期刊
IEEE TRANSACTIONS ON IMAGE PROCESSING
卷 30, 期 -, 页码 6544-6556出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TIP.2021.3093397
关键词
Feature extraction; Face recognition; Image recognition; Faces; Convolution; Image reconstruction; Geometry; Facial expression recognition; deep convolutional neural networks; multi-scale; local attention
资金
- National Natural Science Foundation of China (NSFC) [61825601]
- Natural Science Foundation of Jiangsu Province (NSF-JS) [BK20192004B]
This paper proposed a global multi-scale and local attention network for facial expression recognition in the wild, achieving state-of-the-art results on several benchmarks.
Facial expression recognition (FER) in the wild received broad concerns in which occlusion and pose variation are two key issues. This paper proposed a global multi-scale and local attention network (MA-Net) for FER in the wild. Specifically, the proposed network consists of three main components: a feature pre-extractor, a multi-scale module, and a local attention module. The feature pre-extractor is utilized to pre-extract middle-level features, the multi-scale module to fuse features with different receptive fields, which reduces the susceptibility of deeper convolution towards occlusion and variant pose, while the local attention module can guide the network to focus on local salient features, which releases the interference of occlusion and non-frontal pose problems on FER in the wild. Extensive experiments demonstrate that the proposed MA-Net achieves the state-of-the-art results on several in-the-wild FER benchmarks: CAER-S, AffectNet-7, AffectNet-8, RAFDB, and SFEW with accuracies of 88.42%, 64.53%, 60.29%, 88.40%, and 59.40% respectively. The codes and training logs are publicly available at https://github.com/zengqunzhao/MA-Net.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据