finlay-hudson/TABE
Track Anything Behind Everything: Zero-Shot Amodal Video Object Segmentation
This project helps video analysis professionals, like those in sports analytics or surveillance, to automatically identify and track objects in videos, even when they are partially hidden. You provide a video and an initial visible mask of the object you want to track, and it outputs a complete, 'amodal' mask for that object across all frames, showing its full extent even when obscured. This is ideal for researchers or anyone needing precise object tracking in complex video scenarios.
Use this if you need to accurately track the full shape of objects in videos, even when they are temporarily covered or partially out of view.
Not ideal if you only need to track the visible parts of objects or if you don't have access to high-end GPU hardware.
Stars
9
Forks
—
Language
Python
License
MIT
Category
Last pushed
Nov 06, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/finlay-hudson/TABE"
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