Mask-Guided Multi-Task Network for Face Attribute Recognition

TL;DR

This paper introduces MGMTN, a novel face attribute recognition network that uses adaptive masks to focus on key facial regions, improving accuracy over traditional global-region methods.

Contribution

The paper proposes a new multi-task network with adaptive mask learning and group-global feature fusion to enhance face attribute recognition by focusing on critical facial regions.

Findings

MGMTN outperforms existing methods on benchmark datasets.

Adaptive masks effectively localize key facial features.

Group-global feature fusion improves attribute classification accuracy.

Abstract

Face Attribute Recognition (FAR) plays a crucial role in applications such as person re-identification, face retrieval, and face editing. Conventional multi-task attribute recognition methods often process the entire feature map for feature extraction and attribute classification, which can produce redundant features due to reliance on global regions. To address these challenges, we propose a novel approach emphasizing the selection of specific feature regions for efficient feature learning. We introduce the Mask-Guided Multi-Task Network (MGMTN), which integrates Adaptive Mask Learning (AML) and Group-Global Feature Fusion (G2FF) to address the aforementioned limitations. Leveraging a pre-trained keypoint annotation model and a fully convolutional network, AML accurately localizes critical facial parts (e.g., eye and mouth groups) and generates group masks that delineate meaningful feature regions, thereby mitigating negative transfer from global region usage. Furthermore, G2FF combines group and global features to enhance FAR learning, enabling more precise attribute identification. Extensive experiments on two challenging facial attribute recognition datasets demonstrate the effectiveness of MGMTN in improving FAR performance.