Spatio-temporal prediction in video coding by spatially refined motion compensation

TL;DR

This paper introduces a novel spatio-temporal prediction method for video coding that refines motion compensation by considering spatial and temporal correlations simultaneously, improving compression efficiency.

Contribution

The proposed method combines spatial and temporal correlations in prediction, enhancing existing motion compensation techniques without requiring extra side information.

Findings

Data rate reduction of up to 15%

PSNR increase of up to 0.75 dB

Effective integration into H.264/AVC codec

Abstract

The purpose of this contribution is to introduce a new method of signal prediction in video coding. Unlike most existent prediction methods that either use temporal or use spatial correlations to generate the prediction signal, the proposed method uses spatial and temporal correlations at the same time. The spatio-temporal prediction is obtained by first performing motion compensation for a macroblock, followed by a refinement step that pays attention to the correlations between the macroblock and its surroundings. At the decoder, the refinement step can be performed in the same manner, thus no additional side information has to be transmitted. Implementation of the spatial refinement step into the H.264/AVC video codec leads to reduction in data rate of up to nearly 15% and increase in PSNR of up to 0.75 dB, compared to pure motion compensated prediction.