A Benchmark and Multi-Agent System for Instruction-driven Cinematic Video Compilation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913024652083200 |
|---|---|
| author | Zhang, Peixuan Zhou, Chang Zhang, Ziyuan Liu, Hualuo Zhang, Chunjie Liu, Jingqi Zhou, Xiaohui Chen, Xi Weng, Shuchen Li, Si Shi, Boxin |
| author_facet | Zhang, Peixuan Zhou, Chang Zhang, Ziyuan Liu, Hualuo Zhang, Chunjie Liu, Jingqi Zhou, Xiaohui Chen, Xi Weng, Shuchen Li, Si Shi, Boxin |
| contents | The surging demand for adapting long-form cinematic content into short videos has motivated the need for versatile automatic video compilation systems. However, existing compilation methods are limited to predefined tasks, and the community lacks a comprehensive benchmark to evaluate the cinematic compilation. To address this, we introduce CineBench, the first benchmark for instruction-driven cinematic video compilation, featuring diverse user instructions and high-quality ground-truth compilations annotated by professional editors. To overcome contextual collapse and temporal fragmentation, we present CineAgents, a multi-agent system that reformulates cinematic video compilation into ``design-and-compose'' paradigm. CineAgents performs script reverse-engineering to construct a hierarchical narrative memory to provide multi-level context and employs an iterative narrative planning process that refines a creative blueprint into a final compiled script. Extensive experiments demonstrate that CineAgents significantly outperforms existing methods, generating compilations with superior narrative coherence and logical coherence. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_10456 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | A Benchmark and Multi-Agent System for Instruction-driven Cinematic Video Compilation Zhang, Peixuan Zhou, Chang Zhang, Ziyuan Liu, Hualuo Zhang, Chunjie Liu, Jingqi Zhou, Xiaohui Chen, Xi Weng, Shuchen Li, Si Shi, Boxin Computer Vision and Pattern Recognition The surging demand for adapting long-form cinematic content into short videos has motivated the need for versatile automatic video compilation systems. However, existing compilation methods are limited to predefined tasks, and the community lacks a comprehensive benchmark to evaluate the cinematic compilation. To address this, we introduce CineBench, the first benchmark for instruction-driven cinematic video compilation, featuring diverse user instructions and high-quality ground-truth compilations annotated by professional editors. To overcome contextual collapse and temporal fragmentation, we present CineAgents, a multi-agent system that reformulates cinematic video compilation into ``design-and-compose'' paradigm. CineAgents performs script reverse-engineering to construct a hierarchical narrative memory to provide multi-level context and employs an iterative narrative planning process that refines a creative blueprint into a final compiled script. Extensive experiments demonstrate that CineAgents significantly outperforms existing methods, generating compilations with superior narrative coherence and logical coherence. |
| title | A Benchmark and Multi-Agent System for Instruction-driven Cinematic Video Compilation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.10456 |