Transferring learning from multi-person tracking to person re-identification

Thumbnail Image
Publication date
Defense date
Journal Title
Journal ISSN
Volume Title
IOP Press
Google Scholar
Research Projects
Organizational Units
Journal Issue
Learning to discriminate, whether two person-images correspond to the same person or not, is a daunting challenge when only two images per person are available. This task is called single-shot person re-identification (re-id) and it assumes that each one of the two available images was captured from a different camera view entailing variations in pose, resolution, scale, illumination and background. Addressing this task through supervised training of a deep convolutional neural network is susceptible to model overfitting due to the critical lack of enough labelled data. This paper proposes to exploit the transference of learning previously acquired from a multi-object-tracking (MOT) domain. In this context, a unique deep triplet architecture has been trained on both domains. Six different levels of transfer learning have been implemented and evaluated, proving that the transference of leaning from a different domain remarkably increases the re-id performance. Experimental results validate accuracy and robustness of the proposed method as comparable to other state-of-the-art techniques. These results also confirm that, despite the data problem, deep learning is also applicable to the single-shot re-id task.
Transfer learning, Deep learning, Person re-identification, Multi-object tracking, Pair-wise binary classification
Bibliographic citation
M.J. Gómez-Silva, E. Izquierdo, Escalera, A. de la and Armingol, J.M. “Transferring Learning from Multi-person Tracking to Person Re-identification”, Integrated Computer-Aided Engineering, vol. 26, nº 4, pp. 329-344, 1 Jan. 2019