Improved softmax loss for deep learning-based face and expression recognition

Jiancan Zhou, Xi Jia, Linlin Shen, Zhenkun Wen, Zhong Ming

Research output: Journal PublicationArticlepeer-review

14 Citations (Scopus)

Abstract

In recent years, deep convolutional neural networks (CNN) have been widely used in computer vision and significantly improved the performance of image recognition tasks. Most works use softmax loss to supervise the training of CNN and then adopt the output of last layer as features. However, the discriminative capability of the softmax loss is limited. Here, the authors analyse and improve the softmax loss by manipulating the cosine value and input feature length. As the approach does not change the principle of the softmax loss, the network can easily be optimised by typical stochastic gradient descent. The MNIST handwritten digits dataset is employed to visualise the features learned by the improved softmax loss. The CASIA-WebFace and FER2013 training set are adopted to train deep CNN for face and expression recognition, respectively. Results on both the LFW dataset and FER2013 test set show that the proposed softmax loss can learn more discriminative features and achieve better performance.

Original languageEnglish
Pages (from-to)97-102
Number of pages6
JournalCognitive Computation and Systems
Volume1
Issue number4
DOIs
Publication statusPublished - Dec 2019
Externally publishedYes

ASJC Scopus subject areas

  • Experimental and Cognitive Psychology
  • Computer Vision and Pattern Recognition
  • Computer Science Applications
  • Cognitive Neuroscience
  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Improved softmax loss for deep learning-based face and expression recognition'. Together they form a unique fingerprint.

Cite this