SJTU-DeepVisionLab/LaGa

Tackling View-Dependent Semantics in 3D Language Gaussian Splatting (ICML 2025)

23
/ 100
Experimental

This tool helps researchers and computer vision engineers precisely identify and segment 3D objects within complex digital scenes using simple text descriptions. It takes detailed 3D scene data (like those captured by LERF-OVS or ScanNet) and outputs accurate 3D segmentations, even when objects appear differently from various viewpoints. It's designed for professionals working with advanced 3D scene understanding and virtual environment analysis.

No commits in the last 6 months.

Use this if you need to perform highly accurate, language-driven 3D object segmentation and analysis on complex, real-world 3D scene data, overcoming challenges with how objects look from different angles.

Not ideal if you are looking for a basic 2D image segmentation tool or if your workflow doesn't involve detailed 3D scene reconstruction and analysis.

3D-scene-understanding computer-vision spatial-computing virtual-environment-analysis robotics-perception
No License Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 8 / 25
Maturity 7 / 25
Community 6 / 25

How are scores calculated?

Stars

62

Forks

3

Language

C++

License

Last pushed

Jun 03, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/SJTU-DeepVisionLab/LaGa"

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