Skip to content

Deep Compact Person Re-Identification with Distractor Synthesis via Guided DC-GANs

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Original languageEnglish
Title of host publicationInternational Conference on Image Analysis and Processing
Subtitle of host publicationLecture Notes in Computer Science
Publisher or commissioning bodySpringer
Pages488-498
Number of pages11
Volume11751
ISBN (Electronic)9783030306427
ISBN (Print)9783030306410
DOIs
DateAccepted/In press - 21 Jun 2019
DatePublished (current) - 2 Sep 2019

Publication series

NameLecture Notes in Computer Science
PublisherSpringer Cham
Volume11751
ISSN (Electronic)0302-9743

Abstract

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.

Documents

View research connections

Related faculties, schools or groups