Since the Generative Adversarial Networks (GANs) was proposed, researches on image generation attract many scholars' general attention and good graces. Traditional GANs generate a sample by playing a minimax game ...
Since the Generative Adversarial Networks (GANs) was proposed, researches on image generation attract many scholars' general attention and good graces. Traditional GANs generate a sample by playing a minimax game between generator and discriminator. In this paper, we propose a new method called EmotionGAN for generating facial expression. Specifically, the inverse of the generator is firstly utilized to establish the mapping between the input and feature vector. Then the Generalized Linear Model (GLM) is used to fit the changing direction of different expressions in the feature space, which provide a linear guidance to the feature vector along the expression axis, and thus spatial distribution consistence with the target feature vector is assured. Finally the generator is applied to reconstruct the facial image of the expression. By controlling the intensity of the feature vector, the generated image can be smoothly changed on a specific expression. Experiments have shown that EmotionGAN can quickly generate face images with arbitrary expressions while ensuring identity information is not changed, and the image attributes are more accurate and the resolution is higher.
This paper proposes a simple and efficient regularization method, called PickPatch, for face recognition based on a deep convolutional neural network (DCNN). The proposed method randomly selects patches in the input f...
This paper proposes a simple and efficient regularization method, called PickPatch, for face recognition based on a deep convolutional neural network (DCNN). The proposed method randomly selects patches in the input face image and the intermediate feature maps as the activation region according to facial landmarks during the training phase. PickPatch is an approximation method that trains a series of models for different face patches and provides a combined model. This strategy introduces the idea of model combination for multiple face patches but does not change the model structure, which is both simple and efficient. Experiments on the public LFW database demonstrate that the proposed regularization method based on current deep convolutional neural networks can achieve obvious improvements of face recognition accuracy.
暂无评论