%0 Generic
%A Ponce-Lopez, V
%A Burghardt, T
%A Sun, Y
%A Hannuna, S
%A Damen, D
%A Mirmehdi, M
%C Cham, Switzerland
%D 2019
%E Ricci, E
%E Bulo, SR
%E Snoek, C
%E Lanz, O
%E Messelodi, S
%E Sebe, N
%F discovery:10115347
%I Springer
%K Person Re-ID, GANs, Distractor synthesis, Deep face analysis
%P 488-498
%T Deep Compact Person Re-Identification with Distractor Synthesis via Guided DC-GANs
%U https://discovery.ucl.ac.uk/id/eprint/10115347/
%V 11751
%X We present a dual-stream CNN that learns both appearance and facial features in tandem from still images and, after feature fusion, infers person identities. We then describe an alternative architecture of a single, lightweight ID-CondenseNet where a face detector-guided DC-GAN is used to generate distractor person images for enhanced training. For evaluation, we test both architectures on FLIMA, a new extension of an existing person re-identification dataset with added frame-by-frame annotations of face presence. Although the dual-stream CNN can outperform the CondenseNet approach on FLIMA, we show that the latter surpasses all state-of-the-art architectures in top-1 ranking performance when applied to the largest existing person re-identification dataset, MSMT17. We conclude that whilst re-identification performance is highly sensitive to the structure of datasets, distractor augmentation and network compression have a role to play for enhancing performance characteristics for larger scale applications.
%Z This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions.