Associative affinity network learning for multi-object tracking

Liang Ma,Qiaoyong Zhong,Yingying Zhang,Di Xie,Shiliang Pu

DOI: https://doi.org/10.1631/FITEE.2000272

2021-09-17

Abstract:We propose a joint feature and metric learning deep neural network architecture, called the associative affinity network (AAN), as an affinity model for multi-object tracking (MOT) in videos. The AAN learns the associative affinity between tracks and detections across frames in an end-to-end manner. Considering flawed detections, the AAN jointly learns bounding box regression, classification, and affinity regression via the proposed multi-task loss. Contrary to networks that are trained with ranking loss, we directly train a binary classifier to learn the associative affinity of each track-detection pair and use a matching cardinality loss to capture information among candidate pairs. The AAN learns a discriminative affinity model for data association to tackle MOT, and can also perform single-object tracking. Based on the AAN, we propose a simple multi-object tracker that achieves competitive performance on the public MOT16 and MOT17 test datasets.

engineering, electrical & electronic,computer science, information systems, software engineering