Please use this identifier to cite or link to this item:
Title: Learning disentangled representation implicitly via transformer for occluded person re-identification
Authors: Jia, Mengxi
Cheng, Xinhua
Lu, Shijian
Zhang, Jian
Keywords: Engineering::Computer science and engineering
Issue Date: 2022
Source: Jia, M., Cheng, X., Lu, S. & Zhang, J. (2022). Learning disentangled representation implicitly via transformer for occluded person re-identification. IEEE Transactions On Multimedia, 3141267-.
Journal: IEEE Transactions on Multimedia 
Abstract: Person re-IDentification (re-ID) under various occlusions has been a long-standing challenge as person images with different types of occlusions often suffer from misalignment in image matching and ranking. Most existing methods tackle this challenge by aligning spatial features of body parts according to external semantic cues or feature similarities but this alignment approach is complicated and sensitive to noises. We design DRL-Net, a disentangled representation learning network that handles occluded re-ID without requiring strict person image alignment or any additional supervision. Leveraging transformer architectures, DRL-Net achieves alignment-free re-ID via global reasoning of local features of occluded person images. It measures image similarity by automatically disentangling the representation of undefined semantic components, e.g., human body parts or obstacles, under the guidance of semantic preference object queries in the transformer. In addition, we design a decorrelation constraint in the transformer decoder and impose it over object queries for better focus on different semantic components. To better eliminate interference from occlusions, we design a contrast feature learning technique (CFL) for better separation of occlusion features and discriminative ID features. Extensive experiments over occluded and holistic reID benchmarks show that the DRL-Net achieves superior re-ID performance consistently and outperforms the state-of-the-art by large margins for occluded re-ID dataset.
ISSN: 1520-9210
DOI: 10.1109/TMM.2022.3141267
Schools: School of Computer Science and Engineering 
Rights: © 2021 IEEE. All rights reserved.
Fulltext Permission: none
Fulltext Availability: No Fulltext
Appears in Collections:SCSE Journal Articles

Citations 10

Updated on Jul 2, 2024

Web of ScienceTM
Citations 20

Updated on Oct 27, 2023

Page view(s)

Updated on Jul 12, 2024

Google ScholarTM




Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.