XinyuYanTJU/LawDIS

[ICCV'2025] LawDIS: Language-Window-based Controllable Dichotomous Image Segmentation

36
/ 100
Emerging

This project helps graphic designers, image editors, and researchers accurately extract specific objects or regions from images. You provide an image and a text description, and it generates a precise segmentation mask. It then allows for fine-tuning those masks by interactively adjusting regions, making it ideal for detailed image manipulation and analysis.

Use this if you need to precisely isolate objects in images using natural language, and then meticulously refine those selections for high-accuracy results.

Not ideal if you only need rough object detection or if you don't require the ability to refine segmentation masks with fine-grained control.

image-editing computer-vision-research graphic-design visual-content-creation photo-manipulation
No Package No Dependents
Maintenance 10 / 25
Adoption 8 / 25
Maturity 15 / 25
Community 3 / 25

How are scores calculated?

Stars

53

Forks

1

Language

Python

License

MIT

Last pushed

Mar 08, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/XinyuYanTJU/LawDIS"

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