Tracking the Trackers: An Analysis of the State of the Art in Multiple Object Tracking

Abstract

Standardized benchmarks are crucial for the majority of computer vision applications. Although leaderboards and ranking tables should not be over-claimed, benchmarks often provide the most objective measure of performance and are therefore important guides for research. We present a benchmark for Multiple Object Tracking launched in the late 2014, with the goal of creating a framework for the standardized evaluation of multiple object tracking methods. This paper collects the two releases of the benchmark made so far, and provides an in-depth analysis of almost 50 state-of-the-art trackers that were tested on over 11000 frames. We show the current trends and weaknesses of multiple people tracking methods, and provide pointers of what researchers should be focusing on to push the field forward.

References in corpus (2)

• The Cityscapes Dataset for Semantic Urban Scene Understanding
• Tracking with multi-level features

Cited by in corpus (22)

• Tracking without bells and whistles
• Deep Learning in Video Multi-Object Tracking: A Survey
• Pedestrian Models for Autonomous Driving Part I: Low-Level Models, from Sensing to Tracking
• Multi-Target, Multi-Camera Tracking by Hierarchical Clustering: Recent Progress on DukeMTMC Project
• Video Big Data Analytics in the Cloud: A Reference Architecture, Survey, Opportunities, and Open Research Issues
• Memory Based Online Learning of Deep Representations from Video Streams
• Spatial-Temporal Relation Networks for Multi-Object Tracking
• Quasi-Dense Similarity Learning for Multiple Object Tracking
• Tracking Objects as Points
• Joint Detection and Tracking in Videos with Identification Features
• Locality Aware Appearance Metric for Multi-Target Multi-Camera Tracking
• SG-Net: Spatial Granularity Network for One-Stage Video Instance Segmentation
• Fusion of Head and Full-Body Detectors for Multi-Object Tracking
• The SpaceNet Multi-Temporal Urban Development Challenge
• Learning a Proposal Classifier for Multiple Object Tracking
• Adaptive Affinity for Associations in Multi-Target Multi-Camera Tracking
• ArTIST: Autoregressive Trajectory Inpainting and Scoring for Tracking
• Learning to associate detections for real-time multiple object tracking
• Efficient Global Multi-object Tracking Under Minimum-cost Circulation Framework
• Computational Anatomy for Multi-Organ Analysis in Medical Imaging: A Review
• Handcrafted and Deep Trackers: Recent Visual Object Tracking Approaches and Trends
• A two-stage data association approach for 3D Multi-object Tracking