%P 488-498
%O This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions.
%T Deep Compact Person Re-Identification with Distractor Synthesis via Guided DC-GANs
%L discovery10115347
%I Springer
%D 2019
%C Cham, Switzerland
%J IMAGE ANALYSIS AND PROCESSING - ICIAP 2019, PT I
%S Lecture Notes in Computer Science
%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.
%B Image Analysis and Processing – ICIAP 2019. ICIAP 2019
%E E Ricci
%E SR Bulo
%E C Snoek
%E O Lanz
%E S Messelodi
%E N Sebe
%K Person Re-ID, GANs, Distractor synthesis, Deep face analysis
%V 11751
%A V Ponce-Lopez
%A T Burghardt
%A Y Sun
%A S Hannuna
%A D Damen
%A M Mirmehdi