4.7 Article

A Transfer Learning Approach to Heatmap Regression for Action Unit Intensity Estimation

期刊

IEEE TRANSACTIONS ON AFFECTIVE COMPUTING
卷 14, 期 1, 页码 436-450

出版社

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TAFFC.2021.3061605

关键词

Gold; Task analysis; Heating systems; Correlation; Transfer learning; Estimation; Heat transfer; Facial action unit intensity estimation; heatmap regression; transfer learning

向作者/读者索取更多资源

In this study, a novel approach based on Heatmap Regression is proposed to jointly estimate the localization and intensity of Action Units (AUs). The method utilizes variable size heatmaps to model AUs intensity and employs transfer learning from a network trained on facial landmark datasets. Experimental results demonstrate that the system achieves state-of-the-art performance on three popular datasets, BP4D, DISFA, and FERA2017.
Action Units (AUs) are geometrically-based atomic facial muscle movements known to produce appearance changes at specific facial locations. Motivated by this observation we propose a novel AU modelling problem that consists of jointly estimating their localisation and intensity. To this end, we propose a simple yet efficient approach based on Heatmap Regression that merges both problems into a single task. A Heatmap models whether an AU occurs or not at a given spatial location. To accommodate the joint modelling of AUs intensity, we propose variable size heatmaps, with their amplitude and size varying according to the labelled intensity. Using Heatmap Regression, we can inherit from the progress recently witnessed in facial landmark localisation. Building upon the similarities between both problems, we devise a transfer learning approach where we exploit the knowledge of a network trained on large-scale facial landmark datasets. In particular, we explore different alternatives for transfer learning through a) fine-tuning, b) adaptation layers, c) attention maps, and d) reparametrisation. Our approach effectively inherits the rich facial features produced by a strong face alignment network, with minimal extra computational cost. We empirically validate that our system sets a new state-of-the-art on three popular datasets, namely BP4D, DISFA, and FERA2017.

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。

评论

主要评分

4.7
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据