UniFlowRestore: A General Video Restoration Framework via Flow Matching and Prompt Guidance
This addresses the limitation of traditional single-task methods for video restoration, offering improved generalization and efficiency for real-world applications with diverse degradation types.
The paper tackles the problem of video restoration from complex degradations like blur and noise by proposing UniFlowRestore, a general framework that models restoration as a time-continuous evolution, achieving state-of-the-art performance with a PSNR of 33.89 dB and SSIM of 0.97 on video denoising.
Video imaging is often affected by complex degradations such as blur, noise, and compression artifacts. Traditional restoration methods follow a "single-task single-model" paradigm, resulting in poor generalization and high computational cost, limiting their applicability in real-world scenarios with diverse degradation types. We propose UniFlowRestore, a general video restoration framework that models restoration as a time-continuous evolution under a prompt-guided and physics-informed vector field. A physics-aware backbone PhysicsUNet encodes degradation priors as potential energy, while PromptGenerator produces task-relevant prompts as momentum. These components define a Hamiltonian system whose vector field integrates inertial dynamics, decaying physical gradients, and prompt-based guidance. The system is optimized via a fixed-step ODE solver to achieve efficient and unified restoration across tasks. Experiments show that UniFlowRestore delivers stateof-the-art performance with strong generalization and efficiency. Quantitative results demonstrate that UniFlowRestore achieves state-of-the-art performance, attaining the highest PSNR (33.89 dB) and SSIM (0.97) on the video denoising task, while maintaining top or second-best scores across all evaluated tasks.