Are Image-to-Video Models Good Zero-Shot Image Editors?

TL;DR

IF-Edit is a tuning-free framework that adapts pretrained image-to-video diffusion models for zero-shot image editing, effectively handling prompt misalignment, redundant latents, and blurry frames to enable reasoning-centric and general edits.

Contribution

The paper introduces IF-Edit, a novel method that repurposes image-to-video diffusion models for zero-shot image editing with enhanced reasoning and coherence without additional training.

Findings

Strong performance on reasoning-centric tasks

Competitive results on general-purpose edits

Effective handling of prompt and temporal challenges

Abstract

Large-scale video diffusion models show strong world simulation and temporal reasoning abilities, but their use as zero-shot image editors remains underexplored. We introduce IF-Edit, a tuning-free framework that repurposes pretrained image-to-video diffusion models for instruction-driven image editing. IF-Edit addresses three key challenges: prompt misalignment, redundant temporal latents, and blurry late-stage frames. It includes (1) a chain-of-thought prompt enhancement module that transforms static editing instructions into temporally grounded reasoning prompts; (2) a temporal latent dropout strategy that compresses frame latents after the expert-switch point, accelerating denoising while preserving semantic and temporal coherence; and (3) a self-consistent post-refinement step that sharpens late-stage frames using a short still-video trajectory. Experiments on four public benchmarks, covering non-rigid editing, physical and temporal reasoning, and general instruction edits, show that IF-Edit performs strongly on reasoning-centric tasks while remaining competitive on general-purpose edits. Our study provides a systematic view of video diffusion models as image editors and highlights a simple recipe for unified video-image generative reasoning.