Deep Multi-task Multi-label CNN for Effective Facial Attribute Classification

TL;DR

This paper introduces DMM-CNN, a multi-task multi-label CNN that jointly learns facial landmark detection and attribute classification, using dynamic weighting and adaptive thresholding to improve accuracy on challenging datasets.

Contribution

The paper presents a novel deep multi-task multi-label CNN that jointly optimizes facial landmark detection and attribute classification, with dynamic loss weighting and adaptive thresholding for better performance.

Findings

Outperforms state-of-the-art methods on CelebA and LFWA datasets.

Effectively handles class imbalance in multi-label learning.

Improves facial attribute classification accuracy through joint multi-task learning.

Abstract

Facial Attribute Classification (FAC) has attracted increasing attention in computer vision and pattern recognition. However, state-of-the-art FAC methods perform face detection/alignment and FAC independently. The inherent dependencies between these tasks are not fully exploited. In addition, most methods predict all facial attributes using the same CNN network architecture, which ignores the different learning complexities of facial attributes. To address the above problems, we propose a novel deep multi-task multi-label CNN, termed DMM-CNN, for effective FAC. Specifically, DMM-CNN jointly optimizes two closely-related tasks (i.e., facial landmark detection and FAC) to improve the performance of FAC by taking advantage of multi-task learning. To deal with the diverse learning complexities of facial attributes, we divide the attributes into two groups: objective attributes and subjective attributes. Two different network architectures are respectively designed to extract features for two groups of attributes, and a novel dynamic weighting scheme is proposed to automatically assign the loss weight to each facial attribute during training. Furthermore, an adaptive thresholding strategy is developed to effectively alleviate the problem of class imbalance for multi-label learning. Experimental results on the challenging CelebA and LFWA datasets show the superiority of the proposed DMM-CNN method compared with several state-of-the-art FAC methods.