D$^2$-VR: Degradation-Robust and Distilled Video Restoration with Synergistic Optimization Strategy

TL;DR

D$^2$-VR is a novel video restoration framework that combines diffusion priors with a robust alignment and distillation strategy, achieving high-quality results with significantly reduced inference time and improved temporal stability.

Contribution

It introduces a degradation-robust flow alignment module and an adversarial distillation approach to enable fast, stable, and high-quality video restoration with diffusion models.

Findings

Achieves state-of-the-art performance in video restoration.

Accelerates diffusion sampling by 12 times.

Enhances temporal stability under complex degradations.

Abstract

The integration of diffusion priors with temporal alignment has emerged as a transformative paradigm for video restoration, delivering fantastic perceptual quality, yet the practical deployment of such frameworks is severely constrained by prohibitive inference latency and temporal instability when confronted with complex real-world degradations. To address these limitations, we propose \textbf{D$^2$-VR}, a single-image diffusion-based video-restoration framework with low-step inference. To obtain precise temporal guidance under severe degradation, we first design a Degradation-Robust Flow Alignment (DRFA) module that leverages confidence-aware attention to filter unreliable motion cues. We then incorporate an adversarial distillation paradigm to compress the diffusion sampling trajectory into a rapid few-step regime. Finally, a synergistic optimization strategy is devised to harmonize perceptual quality with rigorous temporal consistency. Extensive experiments demonstrate that D$^2$-VR achieves state-of-the-art performance while accelerating the sampling process by \textbf{12$\times$}