RFDM: Residual Flow Diffusion Model for Efficient Causal Video Editing

TL;DR

RFDM introduces a novel residual flow diffusion approach for efficient, variable-length causal video editing that leverages temporal redundancy and outperforms existing methods in style transfer and object removal tasks.

Contribution

The paper presents RFDM, a new diffusion-based video editing model that predicts residuals between frames, enabling efficient, scalable editing of variable-length videos with improved performance.

Findings

RFDM surpasses I2I-based methods in quality.

RFDM competes with fully spatiotemporal models.

RFDM maintains efficiency regardless of input video length.

Abstract

Instructional video editing applies edits to an input video using only text prompts, enabling intuitive natural-language control. Despite rapid progress, most methods still require fixed-length inputs and substantial compute. Meanwhile, autoregressive video generation enables efficient variable-length synthesis, yet remains under-explored for video editing. We introduce a causal, efficient video editing model that edits variable-length videos frame by frame. For efficiency, we start from a 2D image-to-image (I2I) diffusion model and adapt it to video-to-video (V2V) editing by conditioning the edit at time step t on the model's prediction at t-1. To leverage videos' temporal redundancy, we propose a new I2I diffusion forward process formulation that encourages the model to predict the residual between the target output and the previous prediction. We call this Residual Flow Diffusion Model (RFDM), which focuses the denoising process on changes between consecutive frames. Moreover, we propose a new benchmark that better ranks state-of-the-art methods for editing tasks. Trained on paired video data for global/local style transfer and object removal, RFDM surpasses I2I-based methods and competes with fully spatiotemporal (3D) V2V models, while matching the compute of image models and scaling independently of input video length. More content can be found in: https://smsd75.github.io/RFDM_page/