Region-Adaptive Video Sharpening via Rate-Perception Optimization

2025-08-12

AI Summary

Conventional video sharpening applies uniform intensity across frames, ignoring texture heterogeneity—leading to degraded perceptual quality and increased bitrates. To address this, we propose an end-to-end region-adaptive sharpening model that, for the first time, incorporates Coding Tree Unit (CTU)-level partition masks as structural priors to jointly optimize sharpening gain and bitrate allocation in texture-sensitive regions. Our method integrates a rate-perception-aware joint loss function with a CTU-level adaptive gain prediction module, enabling spatially varying sharpening intensity and bitrate control. Evaluated on multiple benchmark datasets, the model significantly outperforms baselines in PSNR, LPIPS, and subjective quality scores. Crucially, it achieves average bitrate reductions of 3.2%–7.8% while preserving or even enhancing perceptual fidelity—demonstrating both effectiveness and practicality for video coding enhancement.

Technology Category

• Computer Vision: Video Understanding & Activity Analysis
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Learning on the Edge & Model Compression

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Security and Privacy: Privacy-enhancing technologies
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Sharpening is a widely adopted video enhancement technique. However, uniform sharpening intensity ignores texture variations, degrading video quality. Sharpening also increases bitrate, and there's a lack of techniques to optimally allocate these additional bits across diverse regions. Thus, this paper proposes RPO-AdaSharp, an end-to-end region-adaptive video sharpening model for both perceptual enhancement and bitrate savings. We use the coding tree unit (CTU) partition mask as prior information to guide and constrain the allocation of increased bits. Experiments on benchmarks demonstrate the effectiveness of the proposed model qualitatively and quantitatively.

Problem

• Optimize video sharpening for varying texture regions
• Balance sharpening-enhanced bitrate across diverse video areas
• Integrate CTU partition masks for bit allocation guidance

Innovation

• Region-adaptive sharpening via CTU partition masks
• Rate-perception optimization for bitrate savings
• End-to-end model for perceptual enhancement