Global Instance Tracking: Locating Target More Like Humans

• URL: http://arxiv.org/abs/2202.13073v1
• Date: Sat, 26 Feb 2022 06:16:34 GMT
• Title: Global Instance Tracking: Locating Target More Like Humans
• Authors: Shiyu Hu, Xin Zhao, Lianghua Huang, Kaiqi Huang
• Abstract summary: Target tracking, the essential ability of the human visual system, has been simulated by computer vision tasks. Existing trackers perform well in austere experimental environments but fail in challenges like occlusion and fast motion. We propose the global instance tracking (GIT) task, which is supposed to search an arbitrary user-specified instance in a video.
• Score: 47.99395323689126
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• 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.

Related papers

• BEVTrack: A Simple and Strong Baseline for 3D Single Object Tracking in Bird's-Eye View [56.77287041917277]
  3D Single Object Tracking (SOT) is a fundamental task of computer vision, proving essential for applications like autonomous driving. In this paper, we propose BEVTrack, a simple yet effective baseline method. By estimating the target motion in Bird's-Eye View (BEV) to perform tracking, BEVTrack demonstrates surprising simplicity from various aspects, i.e., network designs, training objectives, and tracking pipeline, while achieving superior performance.
  arXiv  Detail & Related papers  (2023-09-05T12:42:26Z)
• Self-Improving Robots: End-to-End Autonomous Visuomotor Reinforcement Learning [54.636562516974884]
  In imitation and reinforcement learning, the cost of human supervision limits the amount of data that robots can be trained on. In this work, we propose MEDAL++, a novel design for self-improving robotic systems. The robot autonomously practices the task by learning to both do and undo the task, simultaneously inferring the reward function from the demonstrations.
  arXiv  Detail & Related papers  (2023-03-02T18:51:38Z)
• Person Monitoring by Full Body Tracking in Uniform Crowd Environment [10.71804432329509]
  In the Middle East, uniform crowd environments are the norm which challenges state-of-the-art trackers. In this work, we develop an annotated dataset with one specific target per video in a uniform crowd environment. The dataset was used in evaluating and fine-tuning a state-of-the-art tracker.
  arXiv  Detail & Related papers  (2022-09-02T21:21:47Z)
• Robot Person Following in Uniform Crowd Environment [13.708992331117281]
  Person-tracking robots have many applications, such as in security, elderly care, and socializing robots. In this work, we focus on improving the perceptivity of a robot for a person following task by developing a robust and real-time applicable object tracker. We present a new robot person tracking system with a new RGB-D tracker, Deep Tracking with RGB-D (DTRD) that is resilient to tricky challenges introduced by the uniform crowd environment.
  arXiv  Detail & Related papers  (2022-05-21T10:20:14Z)
• Single Object Tracking Research: A Survey [44.24280758718638]
  This paper presents the rationale and works of two most popular tracking frameworks in past ten years. We present some deep learning based tracking methods categorized by different network structures. We also introduce some classical strategies for handling the challenges in tracking problem.
  arXiv  Detail & Related papers  (2022-04-25T02:59:15Z)
• Learning to Track Objects from Unlabeled Videos [63.149201681380305]
  In this paper, we propose to learn an Unsupervised Single Object Tracker (USOT) from scratch. To narrow the gap between unsupervised trackers and supervised counterparts, we propose an effective unsupervised learning approach composed of three stages. Experiments show that the proposed USOT learned from unlabeled videos performs well over the state-of-the-art unsupervised trackers by large margins.
  arXiv  Detail & Related papers  (2021-08-28T22:10:06Z)
• Benchmarking high-fidelity pedestrian tracking systems for research, real-time monitoring and crowd control [55.41644538483948]
  High-fidelity pedestrian tracking in real-life conditions has been an important tool in fundamental crowd dynamics research. As this technology advances, it is becoming increasingly useful also in society. To successfully employ pedestrian tracking techniques in research and technology, it is crucial to validate and benchmark them for accuracy. We present and discuss a benchmark suite, towards an open standard in the community, for privacy-respectful pedestrian tracking techniques.
  arXiv  Detail & Related papers  (2021-08-26T11:45:26Z)
• Tracking Without Re-recognition in Humans and Machines [12.591847867999636]
  We investigate if state-of-the-art deep neural networks for visual tracking are capable of the same. We introduce PathTracker, a synthetic visual challenge that asks human observers and machines to track a target object. We model circuit mechanisms in biological brains that are implicated in tracking objects based on motion cues.
  arXiv  Detail & Related papers  (2021-05-27T17:56:37Z)
• MOTChallenge: A Benchmark for Single-Camera Multiple Target Tracking [72.76685780516371]
  We present MOTChallenge, a benchmark for single-camera Multiple Object Tracking (MOT) The benchmark is focused on multiple people tracking, since pedestrians are by far the most studied object in the tracking community. We provide a categorization of state-of-the-art trackers and a broad error analysis.
  arXiv  Detail & Related papers  (2020-10-15T06:52:16Z)
• Kalman Filter Based Multiple Person Head Tracking [10.235946073664818]
  State-of-the-art approaches rely on the deep learning-based visual representation. We come up with a simple yet effective target representation for human tracking.
  arXiv  Detail & Related papers  (2020-06-11T00:54:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.