healthonrails/annolid

An annotation and instance segmentation-based multi-object tracking and behavior analysis package.

68
/ 100
Established

Integrates foundation models (Cutie, Segment Anything, Grounding DINO) and optional EfficientTAM for video segmentation to enable markerless tracking from minimal annotations, even under occlusion. Combines interactive GUI annotation with LLM-powered behavior classification, keypoint tracking, and zone-based spatial analysis—supporting both pre-recorded video and real-time streams with optional GPU acceleration. Extensible via Model Context Protocol (MCP) for custom tools and agent-driven analytics workflows.

Available on PyPI.

Maintenance 13 / 25
Adoption 14 / 25
Maturity 25 / 25
Community 16 / 25

How are scores calculated?

Stars

56

Forks

10

Language

Python

License

—

Category

assistive-vision-ai

Last pushed

Mar 17, 2026

Monthly downloads

435

Commits (30d)

0

Dependencies

44

GitHub PyPI
Assistive Vision Ai · 1 tools

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/healthonrails/annolid"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.

Explore Computer Vision Tools

All categories Trending Computer Vision directory Insights