Separate Loss for Basic and Compound Facial Expression Recognition in the Wild

Yingjian Li, Yao Lu, Jinxing Li, Guangming Lu
Proceedings of The Eleventh Asian Conference on Machine Learning, PMLR 101:897-911, 2019.

Abstract

In the past few years, facial expression recognition has made great progress because of the development of convolutional neural networks. However, the features learned only using the softmax loss are not discriminative enough for highly accurate facial expression recognition in the wild, especially for the compound facial expression recognition. To enhance the discriminative power of the learned features, we propose the separate loss for both basic and compound facial expression recognition in the wild in this paper. Such loss maximizes intra-class similarity while minimizing the similarity between different classes. The qualitative and quantitative analysis shows that the features learned using such loss function are characterized by intra-class compactness and inter-class separation. Experiments are performed on two databases in the wild and the proposed method achieves state-of-the-art results on both basic and compound expressions. Furthermore, another two databases are used to perform cross database experiments to show the generalization ability of our method.

Cite this Paper

BibTeX
@InProceedings{pmlr-v101-li19b, title = {Separate Loss for Basic and Compound Facial Expression Recognition in the Wild}, author = {Li, Yingjian and Lu, Yao and Li, Jinxing and Lu, Guangming}, booktitle = {Proceedings of The Eleventh Asian Conference on Machine Learning}, pages = {897--911}, year = {2019}, editor = {Lee, Wee Sun and Suzuki, Taiji}, volume = {101}, series = {Proceedings of Machine Learning Research}, month = {17--19 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v101/li19b/li19b.pdf}, url = {https://proceedings.mlr.press/v101/li19b.html}, abstract = {In the past few years, facial expression recognition has made great progress because of the development of convolutional neural networks. However, the features learned only using the softmax loss are not discriminative enough for highly accurate facial expression recognition in the wild, especially for the compound facial expression recognition. To enhance the discriminative power of the learned features, we propose the separate loss for both basic and compound facial expression recognition in the wild in this paper. Such loss maximizes intra-class similarity while minimizing the similarity between different classes. The qualitative and quantitative analysis shows that the features learned using such loss function are characterized by intra-class compactness and inter-class separation. Experiments are performed on two databases in the wild and the proposed method achieves state-of-the-art results on both basic and compound expressions. Furthermore, another two databases are used to perform cross database experiments to show the generalization ability of our method.} }
Endnote
%0 Conference Paper %T Separate Loss for Basic and Compound Facial Expression Recognition in the Wild %A Yingjian Li %A Yao Lu %A Jinxing Li %A Guangming Lu %B Proceedings of The Eleventh Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2019 %E Wee Sun Lee %E Taiji Suzuki %F pmlr-v101-li19b %I PMLR %P 897--911 %U https://proceedings.mlr.press/v101/li19b.html %V 101 %X In the past few years, facial expression recognition has made great progress because of the development of convolutional neural networks. However, the features learned only using the softmax loss are not discriminative enough for highly accurate facial expression recognition in the wild, especially for the compound facial expression recognition. To enhance the discriminative power of the learned features, we propose the separate loss for both basic and compound facial expression recognition in the wild in this paper. Such loss maximizes intra-class similarity while minimizing the similarity between different classes. The qualitative and quantitative analysis shows that the features learned using such loss function are characterized by intra-class compactness and inter-class separation. Experiments are performed on two databases in the wild and the proposed method achieves state-of-the-art results on both basic and compound expressions. Furthermore, another two databases are used to perform cross database experiments to show the generalization ability of our method.
APA
Li, Y., Lu, Y., Li, J. & Lu, G.. (2019). Separate Loss for Basic and Compound Facial Expression Recognition in the Wild. Proceedings of The Eleventh Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 101:897-911 Available from https://proceedings.mlr.press/v101/li19b.html.

Related Material