Kaihua-Chen/diffusion-vas
[CVPR 2025] Official code for Using Diffusion Priors for Video Amodal Segmentation
This project helps computer vision researchers and developers analyze videos by separating foreground objects from their backgrounds, even when parts of the objects are hidden or 'occluded'. It takes video footage and segmentations of visible object parts, then outputs a complete mask for each object, revealing its full shape as if transparent. Researchers in object tracking and autonomous systems would find this particularly useful for understanding complex scenes.
113 stars.
Use this if you need to accurately identify and delineate the entire shape of objects in videos, even when they are partially obscured by other elements in the scene.
Not ideal if you are looking for a simple, out-of-the-box solution for general video editing or if you don't have experience with computational vision frameworks.
Stars
113
Forks
5
Language
Python
License
MIT
Category
Last pushed
Nov 13, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/Kaihua-Chen/diffusion-vas"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
hao-ai-lab/FastVideo
A unified inference and post-training framework for accelerated video generation.
ModelTC/LightX2V
Light Image Video Generation Inference Framework
thu-ml/TurboDiffusion
TurboDiffusion: 100–200× Acceleration for Video Diffusion Models
PKU-YuanGroup/Helios
Helios: Real Real-Time Long Video Generation Model
PKU-YuanGroup/MagicTime
[TPAMI 2025🔥] MagicTime: Time-lapse Video Generation Models as Metamorphic Simulators