ETRI-Knowledge Sharing Plaform

KOREAN
논문 검색
Type SCI
Year ~ Keyword

Detail

Journal Article Toward Unbiased Facial Expression Recognition in the Wild via Cross-Dataset Adaptation
Cited 13 time in scopus Download 204 time Share share facebook twitter linkedin kakaostory
Authors
Byungok Han, Woo-Han Yun, Jang-Hee Yoo, Won Hwa Kim
Issue Date
2020-08
Citation
IEEE Access, v.8, pp.159172-159181
ISSN
2169-3536
Publisher
IEEE
Language
English
Type
Journal Article
DOI
https://dx.doi.org/10.1109/ACCESS.2020.3018738
Abstract
Despite various success in computer vision with facial images (e.g., face detection, recognition, and generation), facial expression recognition is still a challenging problem yet to be solved. This is because of simple but fundamental bottlenecks: 1) no global agreement on different facial expressions, 2) significant dataset biases that prevent cross-dataset analysis for a large-scale study, and 3) high class imbalance in in-the-wild datasets that causes inconsistency in predicting expressions in images using a machine learning algorithm. To tackle these issues, we propose a novel Deep Learning approach via adaptive cross-dataset scheme. We combine multiple in-the-wild datasets to secure sufficient training samples while minimizing dataset bias using ideas of reversal gradients to retain generality. For this, we introduce a flexible objective function that can control for skewed label distributions in the dataset. Incorporating these ideas, together with the ResNet pipeline as a backbone, we carried extensive experiments to validate our ideas using three independent in-the-wild facial expression datasets, which first confirmed bias from different datasets and yielded improved performance on facial expression recognition using the multi-site dataset.
KSP Keywords
Computer Vision(CV), Dataset bias, Facial expression recognition(FER), Facial image, Global agreement, Improved performance, In-the-wild, Label distributions, Large-scale study, Learning approach, Machine Learning Algorithms
This work is distributed under the term of Creative Commons License (CCL)
(CC BY)
CC BY