Combining Convolutional Neural Network and Support Vector Machine for Gait-based Gender Recognition

Recently, deep learning based on convolutional neural networks (CNN) has achieved great state-of-the-art performance in many fields such as image classification, semantic analysis and biometric recognition. Normally, the Softmax activation function is used as classifier in the last layer of CNN. However, there some studies try to replace the Softmax layer with the support vector machine (SVM) in an artificial neural network architecture and achieve great results. Inspired by these works, we research the performance of CNN with linear SVM classifier on the gender recognition based on CASIA-B dataset. In the first model, the input image's descriptors are extracted from the fully connected layer of the pre-trained VGGNet-16 model as the features to train the SVM. In the second model, we adjust VGGNet-16 with a hinge loss function using an L2 norm to create a new architecture, namely VGGNet-SVM. The results have shown that SVM shows the better performance than Softmax in VGGNet-16 to work out the gender recognition problem based on gait.