English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Pedestrian attribute recognition (PAR) has received increasing attentionbecause of its wide application in video surveillance and pedestrian analysis.Extracting robust feature representation is one of the key challenges in thistask. The existing methods mainly use the convolutional neural network (CNN) asthe backbone network to extract features. However, these methods mainly focuson small discriminative regions while ignoring the global perspective. Toovercome these limitations, we propose a pure transformer-based multi-task PARnetwork named PARFormer, which includes four modules. In the feature extractionmodule, we build a transformer-based strong baseline for feature extraction,which achieves competitive results on several PAR benchmarks compared with theexisting CNN-based baseline methods. In the feature processing module, wepropose 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 theinter-attribute discriminability in the feature representations. In theviewpoint perception module, we explore the impact of viewpoints on pedestrianattributes, and propose a multi-view contrastive loss (MCVL) that enables thenetwork to exploit the viewpoint information. In the attribute recognitionmodule, we alleviate the negative-positive imbalance problem to generate theattribute predictions. The above modules interact and jointly learn a highlydiscriminative feature space, and supervise the generation of the finalfeatures. Extensive experimental results show that the proposed PARFormernetwork performs well compared to the state-of-the-art methods on severalpublic datasets, including PETA, RAP, and PA100K. Code will be released athttps://github.com/xwf199/PARFormer.

PARFormer: Transformer-based Multi-Task Network for Pedestrian Attribute Recognition