I-Center Loss for Deep Neural Networks

Abstract

Convolutional Neural Network (CNN) have been widely used for image classification and computer vision tasks such as face recognition, target detection. Softmax loss is one of the most commonly used components to train CNN, which only penalizes the classification loss. So we consider how to train intra-class compactness and inter-class separability better. In this paper, we proposed an I-Center Loss to make inter-class having a better separability, which means that I-Center Loss penalizes the difference between each center of classes. With I-Center Loss, we trained a robust CNN to achieve a better performance. Extensive experiments on MNIST, CIFAR10, LFW (face datasets for face recognition) demonstrate the effectiveness of the I-center loss. We have tried different models, visualized the experimental results and showed the effectiveness of our proposed I-center loss.