Pedestrian Attribute Recognition as Label-balanced Multi-label Learning

arXiv:2405.04858v1 Announce Type: new
Abstract: Rooting in the scarcity of most attributes, realistic pedestrian attribute datasets exhibit unduly skewed data distribution, from which two types of model failures are delivered: (1) label imbalance: model predictions lean greatly towards the side of majority labels; (2) semantics imbalance: model is easily overfitted on the under-represented attributes due to their insufficient semantic diversity. To render perfect label balancing, we propose a novel framework that successfully decouples label-balanced data re-sampling from the curse of …

abstract arxiv cs.cv data datasets distribution labels lean pedestrian predictions recognition semantics type types