arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2106.04520cs.CV
收藏

MVT:用于野外面部表情识别的面具视觉Transformer

MVT: Mask Vision Transformer for Facial Expression Recognition in the wild

University of Science and Technology of China(中国科学技术大学)

Hanting Li, Mingzhe Sui, Feng Zhao, Zhengjun Zha, Feng Wu

摘要:

野外人脸表情识别是计算机视觉中一项极具挑战性的任务,这是由于背景多变、人脸图像质量不高以及标注者的主观性等原因造成的。这些不确定性使得神经网络很难在有限规模的数据集上学习鲁棒的特征。此外,网络可以很容易地被上述因素分布,并执行错误的决策。近年来,视觉Transformer(ViT)和数据高效图像Transformers(DeiT)在传统的分类任务中表现出了显著的性能。自注意力机制使Transformers在第一层获得全局感受野,大大增强了特征提取能力。在这项工作中,我们首先提出了一种新的纯transformer为基础的面具视觉Transformer(MVT)在野外的FER,它由两个模块组成:一个transformer为基础的面具生成网络(MGN),以产生一个面具,可以过滤掉复杂的背景和遮挡的人脸图像,和一个动态的重新标记模块,纠正错误的标签在野外FER数据集。大量的实验结果表明,我们的MVT优于国家的最先进的方法在RAF-DB与88.62%,FERPlus与89.22%,和AffectNet-7与64.57%,分别,并实现了相当的结果在AffectNet-8与61.40%。

英文摘要:

Facial Expression Recognition (FER) in the wild is an extremely challenging task in computer vision due to variant backgrounds, low-quality facial images, and the subjectiveness of annotators. These uncertainties make it difficult for neural networks to learn robust features on limited-scale datasets. Moreover, the networks can be easily distributed by the above factors and perform incorrect decisions. Recently, vision transformer (ViT) and data-efficient image transformers (DeiT) present their significant performance in traditional classification tasks. The self-attention mechanism makes transformers obtain a global receptive field in the first layer which dramatically enhances the feature extraction capability. In this work, we first propose a novel pure transformer-based mask vision transformer (MVT) for FER in the wild, which consists of two modules: a transformer-based mask generation network (MGN) to generate a mask that can filter out complex backgrounds and occlusion of face images, and a dynamic relabeling module to rectify incorrect labels in FER datasets in the wild. Extensive experimental results demonstrate that our MVT outperforms state-of-the-art methods on RAF-DB with 88.62%, FERPlus with 89.22%, and AffectNet-7 with 64.57%, respectively, and achieves a comparable result on AffectNet-8 with 61.40%.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑