WesleyHsieh0806/TAO-Amodal
Official Code for Tracking Any Object Amodally
This project helps computer vision researchers and robotics engineers analyze how objects move in video sequences, even when they are partially hidden. It takes in video frames and corresponding annotation files describing objects and their occlusions, then outputs predictions about object identities and their bounding box locations across frames. This allows users to understand the full extent of objects despite visual obstructions.
120 stars. No commits in the last 6 months.
Use this if you need to accurately track the complete form of objects in videos, even when parts of them are obscured by other objects or obstacles.
Not ideal if you only need to track the visible parts of objects or if your application doesn't involve heavy occlusion in video analysis.
Stars
120
Forks
12
Language
Python
License
—
Category
Last pushed
Jul 11, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/WesleyHsieh0806/TAO-Amodal"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
roboflow/rf-detr
[ICLR 2026] RF-DETR is a real-time object detection and segmentation model architecture...
stereolabs/zed-sdk
⚡️The spatial perception framework for rapidly building smart robots and spaces
mikel-brostrom/boxmot
BoxMOT: Pluggable SOTA multi-object tracking modules with support for axis-aligned and oriented...
RizwanMunawar/yolov7-object-tracking
YOLOv7 Object Tracking Using PyTorch, OpenCV and Sort Tracking
google-deepmind/tapnet
Tracking Any Point (TAP)