: Hugging Face Transformers github.com/huggingface/transformers
- 视觉: timm github.com/rwightman/pytorch-image-models
国内开发者在 GitHub 开源的一款 Android 应用：「PoseMon 让爷康康」，可借助 AI 技术，实时监测不良坐姿，并及时给出语音提示。 应用不需要联网使用，所有 AI 特性均在手机本地运行，不需要将视频画面传输至外部服务器，仅需要摄像头权限用于获取姿态画面。 GitHub：github.com/linyiLYi/pose-monitor
【Hua：Stable Diffusion图像编辑器】’Hua - an AI image editor with Stable Diffusion (and more).' by PENG Bo GitHub: github.com/BlinkDL/Hua
使用Parallel-UNet架构，提出一种基于扩散的方法，实现了在保留服装细节的同时处理复杂的身体姿势变化，达到了最先进的效果。 https://arxiv.org/abs/2306.08276 [CV]《TryOnDiffusion: A Tale of Two UNets》L Zhu, D Yang, T Zhu, F Reda, W Chan, C Saharia, M Norouzi, I Kemelmacher-Shlizerman [Google Research & University of Washington] (2023)
【Live Face Recognition App：实时人脸识别应用，在实时视频流中识别人脸，可在浏览器上运行】'Live Face Recognition App - Simple Live Face Recognition Streamlit App' Martin Knoche GitHub: github.com/Martlgap/livefaceidapp