AnchorSync: Global Consistency Optimization for Long Video Editing
This addresses the challenge of editing minute-long videos for video editors or AI researchers, offering a novel method to reduce structural drift and artifacts, though it appears incremental as it builds on diffusion-based approaches.
The paper tackled the problem of maintaining global consistency and temporal coherence in long video editing, introducing AnchorSync, a diffusion-based framework that decouples editing into anchor frame editing and interpolation, resulting in coherent, high-fidelity edits that surpass prior methods in visual quality and temporal stability.
Editing long videos remains a challenging task due to the need for maintaining both global consistency and temporal coherence across thousands of frames. Existing methods often suffer from structural drift or temporal artifacts, particularly in minute-long sequences. We introduce AnchorSync, a novel diffusion-based framework that enables high-quality, long-term video editing by decoupling the task into sparse anchor frame editing and smooth intermediate frame interpolation. Our approach enforces structural consistency through a progressive denoising process and preserves temporal dynamics via multimodal guidance. Extensive experiments show that AnchorSync produces coherent, high-fidelity edits, surpassing prior methods in visual quality and temporal stability.