Coherent Video Inpainting Using Optical Flow-Guided Efficient Diffusion

TL;DR

This paper introduces FloED, a novel optical flow-guided diffusion framework for video inpainting that enhances temporal coherence and computational efficiency through dual-branch architecture and flow-based acceleration techniques.

Contribution

The paper proposes a new dual-branch diffusion model with flow guidance and a training-free interpolation method to improve video inpainting quality and speed.

Findings

FloED outperforms existing methods in quality and efficiency.

The flow attention cache reduces computational costs.

The approach effectively restores backgrounds and removes objects.

Abstract

The text-guided video inpainting technique has significantly improved the performance of content generation applications. A recent family for these improvements uses diffusion models, which have become essential for achieving high-quality video inpainting results, yet they still face performance bottlenecks in temporal consistency and computational efficiency. This motivates us to propose a new video inpainting framework using optical Flow-guided Efficient Diffusion (FloED) for higher video coherence. Specifically, FloED employs a dual-branch architecture, where the time-agnostic flow branch restores corrupted flow first, and the multi-scale flow adapters provide motion guidance to the main inpainting branch. Besides, a training-free latent interpolation method is proposed to accelerate the multi-step denoising process using flow warping. With the flow attention cache mechanism, FLoED efficiently reduces the computational cost of incorporating optical flow. Extensive experiments on background restoration and object removal tasks show that FloED outperforms state-of-the-art diffusion-based methods in both quality and efficiency. Our codes and models will be made publicly available.