Open Source Computer Vision Projects
Computer vision enables computers to interpret and analyze visual information, such as images and video. It supports tasks including object detection, image classification, facial analysis, text recognition, segmentation, and motion tracking. These techniques help automate inspection, improve accessibility, organize visual data, and give machines a way to respond to their surroundings. Methods range from traditional image processing to systems trained with machine learning and deep learning.
Open source tools in this area include libraries for image processing, model training and inference, data annotation, visual search, and real-time analysis. When choosing a tool, consider its license, maintenance activity, documentation, hardware and software requirements, and compatibility with your data and existing systems. These resources are useful to researchers, developers, educators, and organizations building or evaluating visual applications.
65 repositories · updated October 15, 2026
SpatialLM: Turn 3D Point Clouds into Structured Indoor Scenes
SpatialLM uses language models to interpret indoor point clouds and produce structured layouts, including walls, doors, windows, and object boxes. It suits 3D perception researchers and teams building scene-understanding prototypes, but requires a CUDA-oriented setup.
QuietHealth 27/100
claude-video: Give AI Agents Timestamped Video Evidence
claude-video adds a watch skill that turns video URLs or local files into timestamped visual frames and transcripts for AI agents. It suits people who want an agent to analyze video, with options for local processing or Gemini and transcription services.
NewHealth 36/100
AWESOME-OCR-LLM: Track OCR Research in the LLM Era
AWESOME-OCR-LLM is a curated reading list for OCR research shaped by large language and vision-language models. It helps researchers and practitioners follow work across document parsing, understanding, visual text generation, and evaluation.
ActiveHealth 43/100
scikit-video: Process Video with Python
scikit-video is a Python library for video I/O and video-oriented analysis, including quality metrics and temporal processing. It suits researchers and developers who want Python-based tools to experiment with video algorithms alongside SciPy and scikit-image.
InactiveHealth 20/100
vidgear: Capture, Process, Write, and Stream Video in Python
VidGear is a Python framework that unifies video capture, processing, encoding, and streaming through specialized APIs built on tools such as OpenCV and FFmpeg. It suits developers building real-time media applications who want threaded or asynchronous workflows without wiring each backend themselves.
QuietHealth 26/100
Awesome-pytorch-list: Find PyTorch Libraries, Tutorials, and Papers
A categorized directory of PyTorch libraries, learning materials, and paper implementations. Use it to discover resources across NLP, computer vision, probabilistic modeling, and other deep-learning topics.
MaintainedHealth 37/100
PixelRAG: Search Documents by Their Visual Content
PixelRAG indexes screenshots of web pages and documents for visual retrieval, preserving charts, tables, and layout that text extraction can lose. It offers a Python pipeline, command-line renderer, and hosted search API.
NewHealth 55/100
bg-remove: Remove Image Backgrounds in Your Browser
bg-remove is a browser-based image background remover that runs machine-learning models locally with Transformers.js. It suits people who want to edit images without uploading them to a server, with optional WebGPU acceleration where supported.
InactiveHealth 20/100
Qwen3-VL: Understand Images, Video, and Text with Multimodal Models
Qwen3-VL is a family of multimodal language models for interpreting images, video, and text. The repository provides inference examples, deployment guidance, and cookbooks for tasks such as OCR, spatial reasoning, and visual agents.
QuietHealth 39/100
GLM-OCR: Recognize Text and Structure in Documents
GLM-OCR is a multimodal OCR model and SDK for extracting text and layout from complex documents. Use it through a hosted API or deploy the pipeline with supported inference servers for local control.
QuietHealth 35/100
CellProfiler: Measure Biological Phenotypes from Images
CellProfiler helps biologists quantify phenotypes across large collections of images without requiring computer vision or programming expertise. It is suited to repeatable image-analysis workflows, with stable releases for users and source builds for contributors.
ActiveHealth 53/100
UI-TARS-desktop: Control Computers with a Multimodal AI Agent
UI-TARS-desktop is a desktop GUI agent that uses vision-language models to operate computers through natural-language instructions. It offers local and remote computer and browser operators for users building or exploring AI-driven automation.
ActiveHealth 45/100