Matches in SemOpenAlex for { <https://semopenalex.org/work/W4226280080> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W4226280080 endingPage "798" @default.
- W4226280080 startingPage "790" @default.
- W4226280080 abstract "Clustering-based unsupervised domain adaptive (UDA) person re-identification (ReID) reduces exhaustive annotations. However, owing to unsatisfactory feature embedding and imperfect clustering, pseudo labels for target domain data inherently contain an unknown proportion of wrong ones, which would mislead feature learning. In this paper, we propose an approach named probabilistic uncertainty guided progressive label refinery (P2LR) for domain adaptive person re-identification. First, we propose to model the labeling uncertainty with the probabilistic distance along with ideal single-peak distributions. A quantitative criterion is established to measure the uncertainty of pseudo labels and facilitate the network training. Second, we explore a progressive strategy for refining pseudo labels. With the uncertainty-guided alternative optimization, we balance between the exploration of target domain data and the negative effects of noisy labeling. On top of a strong baseline, we obtain significant improvements and achieve the state-of-the-art performance on four UDA ReID benchmarks. Specifically, our method outperforms the baseline by 6.5% mAP on the Duke2Market task, while surpassing the state-of-the-art method by 2.5% mAP on the Market2MSMT task. Code is available at: https://github.com/JeyesHan/P2LR." @default.
- W4226280080 created "2022-05-05" @default.
- W4226280080 creator A5011907016 @default.
- W4226280080 creator A5030965866 @default.
- W4226280080 creator A5086144161 @default.
- W4226280080 date "2022-06-28" @default.
- W4226280080 modified "2023-10-16" @default.
- W4226280080 title "Delving into Probabilistic Uncertainty for Unsupervised Domain Adaptive Person Re-identification" @default.
- W4226280080 doi "https://doi.org/10.1609/aaai.v36i1.19960" @default.
- W4226280080 hasPublicationYear "2022" @default.
- W4226280080 type Work @default.
- W4226280080 citedByCount "6" @default.
- W4226280080 countsByYear W42262800802022 @default.
- W4226280080 countsByYear W42262800802023 @default.
- W4226280080 crossrefType "journal-article" @default.
- W4226280080 hasAuthorship W4226280080A5011907016 @default.
- W4226280080 hasAuthorship W4226280080A5030965866 @default.
- W4226280080 hasAuthorship W4226280080A5086144161 @default.
- W4226280080 hasBestOaLocation W42262800801 @default.
- W4226280080 hasConcept C111368507 @default.
- W4226280080 hasConcept C116834253 @default.
- W4226280080 hasConcept C119857082 @default.
- W4226280080 hasConcept C124101348 @default.
- W4226280080 hasConcept C12725497 @default.
- W4226280080 hasConcept C127313418 @default.
- W4226280080 hasConcept C134306372 @default.
- W4226280080 hasConcept C138885662 @default.
- W4226280080 hasConcept C153180895 @default.
- W4226280080 hasConcept C154945302 @default.
- W4226280080 hasConcept C162324750 @default.
- W4226280080 hasConcept C187736073 @default.
- W4226280080 hasConcept C2776401178 @default.
- W4226280080 hasConcept C2780451532 @default.
- W4226280080 hasConcept C33923547 @default.
- W4226280080 hasConcept C36503486 @default.
- W4226280080 hasConcept C41008148 @default.
- W4226280080 hasConcept C41608201 @default.
- W4226280080 hasConcept C41895202 @default.
- W4226280080 hasConcept C49937458 @default.
- W4226280080 hasConcept C59822182 @default.
- W4226280080 hasConcept C73555534 @default.
- W4226280080 hasConcept C86803240 @default.
- W4226280080 hasConceptScore W4226280080C111368507 @default.
- W4226280080 hasConceptScore W4226280080C116834253 @default.
- W4226280080 hasConceptScore W4226280080C119857082 @default.
- W4226280080 hasConceptScore W4226280080C124101348 @default.
- W4226280080 hasConceptScore W4226280080C12725497 @default.
- W4226280080 hasConceptScore W4226280080C127313418 @default.
- W4226280080 hasConceptScore W4226280080C134306372 @default.
- W4226280080 hasConceptScore W4226280080C138885662 @default.
- W4226280080 hasConceptScore W4226280080C153180895 @default.
- W4226280080 hasConceptScore W4226280080C154945302 @default.
- W4226280080 hasConceptScore W4226280080C162324750 @default.
- W4226280080 hasConceptScore W4226280080C187736073 @default.
- W4226280080 hasConceptScore W4226280080C2776401178 @default.
- W4226280080 hasConceptScore W4226280080C2780451532 @default.
- W4226280080 hasConceptScore W4226280080C33923547 @default.
- W4226280080 hasConceptScore W4226280080C36503486 @default.
- W4226280080 hasConceptScore W4226280080C41008148 @default.
- W4226280080 hasConceptScore W4226280080C41608201 @default.
- W4226280080 hasConceptScore W4226280080C41895202 @default.
- W4226280080 hasConceptScore W4226280080C49937458 @default.
- W4226280080 hasConceptScore W4226280080C59822182 @default.
- W4226280080 hasConceptScore W4226280080C73555534 @default.
- W4226280080 hasConceptScore W4226280080C86803240 @default.
- W4226280080 hasIssue "1" @default.
- W4226280080 hasLocation W42262800801 @default.
- W4226280080 hasLocation W42262800802 @default.
- W4226280080 hasOpenAccess W4226280080 @default.
- W4226280080 hasPrimaryLocation W42262800801 @default.
- W4226280080 hasRelatedWork W2382607599 @default.
- W4226280080 hasRelatedWork W2546942002 @default.
- W4226280080 hasRelatedWork W2955042007 @default.
- W4226280080 hasRelatedWork W2961085424 @default.
- W4226280080 hasRelatedWork W2970216048 @default.
- W4226280080 hasRelatedWork W4290792893 @default.
- W4226280080 hasRelatedWork W4306674287 @default.
- W4226280080 hasRelatedWork W4319453497 @default.
- W4226280080 hasRelatedWork W4364383453 @default.
- W4226280080 hasRelatedWork W4224009465 @default.
- W4226280080 hasVolume "36" @default.
- W4226280080 isParatext "false" @default.
- W4226280080 isRetracted "false" @default.
- W4226280080 workType "article" @default.