Edited endpoints improve video transition quality with fewer artifacts
Edited Endpoints as Partial Goals for Image-to-Video Generation
Computer Vision and Pattern Recognition
Summary
Creating videos that smoothly transition from one image to another can be tricky because sometimes the video shows glitches or unnatural movements. The paper finds that focusing only on the parts of the image that actually change, instead of the entire image, helps reduce these glitches. The authors tested this idea on several video models and showed it keeps the video smooth without affecting how well the transition finishes. This approach treats the edited image as a partial goal, leading to better video quality.
What this means in practice
- •For video editing software developers: Use partial image conditioning to reduce visual glitches when generating videos transitioning from edited images.
- •For visual effects teams: Improve smoothness of synthetic video transitions by focusing video model constraints only on edited parts of frames.
Authors
Sihyun Kim, Sanghyun Jo, Kyungsu Kim
Abstract
Edited endpoint images make the target of an image-to-video transition explicit and can improve transition completion, but also preserve context in the source image's layout. Full-frame endpoint conditioning turns this preserved context into requirements on the video's final state, potentially inducing temporal artifacts. We measure discontinuity with Hard-Cut (HC), late background reversal with Background Motion Reversal (BMR), and motion suppression with Dynamic Degree (DD). These artifacts appear across five video models spanning training-free and learned endpoint conditioning, in model-specific forms. We study Partial, a selective conditioning scheme that constrains only an edit-relevant region estimated from image-editor attention and leaves the remaining terminal content to the video model. Across three backbones on TC-Bench-I2V, Partial reduces temporal artifacts relative to full-frame conditioning, with no significant difference in Transition Completion Ratio (TCR); on Wan2.2-Turbo, uniformly weakening conditioning does not reproduce these reductions. A donor-context intervention keeps full-frame conditioning and replaces only the context outside the edit-relevant region with the final frame of a Partial video. It lowers HC and BMR on HunyuanVideo-1.5 and lowers BMR while restoring DD on Wan2.2-Turbo, with no significant difference in TCR, so constraint extent alone does not account for these artifacts. Together, these findings suggest that edited endpoints should serve as partial future goals, although edit relevance does not guarantee future validity. Code and data will be released at https://github.com/wannagetacat/Edited-Endpoints-I2V.