hkchengrex/Tracking-Anything-with-DEVA
[ICCV 2023] Tracking Anything with Decoupled Video Segmentation
This project helps video editors, content creators, or analysts automatically track and segment any object across video frames. You provide a video and a text description (e.g., "guinea pigs," "capybara," "person with hat"), and the tool outputs a video with the specified objects precisely outlined and followed throughout the footage. It's ideal for anyone needing to isolate or analyze specific elements in video content without manual, frame-by-frame effort.
1,487 stars. No commits in the last 6 months.
Use this if you need to accurately highlight and follow specific objects or categories of objects in a video, even if they appear, disappear, or change position.
Not ideal if your primary goal is to segment objects from static images, or if you need to precisely track objects in very short clips where dedicated, end-to-end video segmentation models trained specifically on that type of data might perform better.
Stars
1,487
Forks
139
Language
Python
License
—
Category
Last pushed
Apr 26, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/hkchengrex/Tracking-Anything-with-DEVA"
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)