OmniWeaving: Towards Unified Video Generation with Free-form Composition and Reasoning
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914439086735360 |
|---|---|
| author | Pan, Kaihang Tian, Qi Zhang, Jianwei Kong, Weijie Xiong, Jiangfeng Long, Yanxin Zhang, Shixue Qiu, Haiyi Wang, Tan Lv, Zheqi Wu, Yue Bo, Liefeng Tang, Siliang Zhong, Zhao |
| author_facet | Pan, Kaihang Tian, Qi Zhang, Jianwei Kong, Weijie Xiong, Jiangfeng Long, Yanxin Zhang, Shixue Qiu, Haiyi Wang, Tan Lv, Zheqi Wu, Yue Bo, Liefeng Tang, Siliang Zhong, Zhao |
| contents | While proprietary systems such as Seedance-2.0 have achieved remarkable success in omni-capable video generation, open-source alternatives significantly lag behind. Most academic models remain heavily fragmented, and the few existing efforts toward unified video generation still struggle to seamlessly integrate diverse tasks within a single framework. To bridge this gap, we propose OmniWeaving, an omni-level video generation model featuring powerful multimodal composition and reasoning-informed capabilities. By leveraging a massive-scale pretraining dataset that encompasses diverse compositional and reasoning-augmented scenarios, OmniWeaving learns to temporally bind interleaved text, multi-image, and video inputs while acting as an intelligent agent to infer complex user intentions for sophisticated video creation. Furthermore, we introduce IntelligentVBench, the first comprehensive benchmark designed to rigorously assess next-level intelligent unified video generation. Extensive experiments demonstrate that OmniWeaving achieves SoTA performance among open-source unified models. The codes and model have already been publicly available. Project Page: https://omniweaving.github.io. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_24458 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | OmniWeaving: Towards Unified Video Generation with Free-form Composition and Reasoning Pan, Kaihang Tian, Qi Zhang, Jianwei Kong, Weijie Xiong, Jiangfeng Long, Yanxin Zhang, Shixue Qiu, Haiyi Wang, Tan Lv, Zheqi Wu, Yue Bo, Liefeng Tang, Siliang Zhong, Zhao Computer Vision and Pattern Recognition While proprietary systems such as Seedance-2.0 have achieved remarkable success in omni-capable video generation, open-source alternatives significantly lag behind. Most academic models remain heavily fragmented, and the few existing efforts toward unified video generation still struggle to seamlessly integrate diverse tasks within a single framework. To bridge this gap, we propose OmniWeaving, an omni-level video generation model featuring powerful multimodal composition and reasoning-informed capabilities. By leveraging a massive-scale pretraining dataset that encompasses diverse compositional and reasoning-augmented scenarios, OmniWeaving learns to temporally bind interleaved text, multi-image, and video inputs while acting as an intelligent agent to infer complex user intentions for sophisticated video creation. Furthermore, we introduce IntelligentVBench, the first comprehensive benchmark designed to rigorously assess next-level intelligent unified video generation. Extensive experiments demonstrate that OmniWeaving achieves SoTA performance among open-source unified models. The codes and model have already been publicly available. Project Page: https://omniweaving.github.io. |
| title | OmniWeaving: Towards Unified Video Generation with Free-form Composition and Reasoning |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.24458 |