Multi-Task Mutual Learning for Vehicle Re-Identification

Aytac Kanaci, Minxian Li, Shaogang Gong, Georgia Rajamanoharan; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2019, pp. 62-70


Vehicle re-identification (Re-ID) aims to search a specific vehicle instance across non-overlapping camera views. The main challenge of vehicle Re-ID is that the visual appearance of vehicles may drastically changes according to diverse viewpoints and illumination. Most existing vehicle Re-ID models cannot make full use of various complementary vehicle information, e.g. vehicle type and orientation. In this paper, we propose a novel Multi-Task Mutual Learning (MTML) deep model to learn discriminative features simultaneously from multiple branches. Specifically, we design a consensus learning loss function by fusing features from the final convolutional feature maps from all branches. Extensive comparative evaluations demonstrate the effectiveness of our proposed MTML method in comparison to the state-of-the-art vehicle Re-ID techniques on a large-scale benchmark dataset, VeRi-776. We also yield competitive performance on the NVIDIA 2019 AI City Challenge Track 2.

Related Material

author = {Kanaci, Aytac and Li, Minxian and Gong, Shaogang and Rajamanoharan, Georgia},
title = {Multi-Task Mutual Learning for Vehicle Re-Identification},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2019}