I2VEdit: First-Frame-Guided Video Editing via Image-to-Video Diffusion Models

TL;DR

I2VEdit leverages image-to-video diffusion models to enable high-quality, temporally consistent video editing guided by a single frame, bridging the gap between image and video editing capabilities.

Contribution

We introduce I2VEdit, a novel method that propagates edits from a single image to entire videos using a pre-trained model, effectively handling various edit types with high quality.

Findings

Outperforms existing methods in fine-grained video editing

Produces high-quality, temporally consistent edited videos

Handles global, local, and shape edits effectively

Abstract

The remarkable generative capabilities of diffusion models have motivated extensive research in both image and video editing. Compared to video editing which faces additional challenges in the time dimension, image editing has witnessed the development of more diverse, high-quality approaches and more capable software like Photoshop. In light of this gap, we introduce a novel and generic solution that extends the applicability of image editing tools to videos by propagating edits from a single frame to the entire video using a pre-trained image-to-video model. Our method, dubbed I2VEdit, adaptively preserves the visual and motion integrity of the source video depending on the extent of the edits, effectively handling global edits, local edits, and moderate shape changes, which existing methods cannot fully achieve. At the core of our method are two main processes: Coarse Motion Extraction to align basic motion patterns with the original video, and Appearance Refinement for precise adjustments using fine-grained attention matching. We also incorporate a skip-interval strategy to mitigate quality degradation from auto-regressive generation across multiple video clips. Experimental results demonstrate our framework's superior performance in fine-grained video editing, proving its capability to produce high-quality, temporally consistent outputs.