VideoDiff: Human-AI Video Co-Creation with Alternatives
arXiv:2502.10190 · doi:10.1145/3706598.3713417
Abstract
To make an engaging video, people sequence interesting moments and add visuals such as B-rolls or text. While video editing requires time and effort, AI has recently shown strong potential to make editing easier through suggestions and automation. A key strength of generative models is their ability to quickly generate multiple variations, but when provided with many alternatives, creators struggle to compare them to find the best fit. We propose VideoDiff, an AI video editing tool designed for editing with alternatives. With VideoDiff, creators can generate and review multiple AI recommendations for each editing process: creating a rough cut, inserting B-rolls, and adding text effects. VideoDiff simplifies comparisons by aligning videos and highlighting differences through timelines, transcripts, and video previews. Creators have the flexibility to regenerate and refine AI suggestions as they compare alternatives. Our study participants (N=12) could easily compare and customize alternatives, creating more satisfying results.
Accepted to CHI 2025
References in corpus (8)
- The Effects of Generative AI on Design Fixation and Divergent Thinking
- The Impact of Multiple Parallel Phrase Suggestions on Email Input and Composition Behaviour of Native and Non-Native English Writers
- Scout: Rapid Exploration of Interface Layout Alternatives through High-Level Design Constraints
- AVscript: Accessible Video Editing with Audio-Visual Scripts
- B-Script: Transcript-based B-roll Video Editing with Recommendations
- ExpressEdit: Video Editing with Natural Language and Sketching
- Wakey-Wakey: Animate Text by Mimicking Characters in a GIF
- PEANUT: A Human-AI Collaborative Tool for Annotating Audio-Visual Data