Convolutional Neural Networks for Video Intra Prediction Using Cross-Component Adaptation

Maria Meyer, Jonathan Wiesner, Jens Schneider, Christian Rohlfing

To be presented at the 44rd IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2019), 12-17 May 2019, Brighton, UK
Conference homepage

Abstract

It has recently been shown that neural networks can improve video and image intra prediction. In this paper the properties of different architectures for neural network-based intra prediction are evaluated. This includes an analysis of the properties of convolutional neural networks used for this purpose, showing that they outperform fully connected ones especially for complex and low resolution content. Also, the usage of separate networks for luma and chroma prediction which are able to perform a learned cross-component prediction is proposed as this is clearly beneficial for the prediction quality. Furthermore, a new way to integrate and signal a neural network-based intra prediction mode is investigated. Combined this improves the compression performance in terms of average BD-rate changes compared to HEVC by -2.0 % for the luma and by -1.5 % for the chroma channels.

Supplementary Material

Datasets and Parameters

NOTICE FOR IEEE PUBLICATIONS: © IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works, must be obtained from the IEEE. Contact: Manager, Copyrights and Permissions / IEEE Service Center / 445 Hoes Lane / P.O. Box 1331 / Piscataway, NJ 08855-1331, USA. Telephone: + Intl. 908-562-3966.