Matches in SemOpenAlex for { <https://semopenalex.org/work/W4366207256> ?p ?o ?g. }
Showing items 1 to 79 of
79
with 100 items per page.
- W4366207256 abstract "Pedestrian attribute recognition (PAR) has received increasing attention because of its wide application in video surveillance and pedestrian analysis. Extracting robust feature representation is one of the key challenges in this task. The existing methods mainly use the convolutional neural network (CNN) as the backbone network to extract features. However, these methods mainly focus on small discriminative regions while ignoring the global perspective. To overcome these limitations, we propose a pure transformer-based multi-task PAR network named PARFormer, which includes four modules. In the feature extraction module, we build a transformer-based strong baseline for feature extraction, which achieves competitive results on several PAR benchmarks compared with the existing CNN-based baseline methods. In the feature processing module, we propose an effective data augmentation strategy named batch random mask (BRM) block to reinforce the attentive feature learning of random patches. Furthermore, we propose a multi-attribute center loss (MACL) to enhance the inter-attribute discriminability in the feature representations. In the viewpoint perception module, we explore the impact of viewpoints on pedestrian attributes, and propose a multi-view contrastive loss (MCVL) that enables the network to exploit the viewpoint information. In the attribute recognition module, we alleviate the negative-positive imbalance problem to generate the attribute predictions. The above modules interact and jointly learn a highly discriminative feature space, and supervise the generation of the final features. Extensive experimental results show that the proposed PARFormer network performs well compared to the state-of-the-art methods on several public datasets, including PETA, RAP, and PA100K. Code will be released at https://github.com/xwf199/PARFormer." @default.
- W4366207256 created "2023-04-19" @default.
- W4366207256 creator A5044594971 @default.
- W4366207256 creator A5083861295 @default.
- W4366207256 creator A5085353674 @default.
- W4366207256 creator A5088931934 @default.
- W4366207256 date "2023-04-14" @default.
- W4366207256 modified "2023-10-11" @default.
- W4366207256 title "PARFormer: Transformer-based Multi-Task Network for Pedestrian Attribute Recognition" @default.
- W4366207256 doi "https://doi.org/10.48550/arxiv.2304.07230" @default.
- W4366207256 hasPublicationYear "2023" @default.
- W4366207256 type Work @default.
- W4366207256 citedByCount "0" @default.
- W4366207256 crossrefType "posted-content" @default.
- W4366207256 hasAuthorship W4366207256A5044594971 @default.
- W4366207256 hasAuthorship W4366207256A5083861295 @default.
- W4366207256 hasAuthorship W4366207256A5085353674 @default.
- W4366207256 hasAuthorship W4366207256A5088931934 @default.
- W4366207256 hasBestOaLocation W43662072561 @default.
- W4366207256 hasConcept C119599485 @default.
- W4366207256 hasConcept C119857082 @default.
- W4366207256 hasConcept C124101348 @default.
- W4366207256 hasConcept C127413603 @default.
- W4366207256 hasConcept C138885662 @default.
- W4366207256 hasConcept C142362112 @default.
- W4366207256 hasConcept C153180895 @default.
- W4366207256 hasConcept C153349607 @default.
- W4366207256 hasConcept C154945302 @default.
- W4366207256 hasConcept C165801399 @default.
- W4366207256 hasConcept C22212356 @default.
- W4366207256 hasConcept C2776035091 @default.
- W4366207256 hasConcept C2776401178 @default.
- W4366207256 hasConcept C2777113093 @default.
- W4366207256 hasConcept C2780156472 @default.
- W4366207256 hasConcept C41008148 @default.
- W4366207256 hasConcept C41895202 @default.
- W4366207256 hasConcept C52622490 @default.
- W4366207256 hasConcept C59404180 @default.
- W4366207256 hasConcept C66322947 @default.
- W4366207256 hasConcept C81363708 @default.
- W4366207256 hasConcept C97931131 @default.
- W4366207256 hasConceptScore W4366207256C119599485 @default.
- W4366207256 hasConceptScore W4366207256C119857082 @default.
- W4366207256 hasConceptScore W4366207256C124101348 @default.
- W4366207256 hasConceptScore W4366207256C127413603 @default.
- W4366207256 hasConceptScore W4366207256C138885662 @default.
- W4366207256 hasConceptScore W4366207256C142362112 @default.
- W4366207256 hasConceptScore W4366207256C153180895 @default.
- W4366207256 hasConceptScore W4366207256C153349607 @default.
- W4366207256 hasConceptScore W4366207256C154945302 @default.
- W4366207256 hasConceptScore W4366207256C165801399 @default.
- W4366207256 hasConceptScore W4366207256C22212356 @default.
- W4366207256 hasConceptScore W4366207256C2776035091 @default.
- W4366207256 hasConceptScore W4366207256C2776401178 @default.
- W4366207256 hasConceptScore W4366207256C2777113093 @default.
- W4366207256 hasConceptScore W4366207256C2780156472 @default.
- W4366207256 hasConceptScore W4366207256C41008148 @default.
- W4366207256 hasConceptScore W4366207256C41895202 @default.
- W4366207256 hasConceptScore W4366207256C52622490 @default.
- W4366207256 hasConceptScore W4366207256C59404180 @default.
- W4366207256 hasConceptScore W4366207256C66322947 @default.
- W4366207256 hasConceptScore W4366207256C81363708 @default.
- W4366207256 hasConceptScore W4366207256C97931131 @default.
- W4366207256 hasLocation W43662072561 @default.
- W4366207256 hasOpenAccess W4366207256 @default.
- W4366207256 hasPrimaryLocation W43662072561 @default.
- W4366207256 hasRelatedWork W2404514746 @default.
- W4366207256 hasRelatedWork W2406522397 @default.
- W4366207256 hasRelatedWork W2518599539 @default.
- W4366207256 hasRelatedWork W2725397116 @default.
- W4366207256 hasRelatedWork W2743258233 @default.
- W4366207256 hasRelatedWork W2806866760 @default.
- W4366207256 hasRelatedWork W2905846897 @default.
- W4366207256 hasRelatedWork W2998168123 @default.
- W4366207256 hasRelatedWork W3204901196 @default.
- W4366207256 hasRelatedWork W4287995534 @default.
- W4366207256 isParatext "false" @default.
- W4366207256 isRetracted "false" @default.
- W4366207256 workType "article" @default.