Rishabh1925/scene-localization-system

Powerful CLIP-based computer vision system for natural language-driven object and scene localization in images. Features smart query expansion, adaptive detection, and interactive web UI.

26
/ 100
Experimental

This system helps professionals like marketers, researchers, or archivists quickly find specific objects or scenes within a collection of images using everyday language. You input an image and a text description (e.g., "red car"), and the system outputs the image with bounding boxes highlighting the detected items, along with cropped images of each detection and detailed metadata. It's designed for anyone who needs to visually analyze images without specialized technical training.

Use this if you need to precisely locate and identify objects or complex scenes within images using natural language queries, such as for content analysis, visual search, or automated tagging.

Not ideal if you need real-time object detection for live video streams or require extremely high precision for safety-critical applications, as analysis can take several minutes.

visual-content-analysis image-search digital-asset-management media-monitoring research-imaging
No Package No Dependents
Maintenance 6 / 25
Adoption 5 / 25
Maturity 15 / 25
Community 0 / 25

How are scores calculated?

Stars

10

Forks

Language

HTML

License

MIT

Last pushed

Oct 26, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Rishabh1925/scene-localization-system"

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