Video Compression With CNN-Based Postprocessing

In recent years, video compression techniques have been significantly challenged by the rapidly increased demands associated with high quality and immersive video content. Among various compression tools, postprocessing can be applied on reconstructed video content to mitigate visible compression artefacts and to enhance the overall perceptual quality. Inspired by advances in deep learning, we propose a new convolutional neural network based postprocessing approach, which has been integrated with two state-of-the-art coding standards, versatile video coding (VVC) and AOMedia Video (AV1). The results show consistent coding gains on all tested sequences at various spatial resolutions, with average bit rate savings of 4.0% and 5.8% against original VVC and AV1, respectively (based on the assessment of peak signal-to-noise ratio). This network has also been trained with perceptually inspired loss functions, which have further improved the reconstruction quality based on perceptual quality assessment (VMAF), with average coding gains of 13.9% over VVC and 10.5% against AV1.