Highly Efficient Non-Separable Transforms for Next Generation Video Coding

TL;DR

This paper introduces a new class of low-complexity, parallelizable transforms called HyGTs for video coding, achieving significant compression gains with reduced computational and memory requirements.

Contribution

It proposes a parametric approach to design HyGT transforms that outperform traditional matrix-based methods in efficiency and complexity.

Findings

HyGTs improve average coding gain by 6% bit rate reduction.

HyGTs use 6.8 times less memory than KLT matrices.

Transform implementation is highly parallelizable.

Abstract

For the last few decades, the application of signal-adaptive transform coding to video compression has been stymied by the large computational complexity of matrix-based solutions. In this paper, we propose a novel parametric approach to greatly reduce the complexity without degrading the compression performance. In our approach, instead of following the conventional technique of identifying full transform matrices that yield best compression efficiency, we look for the best transform parameters defining a new class of transforms, called HyGTs, which have low complexity implementations that are easy to parallelize. The proposed HyGTs are implemented as an extension of High Efficiency Video Coding (HEVC), and our comprehensive experimental results demonstrate that proposed HyGTs improve average coding gain by 6% bit rate reduction, while using 6.8 times less memory than KLT matrices.