- PaddleOCR by PaddlePaddle, 93k stars. OCR toolkit that turns PDFs and images into structured data for LLMs
- opencv by opencv, 92k stars. C++ library for image processing, computer vision and deep learning
- MinerU by opendatalab, 84k stars. Converts PDFs and Office files into markdown or JSON for LLM pipelines
- AI-For-Beginners by microsoft, 71k stars. 12-week course with 24 lessons on AI fundamentals and practice
- supervision by roboflow, 52k stars. Reusable Python helpers for computer vision tasks like detection and annotation
- paperless-ngx by paperless-ngx, 48k stars. Self-hosted document archive that scans, indexes and searches your files
- 500-AI-Machine-learning-Deep-learning-Computer-vision-NLP-Projects-with-code by ashishpatel26, 38k stars. Collection of 500 ML, deep learning, computer vision and NLP project examples with code
- frigate by blakeblackshear, 37k stars. Network video recorder that runs local object detection on IP camera feeds
- OCRmyPDF by ocrmypdf, 35k stars. Adds a searchable text layer to scanned PDFs using Tesseract OCR
- detectron2 by facebookresearch, 35k stars. Library for object detection and segmentation built on PyTorch
- Unlimited-OCR by baidu, 31k stars. OCR project for parsing long documents into text in a single pass
- label-studio by HumanSignal, 29k stars. Annotation tool for labeling text, images, audio and video data
- screenpipe by screenpipe, 22k stars. Records screen and audio history on your computer to give agents context
- surya by datalab-to, 22k stars. OCR and layout analysis models with table recognition in 90+ languages
- labelme by wkentaro, 16k stars. Image annotation tool for polygons, boxes and AI-assisted labels
- ImageToolbox by T8RIN, 16k stars. Android app for image editing with crop, filters, OCR and many tools
- javacv by bytedeco, 8.4k stars. Java bindings for OpenCV, FFmpeg and other native media libraries
- koharu by koharu-rs, 6.7k stars. Machine learning tool that translates manga pages, written in Rust
- Windrecorder by yuka-friends, 4.0k stars. Records screen activity and makes it searchable by OCR text and image descriptions
- manga-ocr by kha-white, 2.8k stars. Japanese OCR model tuned for manga text
- clearcam by roryclear, 2.6k stars. Adds object detection, tracking, alerts and search to any security camera
- spirula-studio by harry7557558, 2.4k stars. Gaussian splat trainer that turns video into splats and meshes on Vulkan or CUDA
- glance by jonnyoo, 1.9k stars. Unlocks a Mac with face recognition using computer vision
- indra by datalab-to, 1.8k stars. 1M-parameter OCR model that runs on CPU
- typesafe-computer-use by awlevin, 1.3k stars. Computer use: OCR the screen, pick the next action with TypeSafe, click. macOS
- capcap by realskyrin, 1.1k stars. Menu bar screenshot tool with region capture, annotation and clipboard copy
- mamma by cuevhv, 898 stars. Official code for multi-person markerless 3D human motion capture from video
- Dalaran by Flaminis, 666 stars. ROS 2 visualizer and data store for robot time-series, reads .rrd files
- light-ocr by arcships, 661 stars. Offline OCR for Node.js and C++ with GPU or Core ML acceleration
- ParseBench by run-llama, 638 stars. Benchmark that scores document parsers on tasks aimed at AI agents
- Hands-On-Learning-in-Computer-Vision by Labellerr, 600 stars. Jupyter notebooks for learning computer vision techniques step by step
- FaceGate-Mac by dweep-desai, 597 stars. Locks Mac apps with face recognition, Touch ID or a password, all local
- watch-skill by oxbshw, 582 stars. Turns video, audio and screen activity into timestamped, searchable evidence for agents
- mac-ocr by privatenumber, 551 stars. Command line OCR for images and searchable PDFs using Apple Vision
- neverclick by LazoVelko, 495 stars. Desktop app for controlling the mouse by keyboard, using local computer vision
- Boxer3D by Barath19, 433 stars. iPhone AR app that detects 3D objects with LiDAR, YOLO and BoxerNet
- ocr-it by thiagotigaz, 429 stars. Chrome extension that OCRs a pinned screen region offline with Tesseract
- gyotaku by xevrion, 397 stars. Offline search over screenshots using OCR text, native Linux app built with gpui
- X-AnyLabeling-Server by CVHub520, 338 stars. Lightweight server that runs vision models for the X-AnyLabeling annotation tool
- textsnap by kouhxp, 198 stars. Extracts plain text from images, screenshots and webpages locally using OCR
New repos every two days, stars refreshed daily. Visits are counted without cookies, and the numbers are public. How it works Data RSS Newsletter