AnyView: Synthesizing Any Novel View in Dynamic Scenes
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914275731177472 |
|---|---|
| author | Van Hoorick, Basile Chen, Dian Iwase, Shun Tokmakov, Pavel Irshad, Muhammad Zubair Vasiljevic, Igor Gupta, Swati Cheng, Fangzhou Zakharov, Sergey Guizilini, Vitor Campagnolo |
| author_facet | Van Hoorick, Basile Chen, Dian Iwase, Shun Tokmakov, Pavel Irshad, Muhammad Zubair Vasiljevic, Igor Gupta, Swati Cheng, Fangzhou Zakharov, Sergey Guizilini, Vitor Campagnolo |
| contents | Modern generative video models excel at producing convincing, high-quality outputs, but struggle to maintain multi-view and spatiotemporal consistency in highly dynamic real-world environments. In this work, we introduce \textbf{AnyView}, a diffusion-based video generation framework for \emph{dynamic view synthesis} with minimal inductive biases or geometric assumptions. We leverage multiple data sources with various levels of supervision, including monocular (2D), multi-view static (3D) and multi-view dynamic (4D) datasets, to train a generalist spatiotemporal implicit representation capable of producing zero-shot novel videos from arbitrary camera locations and trajectories. We evaluate AnyView on standard benchmarks, showing competitive results with the current state of the art, and propose \textbf{AnyViewBench}, a challenging new benchmark tailored towards \emph{extreme} dynamic view synthesis in diverse real-world scenarios. In this more dramatic setting, we find that most baselines drastically degrade in performance, as they require significant overlap between viewpoints, while AnyView maintains the ability to produce realistic, plausible, and spatiotemporally consistent videos when prompted from \emph{any} viewpoint. Results, data, code, and models can be viewed at: https://tri-ml.github.io/AnyView/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_16982 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | AnyView: Synthesizing Any Novel View in Dynamic Scenes Van Hoorick, Basile Chen, Dian Iwase, Shun Tokmakov, Pavel Irshad, Muhammad Zubair Vasiljevic, Igor Gupta, Swati Cheng, Fangzhou Zakharov, Sergey Guizilini, Vitor Campagnolo Computer Vision and Pattern Recognition Machine Learning Robotics Modern generative video models excel at producing convincing, high-quality outputs, but struggle to maintain multi-view and spatiotemporal consistency in highly dynamic real-world environments. In this work, we introduce \textbf{AnyView}, a diffusion-based video generation framework for \emph{dynamic view synthesis} with minimal inductive biases or geometric assumptions. We leverage multiple data sources with various levels of supervision, including monocular (2D), multi-view static (3D) and multi-view dynamic (4D) datasets, to train a generalist spatiotemporal implicit representation capable of producing zero-shot novel videos from arbitrary camera locations and trajectories. We evaluate AnyView on standard benchmarks, showing competitive results with the current state of the art, and propose \textbf{AnyViewBench}, a challenging new benchmark tailored towards \emph{extreme} dynamic view synthesis in diverse real-world scenarios. In this more dramatic setting, we find that most baselines drastically degrade in performance, as they require significant overlap between viewpoints, while AnyView maintains the ability to produce realistic, plausible, and spatiotemporally consistent videos when prompted from \emph{any} viewpoint. Results, data, code, and models can be viewed at: https://tri-ml.github.io/AnyView/ |
| title | AnyView: Synthesizing Any Novel View in Dynamic Scenes |
| topic | Computer Vision and Pattern Recognition Machine Learning Robotics |
| url | https://arxiv.org/abs/2601.16982 |