Vchitect-2.0: Parallel Transformer for Scaling Up Video Diffusion Models
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913651227623424 |
|---|---|
| author | Fan, Weichen Si, Chenyang Song, Junhao Yang, Zhenyu He, Yinan Zhuo, Long Huang, Ziqi Dong, Ziyue He, Jingwen Pan, Dongwei Wang, Yi Jiang, Yuming Wang, Yaohui Gao, Peng Chen, Xinyuan Li, Hengjie Lin, Dahua Qiao, Yu Liu, Ziwei |
| author_facet | Fan, Weichen Si, Chenyang Song, Junhao Yang, Zhenyu He, Yinan Zhuo, Long Huang, Ziqi Dong, Ziyue He, Jingwen Pan, Dongwei Wang, Yi Jiang, Yuming Wang, Yaohui Gao, Peng Chen, Xinyuan Li, Hengjie Lin, Dahua Qiao, Yu Liu, Ziwei |
| contents | We present Vchitect-2.0, a parallel transformer architecture designed to scale up video diffusion models for large-scale text-to-video generation. The overall Vchitect-2.0 system has several key designs. (1) By introducing a novel Multimodal Diffusion Block, our approach achieves consistent alignment between text descriptions and generated video frames, while maintaining temporal coherence across sequences. (2) To overcome memory and computational bottlenecks, we propose a Memory-efficient Training framework that incorporates hybrid parallelism and other memory reduction techniques, enabling efficient training of long video sequences on distributed systems. (3) Additionally, our enhanced data processing pipeline ensures the creation of Vchitect T2V DataVerse, a high-quality million-scale training dataset through rigorous annotation and aesthetic evaluation. Extensive benchmarking demonstrates that Vchitect-2.0 outperforms existing methods in video quality, training efficiency, and scalability, serving as a suitable base for high-fidelity video generation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_08453 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Vchitect-2.0: Parallel Transformer for Scaling Up Video Diffusion Models Fan, Weichen Si, Chenyang Song, Junhao Yang, Zhenyu He, Yinan Zhuo, Long Huang, Ziqi Dong, Ziyue He, Jingwen Pan, Dongwei Wang, Yi Jiang, Yuming Wang, Yaohui Gao, Peng Chen, Xinyuan Li, Hengjie Lin, Dahua Qiao, Yu Liu, Ziwei Computer Vision and Pattern Recognition Machine Learning We present Vchitect-2.0, a parallel transformer architecture designed to scale up video diffusion models for large-scale text-to-video generation. The overall Vchitect-2.0 system has several key designs. (1) By introducing a novel Multimodal Diffusion Block, our approach achieves consistent alignment between text descriptions and generated video frames, while maintaining temporal coherence across sequences. (2) To overcome memory and computational bottlenecks, we propose a Memory-efficient Training framework that incorporates hybrid parallelism and other memory reduction techniques, enabling efficient training of long video sequences on distributed systems. (3) Additionally, our enhanced data processing pipeline ensures the creation of Vchitect T2V DataVerse, a high-quality million-scale training dataset through rigorous annotation and aesthetic evaluation. Extensive benchmarking demonstrates that Vchitect-2.0 outperforms existing methods in video quality, training efficiency, and scalability, serving as a suitable base for high-fidelity video generation. |
| title | Vchitect-2.0: Parallel Transformer for Scaling Up Video Diffusion Models |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2501.08453 |