EVA: Zero-shot Accurate Attributes and Multi-Object Video Editing

Understand

Current diffusion-based video editing primarily focuses on local editing (\textit{e.g.,} object/background editing) or global style editing by utilizing various dense correspondences.

• However, these methods often fail to accurately edit the foreground and background simultaneously while preserving the original layout.
• We find that the crux of the issue stems from the imprecise distribution of attention weights across designated regions, including inaccurate text-to-attribute control and attention leakage.
• To tackle this issue, we introduce EVA, a \textbf{zero-shot} and \textbf{multi-attribute} video editing framework tailored for human-centric videos with complex motions.