CIntellifusion/GeometryForcing

[ICLR26] Official implementation of Geometry Forcing: Marrying Video Diffusion and 3D Representation for Consistent World Modeling

41
/ 100
Emerging

This project helps researchers in computer vision generate videos that are consistent in both motion and 3D structure. It takes an input image and synthesizes a video from it, where objects and scenes maintain their shape and position realistically across frames. This is ideal for those working on realistic 3D scene reconstruction or video synthesis.

160 stars.

Use this if you need to generate high-quality, temporally consistent videos from a single image while ensuring accurate 3D geometry.

Not ideal if your primary goal is simple video generation without strong emphasis on 3D geometric consistency.

3D-reconstruction video-synthesis computer-vision-research generative-AI scene-understanding
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 15 / 25
Community 6 / 25

How are scores calculated?

Stars

160

Forks

4

Language

Python

License

Last pushed

Jan 26, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/CIntellifusion/GeometryForcing"

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