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/

翻译：现代生成式视频模型在生成逼真的高质量输出方面表现出色，但在高度动态的真实世界环境中，难以保持多视角和时空一致性。本文提出 \textbf{AnyView}，一个基于扩散的视频生成框架，用于实现 \emph{动态视角合成}，且仅需最小的归纳偏置或几何假设。我们利用多种监督级别的数据源，包括单目（2D）、多视角静态（3D）和多视角动态（4D）数据集，来训练一个通用的时空隐式表示模型。该模型能够从任意相机位置和轨迹生成零样本的新视频。我们在标准基准测试上评估 AnyView，结果显示其与当前最先进方法相比具有竞争力，并提出了 \textbf{AnyViewBench}，这是一个专为多样化真实世界场景中 \emph{极端} 动态视角合成而设计的、具有挑战性的新基准。在这一更具挑战性的设定下，我们发现大多数基线方法的性能急剧下降，因为它们需要视角之间有显著重叠，而 AnyView 在从 \emph{任意} 视角给出提示时，仍能保持生成逼真、合理且时空一致的视频的能力。结果、数据、代码和模型可访问：https://tri-ml.github.io/AnyView/