Learning Data Association for Multi-Object Tracking using Only Coordinates

Abstract

We propose a novel Transformer-based module to address the data association problem for multi-object tracking. From detections obtained by a pretrained detector, this module uses only coordinates from bounding boxes to estimate an affinity score between pairs of tracks extracted from two distinct temporal windows. This module, named TWiX, is trained on sets of tracks with the objective of discriminating pairs of tracks coming from the same object from those which are not. Our module does not use the intersection over union measure, nor does it requires any motion priors or any camera motion compensation technique. By inserting TWiX within an online cascade matching pipeline, our tracker C-TWiX achieves state-of-the-art performance on the DanceTrack and KITTIMOT datasets, and gets competitive results on the MOT17 dataset. The code will be made available upon publication.

Preprint submitted to Pattern Recognition

References in corpus (8)

• Bootstrap your own latent: A new approach to self-supervised Learning
• YOLOX: Exceeding YOLO Series in 2021
• In Defense of the Triplet Loss for Person Re-Identification
• Tracking without bells and whistles
• TransTrack: Multiple Object Tracking with Transformer
• BoT-SORT: Robust Associations Multi-Pedestrian Tracking
• Semi-TCL: Semi-Supervised Track Contrastive Representation Learning
• SparseTrack: Multi-Object Tracking by Performing Scene Decomposition based on Pseudo-Depth

Cited by in corpus (2)

• Tracking by Detection and Query: An Efficient End-to-End Framework for Multi-Object Tracking
• FeatureSORT: Essential Features for Effective Tracking