arXiv cs.AI· Gaurav Parmar, Zhengqi Li, Richard Zhang, Jun-Yan Zhu, Srinivasa Narasimhan, Eli Shechtman, Yotam Nitzan·· 3 小时前AI 评分42
vid2vid-long：基于显式对应关系实现长时序一致的视频到视频翻译
Consistent Video-to-Video Translation via Explicit Correspondences
AI 导读
研究者提出 vid2vid-long，通过输入视频中的 patch 级对应关系检索先前生成输出，并以 correspondence cross-attention 作为条件，解决固定长度因果窗口导致的远程时序一致性缺失。实验显示该方法在多个近期视频模型上持续提升长时程一致性，同时保持生成质量与实时性能。论文已被 ECCV 2026 接收。
正文 · 原文
Abstract:Interactive video-to-video applications require real-time generation while maintaining long-range temporal consistency. However, recent methods achieve speed by restricting attention to a fixed-length causal window. This limits long-range consistency whenever relevant history falls outside the context window. To address this limitation, we introduce vid2vid-long, a correspondence-based approach grounded in a simple observation. Video-to-video translation provides an explicit cross-frame reference: the regions that match in the input video should match in the output video too. Concretely, we compute patch-level correspondences in the input, guiding retrieval of previously generated outputs in the corresponding regions, and condition the model on these patches through correspondence cross-attention. Our experiments demonstrate that vid2vid-long consistently improves long-range temporal consistency across recent video models while preserving generation quality and real-time performance.
| Comments: | ECCV 2026. Project Website: this https URL |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Graphics (cs.GR); Machine Learning (cs.LG) |
| Cite as: | arXiv:2611.17298 [cs.CV] |
| (or arXiv:2611.17298v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2611.17298 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Gaurav Parmar [view email]
[v1]
Thu, 12 Nov 2026 18:55:09 UTC (10,337 KB)
来源：arXiv cs.AI · arxiv.org