Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2412.03758 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929732623269888 |
|---|---|
| author | Ming, Ruibo Wu, Jingwei Huang, Zhewei Ju, Zhuoxuan HU, Jianming Peng, Lihui Zhou, Shuchang |
| author_facet | Ming, Ruibo Wu, Jingwei Huang, Zhewei Ju, Zhuoxuan HU, Jianming Peng, Lihui Zhou, Shuchang |
| contents | Recent advancements in auto-regressive large language models (LLMs) have led to their application in video generation. This paper explores the use of Large Vision Models (LVMs) for video continuation, a task essential for building world models and predicting future frames. We introduce ARCON, a scheme that alternates between generating semantic and RGB tokens, allowing the LVM to explicitly learn high-level structural video information. We find high consistency in the RGB images and semantic maps generated without special design. Moreover, we employ an optical flow-based texture stitching method to enhance visual quality. Experiments in autonomous driving scenarios show that our model can consistently generate long videos. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_03758 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | ARCON: Advancing Auto-Regressive Continuation for Driving Videos Ming, Ruibo Wu, Jingwei Huang, Zhewei Ju, Zhuoxuan HU, Jianming Peng, Lihui Zhou, Shuchang Computer Vision and Pattern Recognition Recent advancements in auto-regressive large language models (LLMs) have led to their application in video generation. This paper explores the use of Large Vision Models (LVMs) for video continuation, a task essential for building world models and predicting future frames. We introduce ARCON, a scheme that alternates between generating semantic and RGB tokens, allowing the LVM to explicitly learn high-level structural video information. We find high consistency in the RGB images and semantic maps generated without special design. Moreover, we employ an optical flow-based texture stitching method to enhance visual quality. Experiments in autonomous driving scenarios show that our model can consistently generate long videos. |
| title | ARCON: Advancing Auto-Regressive Continuation for Driving Videos |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2412.03758 |