Pedestrian Attribute Recognition as Label-balanced Multi-label Learning

TL;DR

This paper introduces a novel multi-label learning framework for pedestrian attribute recognition that addresses label and semantics imbalance, improving accuracy with minimal computational cost.

Contribution

The paper proposes a new framework that decouples label-balanced data re-sampling from attribute co-occurrence, and introduces Bayesian feature augmentation to enhance attribute diversity.

Findings

Achieves state-of-the-art accuracy on multiple benchmarks.

Effectively mitigates label and semantics imbalance issues.

Operates with minimal computational resources.

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 attributes co-occurrence, i.e., we equalize the sampling prior of an attribute while not biasing that of the co-occurred others. To diversify the attributes semantics and mitigate the feature noise, we propose a Bayesian feature augmentation method to introduce true in-distribution novelty. Handling both imbalances jointly, our work achieves best accuracy on various popular benchmarks, and importantly, with minimal computational budget.