Distraction-Aware Feature Learning for Human Attribute Recognition via Coarse-to-Fine Attention Mechanism

TL;DR

This paper introduces Da-HAR, a distraction-aware deep learning method for human attribute recognition that employs a coarse-to-fine attention mechanism to improve attribute localization amidst occlusions and complex backgrounds.

Contribution

It proposes a novel coarse-to-fine attention mechanism that enhances attribute localization in HAR, addressing challenges of distractions and occlusions.

Findings

Achieves state-of-the-art results on WIDER-Attribute and RAP datasets.

Effectively reduces distractions and irrelevant regions during feature learning.

Improves accuracy of human attribute recognition in complex scenes.

Abstract

Recently, Human Attribute Recognition (HAR) has become a hot topic due to its scientific challenges and application potentials, where localizing attributes is a crucial stage but not well handled. In this paper, we propose a novel deep learning approach to HAR, namely Distraction-aware HAR (Da-HAR). It enhances deep CNN feature learning by improving attribute localization through a coarse-to-fine attention mechanism. At the coarse step, a self-mask block is built to roughly discriminate and reduce distractions, while at the fine step, a masked attention branch is applied to further eliminate irrelevant regions. Thanks to this mechanism, feature learning is more accurate, especially when heavy occlusions and complex backgrounds exist. Extensive experiments are conducted on the WIDER-Attribute and RAP databases, and state-of-the-art results are achieved, demonstrating the effectiveness of the proposed approach.