Matches in SemOpenAlex for { <https://semopenalex.org/work/W3009869848> ?p ?o ?g. }
- W3009869848 abstract "Existing fully-supervised person re-identification (ReID) methods usually suffer from poor generalization capability caused by domain gaps. The key to solving this problem lies in filtering out identity-irrelevant interference and learning domain-invariant person representations. In this paper, we aim to design a generalizable person ReID framework which trains a model on source domains yet is able to generalize/perform well on target domains. To achieve this goal, we propose a simple yet effective Style Normalization and Restitution (SNR) module. Specifically, we filter out style variations (e.g., illumination, color contrast) by Instance Normalization (IN). However, such a process inevitably removes discriminative information. We propose to distill identity-relevant feature from the removed information and restitute it to the network to ensure high discrimination. For better disentanglement, we enforce a dual causal loss constraint in SNR to encourage the separation of identity-relevant features and identity-irrelevant features. Extensive experiments demonstrate the strong generalization capability of our framework. Our models empowered by the SNR modules significantly outperform the state-of-the-art domain generalization approaches on multiple widely-used person ReID benchmarks, and also show superiority on unsupervised domain adaptation." @default.
- W3009869848 created "2020-03-13" @default.
- W3009869848 creator A5049963367 @default.
- W3009869848 creator A5051469345 @default.
- W3009869848 creator A5062901935 @default.
- W3009869848 creator A5066716873 @default.
- W3009869848 creator A5079572598 @default.
- W3009869848 date "2020-05-22" @default.
- W3009869848 modified "2023-09-24" @default.
- W3009869848 title "Style Normalization and Restitution for Generalizable Person Re-identification" @default.
- W3009869848 cites W1518138188 @default.
- W3009869848 cites W1596233070 @default.
- W3009869848 cites W1852255964 @default.
- W3009869848 cites W1949591461 @default.
- W3009869848 cites W1979838576 @default.
- W3009869848 cites W1982925187 @default.
- W3009869848 cites W2047632871 @default.
- W3009869848 cites W2139763424 @default.
- W3009869848 cites W2155858138 @default.
- W3009869848 cites W2161969291 @default.
- W3009869848 cites W2183341477 @default.
- W3009869848 cites W2187089797 @default.
- W3009869848 cites W2194775991 @default.
- W3009869848 cites W2220271458 @default.
- W3009869848 cites W2299668505 @default.
- W3009869848 cites W2300840837 @default.
- W3009869848 cites W2339827301 @default.
- W3009869848 cites W2353169560 @default.
- W3009869848 cites W2502312327 @default.
- W3009869848 cites W2572730214 @default.
- W3009869848 cites W2598634450 @default.
- W3009869848 cites W2603777577 @default.
- W3009869848 cites W2736410039 @default.
- W3009869848 cites W2777534232 @default.
- W3009869848 cites W2783702274 @default.
- W3009869848 cites W2787226294 @default.
- W3009869848 cites W2788846526 @default.
- W3009869848 cites W2798718722 @default.
- W3009869848 cites W2798794112 @default.
- W3009869848 cites W2808134123 @default.
- W3009869848 cites W2808260522 @default.
- W3009869848 cites W2884197239 @default.
- W3009869848 cites W2884366600 @default.
- W3009869848 cites W2890159224 @default.
- W3009869848 cites W2896016251 @default.
- W3009869848 cites W2898047322 @default.
- W3009869848 cites W2904949947 @default.
- W3009869848 cites W2922134245 @default.
- W3009869848 cites W2944780724 @default.
- W3009869848 cites W2949117887 @default.
- W3009869848 cites W2949257576 @default.
- W3009869848 cites W2949436635 @default.
- W3009869848 cites W2953054324 @default.
- W3009869848 cites W2954773727 @default.
- W3009869848 cites W2957037461 @default.
- W3009869848 cites W2962859295 @default.
- W3009869848 cites W2963000559 @default.
- W3009869848 cites W2963043696 @default.
- W3009869848 cites W2963067443 @default.
- W3009869848 cites W2963078173 @default.
- W3009869848 cites W2963180826 @default.
- W3009869848 cites W2963314703 @default.
- W3009869848 cites W2963420686 @default.
- W3009869848 cites W2963438548 @default.
- W3009869848 cites W2963557071 @default.
- W3009869848 cites W2963842104 @default.
- W3009869848 cites W2963983343 @default.
- W3009869848 cites W2964121744 @default.
- W3009869848 cites W2964163358 @default.
- W3009869848 cites W2966586228 @default.
- W3009869848 cites W2967515867 @default.
- W3009869848 cites W2970290171 @default.
- W3009869848 cites W2984145721 @default.
- W3009869848 cites W2988852559 @default.
- W3009869848 cites W2997351135 @default.
- W3009869848 cites W2997987796 @default.
- W3009869848 cites W3007744269 @default.
- W3009869848 cites W3009761962 @default.
- W3009869848 doi "https://doi.org/10.48550/arxiv.2005.11037" @default.
- W3009869848 hasPublicationYear "2020" @default.
- W3009869848 type Work @default.
- W3009869848 sameAs 3009869848 @default.
- W3009869848 citedByCount "2" @default.
- W3009869848 countsByYear W30098698482020 @default.
- W3009869848 countsByYear W30098698482021 @default.
- W3009869848 crossrefType "posted-content" @default.
- W3009869848 hasAuthorship W3009869848A5049963367 @default.
- W3009869848 hasAuthorship W3009869848A5051469345 @default.
- W3009869848 hasAuthorship W3009869848A5062901935 @default.
- W3009869848 hasAuthorship W3009869848A5066716873 @default.
- W3009869848 hasAuthorship W3009869848A5079572598 @default.
- W3009869848 hasBestOaLocation W30098698481 @default.
- W3009869848 hasConcept C106131492 @default.
- W3009869848 hasConcept C119857082 @default.
- W3009869848 hasConcept C121332964 @default.
- W3009869848 hasConcept C134306372 @default.
- W3009869848 hasConcept C136886441 @default.
- W3009869848 hasConcept C144024400 @default.
- W3009869848 hasConcept C154945302 @default.
- W3009869848 hasConcept C177148314 @default.