FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3204665409> ?p ?o ?g. }

• W3204665409 abstract "In this work, we address the problem of unsupervised domain adaptation for person re-ID where annotations are available for the source domain but not for target. Previous methods typically follow a two-stage optimization pipeline, where the network is first pre-trained on source and then fine-tuned on target with pseudo labels created by feature clustering. Such methods sustain two main limitations. (1) The label noise may hinder the learning of discriminative features for recognizing target classes. (2) The domain gap may hinder knowledge transferring from source to target. We propose three types of technical schemes to alleviate these issues. First, we propose a cluster-wise contrastive learning algorithm (CCL) by iterative optimization of feature learning and cluster refinery to learn noise-tolerant representations in the unsupervised manner. Second, we adopt a progressive domain adaptation (PDA) strategy to gradually mitigate the domain gap between source and target data. Third, we propose Fourier augmentation (FA) for further maximizing the class separability of re-ID models by imposing extra constraints in the Fourier space. We observe that these proposed schemes are capable of facilitating the learning of discriminative feature representations. Experiments demonstrate that our method consistently achieves notable improvements over the state-of-the-art unsupervised re-ID methods on multiple benchmarks, e.g., surpassing MMT largely by 8.1%, 9.9%, 11.4% and 11.1% mAP on the Market-to-Duke, Duke-to-Market, Market-to-MSMT and Duke-to-MSMT tasks, respectively." @default.
• W3204665409 created "2021-10-11" @default.
• W3204665409 creator A5001764588 @default.
• W3204665409 creator A5009175141 @default.
• W3204665409 creator A5013097653 @default.
• W3204665409 creator A5013914064 @default.
• W3204665409 creator A5030965866 @default.
• W3204665409 creator A5087306292 @default.
• W3204665409 date "2021-10-01" @default.
• W3204665409 modified "2023-09-30" @default.
• W3204665409 title "Towards Discriminative Representation Learning for Unsupervised Person Re-identification" @default.
• W3204665409 cites W1920259731 @default.
• W3204665409 cites W2025768430 @default.
• W3204665409 cites W2072240081 @default.
• W3204665409 cites W2108598243 @default.
• W3204665409 cites W219040644 @default.
• W3204665409 cites W2194775991 @default.
• W3204665409 cites W2204750386 @default.
• W3204665409 cites W2382313035 @default.
• W3204665409 cites W2470322391 @default.
• W3204665409 cites W2575671312 @default.
• W3204665409 cites W2593768305 @default.
• W3204665409 cites W2770645414 @default.
• W3204665409 cites W2798681837 @default.
• W3204665409 cites W2798991696 @default.
• W3204665409 cites W2883725317 @default.
• W3204665409 cites W2884022415 @default.
• W3204665409 cites W2904427185 @default.
• W3204665409 cites W2949813473 @default.
• W3204665409 cites W2953214814 @default.
• W3204665409 cites W2953684117 @default.
• W3204665409 cites W2962687275 @default.
• W3204665409 cites W2962706983 @default.
• W3204665409 cites W2962859295 @default.
• W3204665409 cites W2962926870 @default.
• W3204665409 cites W2963000559 @default.
• W3204665409 cites W2963047834 @default.
• W3204665409 cites W2963289251 @default.
• W3204665409 cites W2963420272 @default.
• W3204665409 cites W2963557071 @default.
• W3204665409 cites W2963842104 @default.
• W3204665409 cites W2963893396 @default.
• W3204665409 cites W2963910742 @default.
• W3204665409 cites W2963975998 @default.
• W3204665409 cites W2964130064 @default.
• W3204665409 cites W2964285681 @default.
• W3204665409 cites W2965744772 @default.
• W3204665409 cites W2987741655 @default.
• W3204665409 cites W2987748894 @default.
• W3204665409 cites W3007744269 @default.
• W3204665409 cites W3008706867 @default.
• W3204665409 cites W3009238880 @default.
• W3204665409 cites W3009761962 @default.
• W3204665409 cites W3034727830 @default.
• W3204665409 cites W3034771037 @default.
• W3204665409 cites W3035014997 @default.
• W3204665409 cites W3035070480 @default.
• W3204665409 cites W3035294798 @default.
• W3204665409 cites W3035402405 @default.
• W3204665409 cites W3035524453 @default.
• W3204665409 cites W3035732677 @default.
• W3204665409 cites W3089800122 @default.
• W3204665409 cites W3106858249 @default.
• W3204665409 cites W3173777106 @default.
• W3204665409 cites W3189423081 @default.
• W3204665409 cites W343636949 @default.
• W3204665409 cites W4214736485 @default.
• W3204665409 doi "https://doi.org/10.1109/iccv48922.2021.00841" @default.
• W3204665409 hasPublicationYear "2021" @default.
• W3204665409 type Work @default.
• W3204665409 sameAs 3204665409 @default.
• W3204665409 citedByCount "37" @default.
• W3204665409 countsByYear W32046654092021 @default.
• W3204665409 countsByYear W32046654092022 @default.
• W3204665409 countsByYear W32046654092023 @default.
• W3204665409 crossrefType "proceedings-article" @default.
• W3204665409 hasAuthorship W3204665409A5001764588 @default.
• W3204665409 hasAuthorship W3204665409A5009175141 @default.
• W3204665409 hasAuthorship W3204665409A5013097653 @default.
• W3204665409 hasAuthorship W3204665409A5013914064 @default.
• W3204665409 hasAuthorship W3204665409A5030965866 @default.
• W3204665409 hasAuthorship W3204665409A5087306292 @default.
• W3204665409 hasBestOaLocation W32046654092 @default.
• W3204665409 hasConcept C108583219 @default.
• W3204665409 hasConcept C115961682 @default.
• W3204665409 hasConcept C119857082 @default.
• W3204665409 hasConcept C138885662 @default.
• W3204665409 hasConcept C153180895 @default.
• W3204665409 hasConcept C154945302 @default.
• W3204665409 hasConcept C199360897 @default.
• W3204665409 hasConcept C2776401178 @default.
• W3204665409 hasConcept C2778827112 @default.
• W3204665409 hasConcept C41008148 @default.
• W3204665409 hasConcept C41895202 @default.
• W3204665409 hasConcept C43521106 @default.
• W3204665409 hasConcept C59404180 @default.
• W3204665409 hasConcept C73555534 @default.
• W3204665409 hasConcept C8038995 @default.
• W3204665409 hasConcept C97931131 @default.
• W3204665409 hasConcept C99498987 @default.

• next