CNN-based approach for visual quality improvement on HEVC

Deep learning based on Convolutional Neural Network (CNN) is a very hot issue for various recognition problems. In this paper, we propose a scheme to improve the visual quality of the video coding standard. We apply a CNN model to high efficiency video coding (HEVC) encoding system and suggest a combined coding scheme using a CNN model. Through experiment, we verify that the proposed scheme achieves up to 0.24 dB of the quality improvement on HM 16.10 reference software.