Abstract
This article proposes an adversarial reconstruction convolution neural network (ARCNN) for non-uniform illumination frontal face image recovery and recognition. The proposed ARCNN includes a reconstruction network and a discriminative network. The authors employ GAN framework to learn the reconstruction network in an adversarial manner. This article integrates gradient loss and perceptual loss terms, which are able to preserve the detailed and spatial structure image information, into the overall reconstruction loss function to constraint the reconstruction procedure. Experiments are conducted on the typical illumination-sensitive dataset, extended YaleB dataset. The reconstructed results demonstrate that the proposed ARCNN approach can remove the illumination and shadow information and recover natural uniform illuminated face image from non-uniform illuminated ones. Face recognition results on the extended YaleB dataset demonstrate that the proposed ARCNN reconstruction procedure can also preserve the discriminative information of face image for classification task.
Highlights
High quality images are important for improving the performance of image analysis systems and the visualization of human beings
The experimental results both for face image recovery and face recognition on Extended YaleB dataset show that the proposed adversarial reconstruction convolution neural network (ARCNN) is efficient for removing non-uniform illumination information and preserving individual classification information, which is useful for face image visualization and classification tasks
The proposed ARCNN employs the structure of generative adversarial networks (GANs), such that it contains a reconstruction network and a discriminative network that are simultaneously trained using the min-max formulation of GAN shown in Eq 1
Summary
High quality images are important for improving the performance of image analysis systems and the visualization of human beings. The authors propose an adversarial reconstruction convolution neural networks(ARCNN) for non-uniform illumination frontal face image recovery. The authors employ the GAN framework for training the reconstruction network and the discriminative system in an adversarial manner to prompt the quality of reconstructed images The experimental results both for face image recovery and face recognition on Extended YaleB dataset show that the proposed ARCNN is efficient for removing non-uniform illumination information and preserving individual classification information, which is useful for face image visualization and classification tasks. The paper is organized as follows: Section 2 introduces the proposed adversarial reconstruction convolution neural networks, and section 3 provides experimental results on two non-uniform illumination face image datasets.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: International Journal of Cognitive Informatics and Natural Intelligence
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
