A Deep Image Compression Framework for Face Recognition

Face recognition technology has advanced rapidly and has been widely used in various applications. Due to the huge amount of data of face images in large-scale face recognition tasks and the large computing resource cost required correspondingly, there is a requirement for a face image compression approach that is highly suitable for face recognition tasks. In this paper, we propose a deep convolutional autoencoder compression network for face recognition tasks. In compression process, deep features are extracted from the original image by a Compression Network (CompNet) to produce a compact representation of the original image, which is then encoded and saved by an existing codec PNG. In reconstruction process, this compact representation is utilized by a Reconstruction Network (RecNet) to generate a restored image of the original one. In order to improve the face recognition accuracy when the compression framework is used in a face recognition system, we combine the CompNet and RecNet with an existing face recognition network for joint optimization. We test the proposed scheme and find that after joint optimization, the Labeled Faces in the Wild (LFW) dataset compressed by our compression framework has higher face verification accuracy than that compressed by JPEG2000, and is much higher than that compressed by JPEG.