Temporal Regularization Makes Your Video Generator Stronger
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_ | 1866909543338868736 |
|---|---|
| author | Chen, Harold Haodong Huang, Haojian Wu, Xianfeng Liu, Yexin Bai, Yajing Shu, Wen-Jie Yang, Harry Lim, Ser-Nam |
| author_facet | Chen, Harold Haodong Huang, Haojian Wu, Xianfeng Liu, Yexin Bai, Yajing Shu, Wen-Jie Yang, Harry Lim, Ser-Nam |
| contents | Temporal quality is a critical aspect of video generation, as it ensures consistent motion and realistic dynamics across frames. However, achieving high temporal coherence and diversity remains challenging. In this work, we explore temporal augmentation in video generation for the first time, and introduce FluxFlow for initial investigation, a strategy designed to enhance temporal quality. Operating at the data level, FluxFlow applies controlled temporal perturbations without requiring architectural modifications. Extensive experiments on UCF-101 and VBench benchmarks demonstrate that FluxFlow significantly improves temporal coherence and diversity across various video generation models, including U-Net, DiT, and AR-based architectures, while preserving spatial fidelity. These findings highlight the potential of temporal augmentation as a simple yet effective approach to advancing video generation quality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_15417 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Temporal Regularization Makes Your Video Generator Stronger Chen, Harold Haodong Huang, Haojian Wu, Xianfeng Liu, Yexin Bai, Yajing Shu, Wen-Jie Yang, Harry Lim, Ser-Nam Computer Vision and Pattern Recognition Artificial Intelligence Temporal quality is a critical aspect of video generation, as it ensures consistent motion and realistic dynamics across frames. However, achieving high temporal coherence and diversity remains challenging. In this work, we explore temporal augmentation in video generation for the first time, and introduce FluxFlow for initial investigation, a strategy designed to enhance temporal quality. Operating at the data level, FluxFlow applies controlled temporal perturbations without requiring architectural modifications. Extensive experiments on UCF-101 and VBench benchmarks demonstrate that FluxFlow significantly improves temporal coherence and diversity across various video generation models, including U-Net, DiT, and AR-based architectures, while preserving spatial fidelity. These findings highlight the potential of temporal augmentation as a simple yet effective approach to advancing video generation quality. |
| title | Temporal Regularization Makes Your Video Generator Stronger |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2503.15417 |