MoviEdit improves editing for long multi-scene videos with coherence
MoviEdit: Long-form Multi-Shot Video Editing Made Easy
Computer Vision and Pattern RecognitionArtificial IntelligenceMachine Learning
Summary
Editing videos with many different scenes and viewpoints is hard because the changes can make the edits look disconnected or inconsistent. The authors created MoviEdit, a new tool that helps keep edits smooth and consistent across these changes by learning from a large collection of edited videos made without manual work. MoviEdit makes a rough, coherent edit first and then carefully improves it over the whole video, working well for long and complex clips. The authors also made a benchmark to fairly test this kind of editing and showed that MoviEdit matches or beats other top tools.
What this means in practice
- •For video production teams: Enable editing of long, multi-scene videos with consistent style and content across abrupt scene changes using automated scalable methods.
- •For advertising agencies: Create cohesive multi-shot video ads from different shots with minimal manual coordination by leveraging automated editing drafts and refinements.$Commercial implications: Offers a scalable video editing product for agencies to produce polished multi-shot ads efficiently using automated tools.
Authors
Suhyeon Lee, Zhengang Li, Soo Ye Kim, Peihao Zhu, Jong Chul Ye, Chun-Hao Paul Huang, Duygu Ceylan, Yan Kang, Hyeonho Jeong
Abstract
Long-form multi-shot video editing requires an intended edit to remain coherent across abrupt changes in scene, viewpoint, and subject appearance, while processing substantially longer visual context than conventional single-shot editors. We introduce MoviEdit, an end-to-end framework for scalable long-form multi-shot video editing that jointly addresses the challenges of data, modeling, and evaluation. To provide cross-shot supervision data at scale, we develop automatic data-generation pipelines spanning synchronized multi-view footage, real-world multi-shot videos, and generated multi-shot content, yielding 87K paired editing videos without manual annotation. Architecturally, MoviEdit employs a cascaded bidirectional-autoregressive design: a lightweight bidirectional stage first establishes a globally coherent editing draft over a spatiotemporally compressed representation, which is then autoregressively propagated and refined over the full video. We further introduce MoviEdit-Bench, a dedicated benchmark for systematic evaluation of long-form multi-shot video editing. Extensive experiments show that MoviEdit substantially outperforms existing open-source models and performs on par with industry frontier models across text-guided and image-guided editing. Visit our project page for complete results and to access the benchmark: https://multi-shot-edit.github.io/