English

Global Instance Tracking: Locating Target More Like Humans

Computer Vision and Pattern Recognition 2022-03-01 v1

Abstract

Target tracking, the essential ability of the human visual system, has been simulated by computer vision tasks. However, existing trackers perform well in austere experimental environments but fail in challenges like occlusion and fast motion. The massive gap indicates that researches only measure tracking performance rather than intelligence. How to scientifically judge the intelligence level of trackers? Distinct from decision-making problems, lacking three requirements (a challenging task, a fair environment, and a scientific evaluation procedure) makes it strenuous to answer the question. In this article, we first propose the global instance tracking (GIT) task, which is supposed to search an arbitrary user-specified instance in a video without any assumptions about camera or motion consistency, to model the human visual tracking ability. Whereafter, we construct a high-quality and large-scale benchmark VideoCube to create a challenging environment. Finally, we design a scientific evaluation procedure using human capabilities as the baseline to judge tracking intelligence. Additionally, we provide an online platform with toolkit and an updated leaderboard. Although the experimental results indicate a definite gap between trackers and humans, we expect to take a step forward to generate authentic human-like trackers. The database, toolkit, evaluation server, and baseline results are available at http://videocube.aitestunion.com.

Keywords

Cite

@article{arxiv.2202.13073,
  title  = {Global Instance Tracking: Locating Target More Like Humans},
  author = {Shiyu Hu and Xin Zhao and Lianghua Huang and Kaiqi Huang},
  journal= {arXiv preprint arXiv:2202.13073},
  year   = {2022}
}

Comments

This paper is published in IEEE TPAMI (refer to DOI). Please cite the published IEEE TPAMI

R2 v1 2026-06-24T09:54:43.536Z