GCNNMatch: Graph Convolutional Neural Networks for Multi-Object Tracking via Sinkhorn Normalization

Abstract

This paper proposes a novel method for online Multi-Object Tracking (MOT) using Graph Convolutional Neural Network (GCNN) based feature extraction and end-to-end feature matching for object association. The Graph based approach incorporates both appearance and geometry of objects at past frames as well as the current frame into the task of feature learning. This new paradigm enables the network to leverage the "context" information of the geometry of objects and allows us to model the interactions among the features of multiple objects. Another central innovation of our proposed framework is the use of the Sinkhorn algorithm for end-to-end learning of the associations among objects during model training. The network is trained to predict object associations by taking into account constraints specific to the MOT task. Experimental results demonstrate the efficacy of the proposed approach in achieving top performance on the MOT '15, '16, '17 and '20 Challenges among state-of-the-art online approaches. The code is available at https://github.com/IPapakis/GCNNMatch.

References in corpus (10)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• MOT16: A Benchmark for Multi-Object Tracking
• Fast Graph Representation Learning with PyTorch Geometric
• Deep Learning in Video Multi-Object Tracking: A Survey
• Multi-Object Tracking with Multiple Cues and Switcher-Aware Classification
• Survey on Deep Learning Techniques for Person Re-Identification Task
• Graph Neural Based End-to-end Data Association Framework for Online Multiple-Object Tracking
• Similarity Mapping with Enhanced Siamese Network for Multi-Object Tracking
• Unsupervised Multiple Person Tracking using AutoEncoder-Based Lifted Multicuts

Cited by in corpus (7)

• Learnable Graph Matching: Incorporating Graph Partitioning with Deep Feature Learning for Multiple Object Tracking
• Online and Real-Time Tracking in a Surveillance Scenario
• Learning a Proposal Classifier for Multiple Object Tracking
• Local Metrics for Multi-Object Tracking
• Simple online and real-time tracking with occlusion handling
• Split and Connect: A Universal Tracklet Booster for Multi-Object Tracking
• Synthetic Data Are as Good as the Real for Association Knowledge Learning in Multi-object Tracking