D$^2$-VR: Degradation-Robust and Distilled Video Restoration with Synergistic Optimization Strategy
This addresses practical deployment challenges in video restoration for applications requiring efficient and stable processing, though it is incremental as it builds on existing diffusion and temporal alignment paradigms.
The paper tackled the problem of high inference latency and temporal instability in diffusion-based video restoration under complex real-world degradations, resulting in a method that achieves state-of-the-art performance while accelerating sampling by 12×.
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$\times$}