自动判断目标成熟程度的甜椒采摘机器人 http://sweeper-robot.eu/
Docker, PyTorch & Flask 机器学习模型训练与web界面部署实例 https://github.com/imadelh/ML-web-app
Keras)简单图片搜索 https://github.com/matsui528/sis
霸榜 GitHub:200 行代码让人瞬间消失 https://mp.weixin.qq.com/s/-QQvvPkBDR5C0JPjzJ73vg
一款优秀入门级 AI 项目以及教程,内容涵盖:人脸、视频、文字的检测和识别。他不仅包含最基本的人脸检测、识别(图片、视频)、轮廓标识、头像合成(给人戴帽子),还有表情识别(生气、厌恶、恐惧等)、视频对象提取、图片修复(可用于水印去除)、图片自动上色等等。推荐这个开源项目不是因为它的内容强大,而是它的教程写的实在太好了,真 · 入门级。手把手教你如何使用这个项目,做出上述功能来。在每篇功能文章的教程里,不仅仅写了每个功能的技术实现方案,还有具体重点关键代码的注释和解释以及具体实现,让你非常轻松的能够看懂、学习和使用。 https://github.com/vipstone/faceai
【机器人操作视觉迁移学习】《Visual Transfer Learning for Robotic Manipulation | Google AI Blog》 https://ai.googleblog.com/2020/03/visual-transfer-learning-for-robotic.html
https://github.com/diovisgood/agender
让蒙娜丽莎“返老还童”:StyleGAN2人脸修改器 https://github.com/tg-bomze/StyleGAN2-Face-Modificator
基于First Order模型的人脸动态模型,让Doom小子变清晰/活起来(Remini/StyleGan2) https://github.com/tg-bomze/Face-Image-Motion-Model https://papers.nips.cc/paper/8935-first-order-motion-model-for-image-animation
基于Opencv的口罩佩戴识别系统 https://blog.csdn.net/cj151525/article/details/104984897
用MediaPipe/TensorFlow.js实现人脸/手势跟踪(facemesh&handpose) https://blog.tensorflow.org/2020/03/face-and-hand-tracking-in-browser-with-mediapipe-and-tensorflowjs.html?linkId=83996111
【基于Flask + Gunicorn + Nginx的机器学习模型部署模板(Docker)】 https://github.com/ivanpanshin/flask_gunicorn_nginx_docker
基于TensorFlow.js在浏览器里实现复杂背景上的实时人像去除
https://github.com/jasonmayes/Real-Time-Person-Removal
https://github.com/wuba/dl_inference
https://github.com/Dhanush45/Realtime-emotion-detectionusing-python
【世界就是一块大绿幕:来看看最先进的视频背景抠图】《Background Matting: The World is Your Green Screen》 https://github.com/senguptaumd/Background-Matting https://gist.github.com/andreyryabtsev/243aa3eefa6e06891dda7b1583d1d08f
用OpenCV实现人脸自动马赛克 https://www.pyimagesearch.com/2020/04/06/blur-and-anonymize-faces-with-opencv-and-python/
【runx:深度学习实验管理工具】’runx - An experiment management tool' https://github.com/NVIDIA/runx
【RGB摄像头“社交距离”检测】’Social-Distancing - Code for estimating social distances from RGB cameras.' https://github.com/IIT-PAVIS/Social-Distancing
【WebXR + TensorFlow.js 实现的拍照创建全息Demo】 https://nosy-b.github.io/holography/
在线会议玩变身!Zoom/Skype虚拟化身 https://github.com/alievk/avatarify
https://github.com/prabodhhere/tsne-grid
https://github.com/Open-Debin/Emotion-FAN
https://github.com/sbdcv/sbd_mask
FaceMaskDetection - 开源人脸口罩检测模型和数据 Detect faces and determine whether people are wearing mask.' https://github.com/AIZOOTech/FaceMaskDetection
https://github.com/incluit/OpenVino-For-SmartCity
https://github.com/MegviiDetection/video_analyst
无需代码,训练自己的学习机器 https://teachablemachine.withgoogle.com/ https://github.com/googlecreativelab/teachablemachine-community
【手把手:将机器学习模型部署为REST API 】《Deploy Your Machine Learning Model as a REST API 》 https://pan.baidu.com/s/1zGztpScX2JxJKpeNNhOBjA
【收集各类深度学习项目,快速搭建demo.用最小代码量实现功能展现】 https://github.com/zhangqijun/deeplearningdemo
Comet ML — 你必須知道的 ML 實驗管理神器 https://pan.baidu.com/s/1lK8ZuEInFDl7D0JHCkmG6w
https://github.com/avinassh/pytorch-flask-api
面向可视化和传感器融合的深度学习实用程序库 https://github.com/jinfagang/alfred
机器学习常用库基础(附PPT下载链接) https://bigquant.com/community/t/topic/163807?utm_source=weibo&utm_campaign=191114_l1
用Angular/Tensorflow.js构建和部署图像分类应用 https://medium.com/m/global-identity?redirectUrl=https%3A%2F%2Fblog.angularindepth.com%2Fcreate-your-own-image-classifier-with-angular-and-tensorflow-js-5b1bc2391424
Contextual Attention for Hand Detection in the Wild https://github.com/SupreethN/Hand-CNN
DeblurGAN-v2: Deblurring (Orders-of-Magnitude) Faster and Better https://github.com/TAMU-VITA/DeblurGANv2
Mobilenet-SSD License Plate Detection & Recognition - 基于mobilenet-ssd车牌检测,全新模型仅3.4MB. https://github.com/zeusees/Mobilenet-SSD-License-Plate-Detection
(Pytorch)面向行为识别和目标检测的视频平台 https://github.com/MichiganCOG/ViP
yolo_nano https://blog.csdn.net/qq_33270279/article/details/102522025 https://arxiv.org/abs/1910.01271
https://github.com/liux0614/yolo_nano
【男宝/女宝👶分类器】’boyorgirl - A web application + neural net to distinguish baby boys and girls' https://github.com/artoby/boyorgirl
用于重复/近似图片发现的Python包 https://github.com/idealo/imagededup
用 FastAPI + Uvicorn 快速部署机器学习模型 https://medium.com/m/global-identity?redirectUrl=https%3A%2F%2Ftowardsdatascience.com%2Fhow-to-deploy-a-machine-learning-model-dc51200fe8cf
https://github.com/taki0112/FUNIT-Tensorflow https://github.com/NVlabs/FUNIT https://github.com/znxlwm/FUNIT-pytorch
深度模型推理服务工具 https://github.com/awslabs/multi-model-server
Service Streamer:加速深度学习web服务(将服务请求排队组成一个完整的batch,再送进GPU运算) https://github.com/ShannonAI/service-streamer
支持 1-100 用户的小型 JupyterHub 版本 https://github.com/jupyterhub/the-littlest-jupyterhub
将机器学习模型发布为REST API https://towardsdatascience.com/deploying-a-machine-learning-model-as-a-rest-api-4a03b865c166
fullstack.ai:端到端的机器学习项目全程实例,展示了从开发到部署的真实机器学习驱动应用开发的关键环节 https://github.com/xadrianzetx/fullstack.ai
深度学习web应用:从数据采集到应用部署 https://medium.com/m/global-identity?redirectUrl=https%3A%2F%2Ftowardsdatascience.com%2Fdeep-learning-based-web-application-from-data-collection-to-deployment-9e89f29b67df
###【YOLOv3的简化训练API】 https://github.com/BMW-InnovationLab/BMW-YOLOv3-Training-Automation
lightKG,lightsmile个人的知识图谱技术框架 https://github.com/smilelight/lightKG
服务侧深度学习部署案例 https://github.com/layerism/TensorRT-Inference-Server-Tutorial
上传一份PPT,它会自动分析里边的演讲者注释,用人工智能生成语音,然后再合成视频。非常适合用来做产品介绍、流程介绍、以及技术课程。 Slide.ftqq.com
(Colab)深度学习神经网络手把手实例入门 https://colab.research.google.com/drive/1VdwQq8JJsonfT4SV0pfXKZ1vsoNvvxcH#scrollTo=V8Vq_6Q3zivl
如何学习SVM(支持向量机)以及改进实现SVM算法程序? https://mp.weixin.qq.com/s/HewogcuiDhemMGagvwGXEA
卷积神经网络详解 https://github.com/ss-is-master-chief/MNIST-Digit.Recognizer-CNNs
用 Pytorch 理解卷积网络 https://builtin.com/data-science/convolutional-neural-networks-explained
一份简短的深度学习笔记 https://github.com/jianzhu/dl-notes
Advanced Topics in Neural Networks https://medium.com/m/global-identity?redirectUrl=https%3A%2F%2Ftowardsdatascience.com%2Fadvanced-topics-in-neural-networks-f27fbcc638ae
深度学习的秘密 https://www.bilibili.com/video/av67000954/
超全!19 种损失函数,你能认识几个? https://mp.weixin.qq.com/s/ZObRXiCqKH1HEqByW8F5AQ
深度网络权重初始化好文(Xavier和Kaiming He方法解析) https://link.medium.com/IRdTdYN6zV
比个手势,AI 自动识别 Emoji! https://mp.weixin.qq.com/s/46P-M2wHx4277EhQAhBGEA
为工程师设计的机器学习课程和教学项目 https://github.com/lukas/ml-class
Keras深度学习实践:交通标志分类 https://www.pyimagesearch.com/2019/11/04/traffic-sign-classification-with-keras-and-deep-learning/
https://github.com/davidpfahler/react-ml-app https://davidpfahler.com/fastai-in-the-browser
(Python)白手起家机器学习 https://github.com/eriklindernoren/ML-From-Scratch
非常好的优化器选择指南 https://ruder.io/optimizing-gradient-descent/ https://arxiv.org/abs/1609.04747
推荐 GitHub 上几个比较实用的机器学习项目 https://mp.weixin.qq.com/s/M3fDPJKvFt6oUetTKnCN3w
如何管理机器学习实验 https://medium.com/@hadyelsahar/how-do-you-manage-your-machine-learning-experiments-ab87508348ac
100个精品项目带你掌握人工智能 https://github.com/Fabsqrt/BitTigerLab/blob/master/ArtificialIntelligent/README.md
Python机器学习实战教程 https://www.bilibili.com/video/av65888723/
深度学习领域涌现的新挑战 https://www.bilibili.com/video/av65728457/
《机器学习》理解思路与扩展知识点 https://github.com/Vay-keen/Machine-learning-learning-notes
数据科学里最强大的思想 “信号:不仅存在于数据集中的模式。 噪声:仅存在于数据集中的模式。” https://medium.com/@kozyrkov/the-most-powerful-idea-in-data-science-78b9cd451e72
那些酷炫的深度学习网络图怎么画出来的? https://mp.weixin.qq.com/s/4QZSMoGD-w4LbqDga8vfgQ
用 GitHub Classroom 进行统计学教学 https://www.tandfonline.com/doi/full/10.1080/10691898.2019.1617089 https://classroom.github.com/
动手学深度学习一天训练营 https://github.com/mli/d2l-1day-zh#part-1-deep-learning-basic
AI 落地思考集 https://github.com/lonelygo/Shift-AI-models-to-real-world-products
《深度学习入门:基于Python的理论与实现》 https://bigquant.com/community/t/topic/160266
PCA、LDA、MDS、LLE、TSNE等降维算法的python实现 https://github.com/heucoder/dimensionality_reduction_alo_codes
李宏毅机器学习完整笔记 https://github.com/datawhalechina/leeml-notes https://www.bilibili.com/video/av59538266
数据科学的笔记以及资料搜集 https://github.com/fengdu78/Data-Science-Notes
神经网络:前馈、反向传播的解释和优化 https://mlfromscratch.com/neural-networks-explained/
如何给8岁女儿解释深度学习 https://www.ovh.com/blog/deep-learning-explained-to-my-8-year-old-daughter/
深度网络排错实践指南 http://josh-tobin.com/assets/pdf/troubleshooting-deep-neural-networks-01-19.pdf
(CS231N)Python Numpy教程 http://cs231n.github.io/python-numpy-tutorial/
与数据融为一体:机器学习之前需要了解的熟悉数据、转换原始数据、探索性分析、建立人工基准 https://blog.floydhub.com/becoming-one-with-the-data/
神经网络参数优化交互可视化解析 https://www.deeplearning.ai/ai-notes/optimization/
一年个人机器学习研究体会分享 https://pan.baidu.com/s/1WaCcZhCmlRN_NVTPKCvnDg
开源免费机器学习书籍《南瓜书 PumpkinBook》。该书是对机器学习领域的经典入门教材周志华老师的《机器学习》书籍的一些细微补充,比如难理解的公式加以解析,以及补充具体的推导细节。 https://github.com/datawhalechina/pumpkin-book https://datawhalechina.github.io/pumpkin-book/#/
完整的目标检测项目。结构简洁明了,中文注释。适宜新手入门、目标检测任务参考,甚至直接基于本项目实现目标检测任务。 https://github.com/yatengLG/SSD-Pytorch
各种深度学习架构、模型和技巧的集合。TensorFlow 和 PyTorch 的各种深度学习架构、模型和技巧的 Jupyter 集合,非常适合学习 https://github.com/rasbt/deeplearning-models
一套机器学习课程。课程全面细致,同时带有 demo 以及进阶的 Kaggle 比赛的样例,非常适合初学者逐步的深入学习 https://github.com/Yorko/mlcourse.ai
机器学习领域经典书籍《统计学习方法》的课件和代码。这个项目提供了课件、代码资源,叙述从具体问题或实例入手,由浅入深,阐明思路,给出必要的数学推导,便于读者掌握统计学习方法的实质,学会运用 https://github.com/fengdu78/lihang-code
使用Python进行机器学习的假设检验 https://mp.weixin.qq.com/s/iC6I66_bBEHlnwZMEZsN_Q
图像分类任务中的损失 https://medium.com/@lightsanweb/losses-in-image-classification-task-7401a8348927
超参优化概览 https://pan.baidu.com/s/14s8zUNNkc6hladTY2BwH_Q
8个深度学习/计算机视觉常见错误及如何避免 https://pan.baidu.com/s/12nfLHjUW4VFAvYTlL5IDoQ
用 Tensorflow.js、ONNX.js 在浏览器里运行深度学习模型 https://medium.com/datadriveninvestor/running-your-deep-learning-models-in-a-browser-using-tensorflow-js-and-onnx-js-a35256d3933
【机器学习应用开发框架,可大大简化前端开发】’Streamlit - Streamlit — The fastest way to build custom ML tools' Streamlit:把Python脚本变成漂亮的机器学习工具 https://github.com/streamlit/streamlit https://weibo.com/1402400261/I9CCR0u8v?mid=4423196612519378&ouid=1402400261&type=comment#_rnd1581127724731
'Streamlit Demo: The Udacity Self-driving Car Image Browser - Streamlit app demonstrating an image browser for the Udacity self-driving-car dataset with realtime object detection using YOLO.' https://github.com/streamlit/demo-self-driving
fastai v2 基本使用实例教程 https://www.kaggle.com/jhoward/fastai-v2-pipeline-tutorial
用 DeepStream SDK 加速实时AI视频/图像分析 https://www.bilibili.com/video/av69983350
【机器学习数据集版本控制入门】 https://towardsdatascience.com/start-version-controlling-your-machine-learning-datasets-2b872e109856
Deepnote:面向团队的notebook,可实时协作 https://www.deepnote.com/
https://github.com/nbgallery/nbgallery
如何用简单易懂的例子解释隐马尔可夫模型? https://www.zhihu.com/question/20962240/answer/33438846
上手实战:Jupyter Notebook安装与机器学习起步 https://pan.baidu.com/s/1ebTJkBKtlHk55OHy3ysO1g https://medium.com/installation-of-jupyter-notebook-and-getting/installation-of-jupyter-notebook-getting-started-with-machine-learning-7443a5e09309
图像迷你实验室,可进行图像实验、处理、分析。使用 PyQt5 结合 opencv-python 实现代码简洁易读,通过该工具可以快速演示图像算法效果,便于 cv 入门学习 https://github.com/itisyang/ImageMiniLab
计算机视觉最佳实践、代码示例和相关文档 https://github.com/microsoft/computervision-recipes
图像处理入门100题实例解析中文版 https://github.com/gzr2017/ImageProcessing100Wen
一款优秀入门级 AI 项目以及教程,内容涵盖:人脸、视频、文字的检测和识别。他不仅包含最基本的人脸检测、识别(图片、视频)、轮廓标识、头像合成(给人戴帽子),还有表情识别(生气、厌恶、恐惧等)、视频对象提取、图片修复(可用于水印去除)、图片自动上色等等。推荐这个开源项目不是因为它的内容强大,而是它的教程写的实在太好了,真 · 入门级。手把手教你如何使用这个项目,做出上述功能来。在每篇功能文章的教程里,不仅仅写了每个功能的技术实现方案,还有具体重点关键代码的注释和解释以及具体实现,让你非常轻松的能够看懂、学习和使用。对于想要入门或者了解机器学习的初学者不能再友好了 https://github.com/vipstone/faceai
上手快速、简单易用的人脸识别库。该库使用 dlib 顶尖的深度学习人脸识别技术构建,在户外脸部检测数据库基准(Labeled Faces in the Wild benchmark)上的准确率高达 99.38%。同时提供了一个简单的面部识别命令行工具,允许您对来自命令行的图像文件夹进行面部识别。完整的开发文档和应用案例,并且兼容树莓派系统(对配置要求低),对于初学者来说可以通过这个项目感受人脸识别或机器学习带来的乐趣。 https://github.com/ageitgey/face_recognition
《 解析卷积神经网络 ——深度学习实践手册》 https://bigquant.com/community/t/topic/161186
机器学习超级笔记:基本概念、算法、流行模型等 https://createmomo.github.io/2018/01/23/Super-Machine-Learning-Revision-Notes/#tableofcontents
神经网络不起作用的37个理由 https://mp.weixin.qq.com/s/WX_4wEIX8JWdsJQNdkLkJw
多图带你读懂各种常见卷积类型 https://www.yanxishe.com/TextTranslation/1488
像堆乐高一样:从零开始解释神经网络的数学过程 https://mp.weixin.qq.com/s/tz-Gzh1TifGDjBLE6HBKdQ
(Jupyter notebooks)机器学习/数据科学算法交互式探索与可视化 https://github.com/PyMLVizard/PyMLViz
计算机视觉/深度学习/OpenCV 入门指南 https://www.pyimagesearch.com/start-here/
CV岗常见面试题 https://github.com/donnyyou/cv-interview
CVPR 2019 计算机视觉最新趋势 https://towardsdatascience.com/latest-computer-vision-trends-from-cvpr-2019-c07806dd570b
深度学习应用思路分享 https://github.com/sladesha/deep_learning
如何向投资人和客户展示你的产品 https://medium.com/@oyper/how-to-present-your-product-to-investors-and-customers-7b4974b03fbf
《Deep Learning vs. Traditional Computer Vision》 https://arxiv.org/abs/1910.13796
深度学习框架的技术挑战和发展趋势 https://v.qq.com/x/page/e302689p3l0.html?&__t=1&ptag=1.sina&_out=1
Yoshua Bengio:深度学习——从1型系统(直觉/无意识/感知)到2型系统(推理/规划/因果) https://www.bilibili.com/video/av79448071?p=1
【NeurIPS 2019 关键趋势总结】 https://huyenchip.com/2019/12/18/key-trends-neurips-2019.html https://venturebeat.com/2019/12/13/google-ai-chief-jeff-dean-interview-machine-learning-trends-in-2020/ 1、机器学习专用芯片将会更有优势; 2、社区不应该痴迷于 SOTA; 3、多模态、多任务和多模型学习会是一个有趣的方向。
“Melanie Mitchell访谈:概念、类比、常识与AI未来” https://www.bilibili.com/video/av65677422?p=9
【2千万数据集唾手可及数据科学新时代的思考】 https://pan.baidu.com/s/1n4oVSl44gwDoOWcpDpZwMQ
【来自AI初创公司创始人的2020年机器学习趋势预测】 https://pan.baidu.com/s/1ou6xUBgxsMSMO494rYBmpw https://medium.com/@ivanfokeev/machine-learning-in-2020-from-the-ai-startup-founder-4ecb6745b6ce
【2020年AI将走向何方:Soumith Chintala, Celeste Kidd, Jeff Dean, Anima Anandkumar, Dario Gil 观点】 https://venturebeat.com/2020/01/02/top-minds-in-machine-learning-predict-where-ai-is-going-in-2020/
用OpenCV、Keras/TensorFlow实现口罩检测 https://www.pyimagesearch.com/2020/05/04/covid-19-face-mask-detector-with-opencv-keras-tensorflow-and-deep-learning/
用于投篮分析的网页&API https://github.com/chonyy/AI-basketball-analysis
https://github.com/broutonlab/face-id-with-medical-masks
【OpenCV/Tensorflow/Keras口罩检测】 https://github.com/chandrikadeb7/Face-Mask-Detection
【用PyTorch Lightning实现口罩检测器】《How I built a Face Mask Detector for COVID-19 using PyTorch Lightning》 https://towardsdatascience.com/how-i-built-a-face-mask-detector-for-covid-19-using-pytorch-lightning-67eb3752fd61
https://github.com/littlemountainman/modeld
【3D车道检测合成数据集】 https://github.com/yuliangguo/3D_Lane_Synthetic_Dataset https://arxiv.org/abs/2003.10656
OpenCV实现的实时“阿凡达”特效(First Order Motion Model) https://github.com/anandpawara/Real_Time_Image_Animation
https://github.com/AutoViML/AutoViz https://towardsdatascience.com/autoviz-a-new-tool-for-automated-visualization-ec9c1744a6ad
【Auto Painter的web demo代码】 https://github.com/irfanICMLL/Auto_painter_demo
https://github.com/Alpha-Video/AlphaVideo
【迁移学习动物类密集姿态估计】 https://github.com/facebookresearch/detectron2 https://research.fb.com/publications/transferring-dense-pose-to-proximal-animal-classes/
https://qiita.com/yukihiko_a/items/43d09db5628334789fab https://github.com/digital-standard/ThreeDPoseTracker
【TengineKit:免费手机端人脸212特征点实时标定】' TengineKit - Free RealTime Face Landmarks 212 Points For Mobile' https://github.com/OAID/TengineKit
'静默活体检测 (Silent-Face-Anti-Spoofing)' https://github.com/minivision-ai/Silent-Face-Anti-Spoofing
微软的 AI Lab 项目 AI Lab helps our large fast-growing community of developers get started on AI. You can experience, learn and code the latest and greatest innovations from Microsoft AI here. AI Lab currently houses eight projects that showcase the latest in custom vision, attnGAN, Visual Studio tools for AI, Cognitive Search, Machine Reading Comprehension and more. Each lab gives you access to the experimentation playground, source code on GitHub, a crisp developer-friendly video, and insights into the underlying developer/ organizational challenge and solution.
- 微软数学:基于 OCR 技术从学生笔记中快速提取数学方程式,并提供解题步骤与相识问题
- 明星脸匹配:基于深度神经网络(DNN)模型和必应图库自动完成明星脸的相似度匹配
- Sketch2Code:可将白板上的手写图形快速转化为可运行的 HTML 代码,帮助设计、开发人员提升与客户间的协作效率 https://github.com/microsoft/ailab
https://github.com/Adityaojas/sudoku-solver
【在Google街景里“漫步”】“Virtual walks in Google Street View” https://github.com/Moving-AI/virtual-walk
【计算机视觉预训练模型集合(Tensorflow/Keras/PyTorch/Caffe/MXNet)】 https://github.com/balavenkatesh3322/CV-pretrained-model
Lane Detection in Low-light Conditions Using an Efficient Data Enhancement : Light Conditions Style Transfer (IV 2020)
https://github.com/Chenzhaowei13/Light-Condition-Style-Transfer
key points estimation and point instance segmentation approach for lane detection https://github.com/koyeongmin/PINet
Ultra Fast Structure-aware Deep Lane Detection (ECCV 2020) https://github.com/cfzd/Ultra-Fast-Lane-Detection
https://github.com/huangzhikun1995/IPM-Net
Talking With Hands 16.2M: A Large-Scale Dataset of Synchronized Body-Finger Motion and Audio for Conversational Motion Analysis and Synthesis . https://github.com/facebookresearch/TalkingWithHands32M
https://github.com/google-research/ssl_detection
Official implementation for "CP-VTON+: Clothing Shape and Texture Preserving Image-Based Virtual Try-On" from CVPRW 2020.
https://github.com/minar09/cp-vton-plus
GPU-Jupyter:使用方便的Jupyter+GPU的Docker镜像 https://github.com/iot-salzburg/gpu-jupyter
Visual Chirality (CVPR2020 Best Paper Nominee) https://github.com/linzhiqiu/digital_chirality
NVIDIA针对自家GPU优化的云推断解决方案,通过HTTP/REST或GRPC端点提供推理服务,也可以作为带API的共享库使用,支持各种深度学习框架 https://github.com/triton-inference-server/server
https://github.com/Media-Smart/flexinfer volksdep is an open-source toolbox for deploying and accelerating PyTorch, Onnx and Tensorflow models with TensorRT. https://github.com/Media-Smart/volksdep
PyTorch open-source toolbox for image-based localization (place recognition), including the implementation of our ECCV-2020 spotlight paper
https://github.com/yxgeee/OpenIBL
https://www.pyimagesearch.com/2020/08/10/opencv-sudoku-solver-and-ocr/
https://github.com/drivendataorg/open-cities-ai-challenge
https://github.com/lilipads/gradient_descent_viz (已在macOS下安装成功,效果不错,可以演示 Work/Deeplearning目录)
【视频会议实时自动手语检测】《Developing Real-Time, Automatic Sign Language Detection for Video Conferencing | Google AI Blog》
https://ai.googleblog.com/2020/10/developing-real-time-automatic-sign.html https://research.google/pubs/pub49425/
https://github.com/TwentyBN/sense
https://towardsdatascience.com/a-must-have-tool-for-every-data-scientist-5e7c76f1916f
【Python分层感知库】'(PAZ) Perception for Autonomous Systems - Hierarchical perception library in Python' https://github.com/oarriaga/paz
NARUTO-HandSignDetection:深度结印——忍术结印手势识别 https://github.com/Kazuhito00/NARUTO-HandSignDetection
基于 Python 和 FastAI 做了个 NSFW(不宜在工作场所打开)图像分类器,可对黄色暴力图像进行分类筛选 https://github.com/ClarityCoders/RedditBot-FastAI
最先进的神经网络火焰检测(基于Yolov5 PyTorch) https://github.com/robmarkcole/fire-detection-from-images
DeepLab v3+OpenCV的视频会议虚拟背景实战 https://towardsdatascience.com/virtual-background-for-video-conferencing-using-machine-learning-dfba17d90aa9
adversarial.js:浏览器里演示神经网络对抗攻击 https://github.com/kennysong/adversarial.js
Bringing-Old-Photos-Back-to-Life,可借助 AI 对老照片快速进行修复。 项目基于 PyTorch 实现,支持划痕检测、面部修饰、全局还原等操作。 https://github.com/microsoft/Bringing-Old-Photos-Back-to-Life
https://github.com/LeonLok/Multi-Camera-Live-Object-Tracking
Image Classification Training GUI - CPU/GPU for linux https://github.com/BMW-InnovationLab/BMW-Classification-Training-GUI
Hand detection models trained on 100DOH (100 Days of Hands) dataset. https://github.com/ddshan/hand_detector.d2
Train robotic agents to learn pick and place with deep learning for vision-based manipulation in PyBullet. Transporter Nets, CoRL 2020. https://github.com/google-research/ravens
https://medium.com/roonyx/ml-model-deployment-option-with-concurrency-flask-uwsgi-4aeafea0d07e
https://github.com/ecsplendid/rembg-greenscreen
用CLIP/StyleGAN实现的基于文本的图片编辑 https://github.com/orpatashnik/StyleCLIP
QDServing - 机器学习模型分布式服务部署:grpc,flask;docke https://github.com/MachineLP/QDServing
https://github.com/exadel-inc/CompreFace
https://paperswithcode.com/newsletter/5
https://petewarden.com/2021/02/28/how-screen-scraping-and-tinyml-can-turn-any-dial-into-an-api/ https://github.com/jomjol/AI-on-the-edge-device
ultimateALPR-SDK:CPU/GPU/VPU/FPGA上最快的多平台深度学习车牌识别 https://github.com/DoubangoTelecom/ultimateALPR-SDK
https://github.com/pkhungurn/talking-head-anime-2-demo
github.com/SthPhoenix/InsightFace-REST
'DeepVision - CV算法快速验证框架项目' github.com/peng-zhihui/DeepVision
【Human Library:基于TensorFlow的JS 3D人脸检测、人体姿态、手和手指跟踪、虹膜跟踪及年龄和性别预测】 github.com/vladmandic/human
MIVisionX:一套综合的计算机视觉和机器智能库、实用程序和应用工具包 https://github.com/GPUOpen-ProfessionalCompute-Libraries/MIVisionX
https://github.com/artperrin/image2csv
github.com/luxonis/depthai-experiments
FaceBlit:基于样例的实时人脸画风迁移 ondrejtexler.github.io/faceblit/
【DetecTree:卫星图片树木检测】 github.com/martibosch/detectree
github.com/poloclub/magic-crop
OpenVINO Inference API:基于OpenVINO的免代码目标检测API github.com/BMW-InnovationLab/BMW-IntelOpenVINO-Inference-API
SOLD2: Self-supervised Occlusion-aware Line Description and Detection github.com/cvg/SOLD2
Slideo: 用OpenCV实现的视频与PPT自动同步 github.com/hediet/slideo
github.com/Cambricon/CNStream
ailia-models:高速推理引擎ailia SDK的预训练模型集 github.com/axinc-ai/ailia-models
Opyrator:将机器学习代码转化为具有网络API、交互式GUI等的微服务 github.com/ml-tooling/opyrator
Machine Learning Collection:微软机器学习/深度学习开源贡献集,包括库、工具、示例代码等 github.com/microsoft/machine-learning-collection
BCNet: Learning Body and Cloth Shape from A Single Image github.com/jby1993/BCNet
reflective-clothes-detect-dataset、helemet detection yolov5、工作服(反光衣)检测数据集、安全帽检测、施工人员穿戴检测' github.com/gengyanlei/reflective-clothes-detect-yolov5
YOLOv4 - ANPR/ALPR:基于YOLOv4, Tesseract, PaddleOCR的车牌识别项目 github.com/Guiflayrom/ALPR-YOLOv4
一个比较丰富的 C++ 库:Lite.AI,里面收录了 60 多个实用的 AI 模型,其中包含对象检测、人脸检测、风格迁移、人脸识别、图像分割、图像分类等类别。 github.com/DefTruth/lite.ai Yolo-FastestV2:适用于移动设备的超轻量高性能Yolo目标检测库 github.com/dog-qiuqiu/Yolo-FastestV2
shufflev2-yolov5:更轻量、更快速、更易部署的yolov5改进版 github.com/ppogg/shufflev2-yolov5
用MediaPipe BlazePose GHUM 和 TensorFlow.js实现3D姿态检测 https://blog.tensorflow.org/2021/08/3d-pose-detection-with-mediapipe-blazepose-ghum-tfjs.html
End-to-End Monocular Vanishing Point Detection Exploiting Lane Annotations https://arxiv.org/abs/2108.13699
ONNX运行时推理示例 github.com/microsoft/onnxruntime-inference-examples
ONNX运行时网页Demo github.com/microsoft/onnxruntime-web-demo
在Nvidia Triton服务器上部署优化版Hugging Face Transformer github.com/ELS-RD/triton_transformers
iris:基于PyTorch的开源照片平台 github.com/prabhuomkar/iris
BigPretrain - A Simple Framwork for CV Pre-training Model (SOCO, VirTex, BEiT) github.com/Sense-GVT/BigPretrain
MLServer:机器学习模型开源推理服务器 github.com/SeldonIO/MLServer
ONNX LSTR Lane Detection:ONNX实现的LSTR车道检测 github.com/ibaiGorordo/ONNX-LSTR-Lane-Detection
人脸画风 https://github.com/Sxela/ArcaneGAN https://huggingface.co/spaces/akhaliq/ArcaneGAN
PaddleBoBo:基于飞桨开发的虚拟主播 github.com/JiehangXie/PaddleBoBo
卡通脸转换模型 https://colab.research.google.com/github/williamyang1991/VToonify/blob/master/notebooks/inference_playground.ipynb https://huggingface.co/PKUWilliamYang/VToonify https://huggingface.co/spaces/PKUWilliamYang/VToonify https://arxiv.org/abs/2209.11224
【《深度学习产品应用》随书代码】’Production-Ready Applied Deep Learning - Production-Ready Applied Deep Learning' GitHub: github.com/PacktPublishing/Production-Ready-Applied-Deep-Learning
'MLOps for Vision Models (TensorFlow) from Transformers with TensorFlow Extended (TFX) - shows how to build Machine Learning pipeline for a vision model (TensorFlow) from Transformers using the TensorFlow Ecosystem' by Chansung Park GitHub: github.com/deep-diver/mlops-hf-tf-vision-models
【CircularNet垃圾识别:用机器学习从废品中分拣可回收材料】《CircularNet: Reducing waste with Machine Learning | The TensorFlow Blog》
【labml.ai深度学习论文注释与复现集】’labml.ai Deep Learning Paper Implementations - 50! Implementations/tutorials of deep learning papers with side-by-side notes; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, ...), gans(cyclegan, stylegan2, ...), reinforcement learning (ppo, dqn), capsnet, distillation, ...' GitHub: github.com/labmlai/annotated_deep_learning_paper_implementations
’Creative AI’ by Sony GitHub: github.com/sony/creativeai
【深度学习不良内容审核(内容审核,敏感内容检测,场景类型分类,裸体检测,暴力检测,物质检测)相关资源列表】’deep-learning-content-moderation - Deep learning based content moderation from text, audio, video & image input modalities.' by fatih GitHub: github.com/fcakyon/content-moderation-deep-learning
【给时装设计师的AI工具VisualHound:1、选择想要设计的产品 2、添加自定义特色(颜色,材质...) 3、自动生成设计样例】“VisualHound - Prototype your fashion design ideas with AI” https://visualhound.com/create-products
【大规模车辆分类(与价格预测)】《Large-Scale Vehicle Classification – Pickled ML》 https://blog.aqnichol.com/2022/12/31/large-scale-vehicle-classification/ GitHub: github.com/unixpickle/car-data
【AI 服务&工具集市】“All Things AI - The Complete Resource Of Artificial Intelligence Tools & Services” https://allthingsai.com/
【DeepPrivacy2:真实图片匿名化工具箱,自动改变图片中人脸等身份和隐私相关信息】’DeepPrivacy2 - A Toolbox for Realistic Image Anonymization - DeepPrivacy2 - A Toolbox for Realistic Image Anonymization' Håkon Hukkelås GitHub: github.com/hukkelas/deep_privacy2
【RestorePhotos.io:肖像照片自动修复】’RestorePhotos.io - Restoring old and blurry face photos with AI.' Hassan El Mghari GitHub: github.com/Nutlope/restorePhotos
'Chinese_license_plate_detection_recognition - yolov5 车牌检测 车牌识别 中文车牌识别 检测 支持13种中文车牌 支持双层车牌' we0091234 GitHub: github.com/we0091234/Chinese_license_plate_detection_recognition
'yolov7 车牌检测 车牌识别 中文车牌识别 检测 支持双层车牌 支持13种中文车牌' we0091234 GitHub: github.com/we0091234/yolov7_plate
【YOLNP - YOLO车牌检测项目】’YOLNP - You Only Look Number Plates - Yolnp is a project based in YOLO to detect plates' Guilherme D. Alves GitHub: github.com/Guiflayrom/yolnp
'YoloV8 Raspberry Pi 4 - YoloV8 for a bare Raspberry Pi 4' Q-engineering GitHub: github.com/Qengineering/YoloV8-ncnn-Raspberry-Pi-4
【Web AI:在浏览器里运行最新的深度网络】’Web AI - Run modern deep learning models in the browser.' Alexander Visheratin GitHub: github.com/visheratin/web-ai
'BallonTranslator - 深度学习辅助漫画翻译工具, 支持一键机翻和简单的图像/文本编辑 | Yet another computer-aided comic/manga translation tool powered by deeplearning' dmMaze GitHub: github.com/dmMaze/BallonsTranslator
一款开源的流程可视化 Web 应用:FUXA,可为机器的实时数据,构建多种不同设计风格的可视化流程图。 GitHub:github.com/frangoteam/FUXA
如果你想做一个在线服务,用户涂鸦后AI出图,那么这里有一个开源项目Scribble Diffusion你可以参考。 网页链接 🔗 github.com/replicate/scribble-diffusion 像我昨天推荐的另一个用照片让AI生成房间布局的AI应用roomGPT 微博正文 类似,它也是基于Replicate (replicate.com)提供的API来实现的,同样也需要ControlNet来固定图形布局。 网站程序是基于Nextjs,很方便就能部署到Vercel上。
Temporalkit + Ebsynth + 后期处理 以后真的别轻易相信网上的视频了😄 原贴:🔗twitter.com/yokohara_h/status/1657191627452981248
【Willow - 新一代开源、本地自托管的语音助手,匹敌Amazon Echo/Google Home,高效、准确、廉价且功能丰富。在本地设备上完全自掌控,无需担忧隐私问题】’Willow - A Practical, Open Source, Privacy-Focused Platform for Voice Assistants and other Applications - Open source, local, and self-hosted Amazon Echo/Google Home competitive Voice Assistant alternative' Tovera GitHub: github.com/toverainc/willow
'本地AI - Mac和Windows一键安装Stable Diffusion WebUI,LamaCleaner,SadTalker,ChatGLM2-6B,等AI工具,使用国内镜像,无需魔法。' dxcweb GitHub: github.com/dxcweb/local-ai
【EcoAssist:一款开源应用,旨在简化生态学家处理相机陷阱图像的工作。EcoAssist是一个人工智能平台,可以对自动物种检测的自定义模型进行标注、训练和部署,为生态学家提供了一种节省图像审查时间并专注于保护工作的方法】'EcoAssist - an open-source application designed to streamline the work of ecologists dealing with camera trap images' Peter van Lunteren GitHub: github.com/PetervanLunteren/EcoAssist
【pyvideotrans:一个视频翻译工具,可将一种语言的视频翻译为另一种语言和配音的视频。 语音识别基于 openai-whisper 离线模型、文字翻译使用google翻译接口,文字合成语音使用 Microsoft Edge tts,背景音乐去除使用 Spleeter,无需购买任何商业接口,也无需付费】'pyvideotrans - Translate the video from one language to another and add dubbing. ' okaymyworld GitHub: github.com/jianchang512/pyvideotrans
'MyHeyGen:平民版视频翻译工具,音频翻译,翻译校正,视频唇纹合成全流程解决方案’ by AIFSH GitHub: github.com/AIFSH/MyHeyGen
'基于AI的视频硬字幕去除、水印去除,无损分辨率生成去字幕、去水印后的视频文件。无需申请第三方API,本地实现。AI-based tool for removing hard-coded subtitles and watermarks from videos.' YaoFANGUK GitHub: github.com/YaoFANGUK/video-subtitle-remover
【FunASR-APP:基于阿里巴巴FunASR开源语音模型的多功能语音应用库,当前包含两个主要功能模块:ClipVideo和TransAudio:ClipVideo可以根据语音识别结果帮助提取视频或音频中的指定片段,生成剪辑后的视频或音频;TransAudio支持语音转写】’FunASR-APP - Applications based on speech related models from FunASR (Modelscope).' Alibaba Damo Academy GitHub: github.com/alibaba-damo-academy/FunASR-APP
【一个python脚本,可以高效地将YouTube播放列表或单个视频转变为高质量的文字脚本,整合了Whisper模型、SpaCy和CUDA等技术进行视频内容的转录】'Bulk Transcribe Youtube Videos from Playlists - Easily take an entire YouTube playlist and turn it into high quality transcripts using Whisper.' Jeff Emanuel GitHub: github.com/Dicklesworthstone/bulk_transcribe_youtube_videos_from_playlist