Track Anything: Segment Anything Meets Videos

Jinyu Yang, Mingqi Gao, Zhe Li, Shang Gao, Fangjing Wang +1 more
4/24/2023
cs.CV

Official code: gaomingqi/Track-Anything (7,012 stars, MIT, last commit 2025-12-13).

Abstract

Recently, the Segment Anything Model (SAM) gains lots of attention rapidly due to its impressive segmentation performance on images. Regarding its strong ability on image segmentation and high interactivity with different prompts, we found that it performs poorly on consistent segmentation in videos. Therefore, in this report, we propose Track Anything Model (TAM), which achieves high-performance interactive tracking and segmentation in videos. To be detailed, given a video sequence, only with very little human participation, i.e., several clicks, people can track anything they are interested in, and get satisfactory results in one-pass inference. Without additional training, such an interactive design performs impressively on video object tracking and segmentation. All resources are available on {https://github.com/gaomingqi/Track-Anything}. We hope this work can facilitate related research.

View on arXivView PDF

Code Implementations(1)

gaomingqi/Track-AnythingOfficial100%
MIT

Track-Anything is a flexible and interactive tool for video object tracking and segmentation, based on Segment Anything, XMem, and E2FGVI.

7,012510HTML, PythonApr 12, 202310 months agoMIT
inpaint-anythinginteractive-trackingsegment-anythingtrack-anythingvideo-object-segmentation+1 more

Similar Papers

Segment and Track Anything

87% match1 repo3.1k2023

Training-Free Robust Interactive Video Object Segmentation

87% match2024

Tracking with Human-Intent Reasoning

86% match1 repo772023

Tracking Anything in High Quality

86% match1 repo7522023

SAM2MOT: A Novel Paradigm of Multi-Object Tracking by Segmentation

86% match1 repo1822025

SiamMask: A Framework for Fast Online Object Tracking and Segmentation

85% match1 repo3.6k2022

Cite this paper

@article{yang2023track,
  title  = {Track Anything: Segment Anything Meets Videos},
  author = {Jinyu Yang and Mingqi Gao and Zhe Li and Shang Gao and Fangjing Wang and Feng Zheng},
  year   = {2023},
  eprint = {2304.11968},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2304.11968v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2304.11968.svg)](https://codeofpaper.com/paper/2304.11968)

Discussion