Attention-Based Pedestrian Attribute Analysis

• Published in: IEEE transactions on image processing Vol. 28; no. 12; pp. 6126 - 6140
• Main Authors: Tan, Zichang, Yang, Yang, Wan, Jun, Hang, Hanyuan, Guo, Guodong, Li, Stan Z.
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.12.2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Aggregates; attention mechanism; Biometrics (access control); Deep learning; Feature extraction; Image recognition; Learning; Pedestrian attribute analysis; pedestrian parsing; Pixels; Semantics; Task analysis

Recognizing the pedestrian attributes in surveillance scenes is an inherently challenging task, especially for the pedestrian images with large pose variations, complex backgrounds, and various camera viewing angles. To select important and discriminative regions or pixels against the variations, th...