Deformable Wiener Filter for Future Video Coding

2026-05-31

AI Summary

This work addresses the limitations of existing in-loop filters in video coding, which predominantly rely on local similarity and fail to fully exploit non-local correlations due to insufficient supervision. To overcome this, the paper proposes a Deformable Wiener Filter (DWF), the first supervised learning-based in-loop filtering approach that effectively integrates both local and non-local information. Grounded in Wiener filtering theory, DWF adaptively groups pixels according to their noise and content characteristics, selects and fuses reference samples accordingly, and incorporates outlier constraints to optimize filter coefficients. Evaluated on VTM-11.0, the method achieves significant compression gains, yielding average bitrate savings of 1.16%, 1.92%, and 2.67% under the All Intra, Random Access, and Low-Delay B configurations, respectively.

Abstract

In-loop filters have attracted increasing attention due to the remarkable noise-reduction capability in the hybrid video coding framework. However, the existing in-loop filters in Versatile Video Coding (VVC) mainly take advantage of the image local similarity. Although some non-local based in-loop filters can make up for this shortcoming, the widely-used unsupervised parameter estimation method by non-local filters limits the performance. In view of this, we propose a deformable Wiener Filter (DWF). It combines the local and non-local characteristics and supervisedly trains the filter coefficients based on the Wiener Filter theory. In the filtering process, local adjacent samples and non-local similar samples are first derived for each sample of interest. Then the to-be-filtered samples are classified into specific groups based on the patch level noise and sample-level characteristics. Samples in each group share the same filter coefficients. After that, the local and non-local reference samples are adaptively fused based on the classification results. Finally, the filtering operation with outlier data constraints is conducted for each to-be-filtered sample. Moreover, the performance of the proposed DWF is analyzed with different reference sample derivation schemes in detail. Simulation results show that the proposed approach achieves 1.16%, 1.92%, and 2.67% bit-rate savings on average compared to the VTM-11.0 for All Intra, Random Access, and Low-Delay B configurations, respectively.

Problem

• in-loop filter
• video coding
• non-local similarity
• noise reduction
• parameter estimation

Innovation

• Deformable Wiener Filter
• in-loop filtering
• supervised coefficient training
• local-nonlocal fusion
• video coding