Abstract
Existing zero-shot video editing methods rely on pre-trained diffusion models, successfully achieving spatial control and basic temporal consistency but fundamentally fail to preserve the video’s original temporal structure. This distinction is critical: temporal consistency ensures visual smoothness, but temporal structure dictates the video’s high-level narrative, rhythm, and semantic flow. Without this preservation, the edited output, especially for long videos with complex semantic variations, becomes narratively incoherent and semantically ambiguous. To address this limitation, we introduce a novel zero-shot editing approach that, for the first time, explicitly focuses on preserving the source video’s temporal structure. We achieve this by adaptively partitioning the video into semantically distinct clips based on feature similarity and selecting a representative anchor frame for each clip. To enhance both intra-clip fidelity and computational efficiency, we design a clip-adaptive token merging strategy which leverages the anchor’s semantic dominance to stabilize the editing. Furthermore, we employ an alternating combination strategy that ensures seamless inter-clip transitions while maintaining semantic distinction. Extensive experiments demonstrate that our method achieves state-of-the-art results, successfully balancing the preservation of original temporal structure with computational efficiency, and setting a new benchmark for zero-shot video editing fidelity.
•Temporally-structured diffusion: Preserves semantic dynamics beyond mere consistency.•Lightweight token merging: O(N) complexity via spatiotemporal windows for intra-clip stability.•Alternating cross-clip fusion: Balances coherence and uniqueness without extra cost.•Training-free integration: Works with attention control/ControlNet based image editing approaches for diverse video editing tasks.