CausNVS：用于灵活三维新视角合成的自回归多视角扩散模型
CausNVS: Autoregressive Multi-view Diffusion for Flexible 3D Novel View Synthesis
AI总结：
针对非自回归多视角扩散模型在视角数量和推理速度上的局限，提出自回归模型CausNVS，支持任意视角配置并顺序生成，结合因果掩码、逐帧噪声和CaPE编码，实现灵活高质量的新视角合成。
摘要：
多视角扩散模型在三维新视角合成中展现出潜力，但现有大多数方法采用非自回归形式。这限制了它们在世界建模中的应用，因为它们仅支持固定数量的视角，并且由于同时对所有帧进行去噪，导致推理速度较慢。为解决这些局限性，我们提出了CausNVS，一种自回归设置下的多视角扩散模型，它支持任意的输入-输出视角配置，并顺序生成视角。我们使用因果掩码和逐帧噪声训练CausNVS，并采用成对相对相机姿态编码（CaPE）实现精确的相机控制。在推理时，我们结合空间感知的滑动窗口与键值缓存和噪声条件增强来缓解漂移。我们的实验表明，CausNVS支持广泛的相机轨迹，实现灵活的自回归新视角合成，并在各种设置下持续获得强大的视觉质量。项目页面：https://kxhit.github.io/CausNVS.html。
英文摘要：
Multi-view diffusion models have shown promise in 3D novel view synthesis, but most existing methods adopt a non-autoregressive formulation. This limits their applicability in world modeling, as they only support a fixed number of views and suffer from slow inference due to denoising all frames simultaneously. To address these limitations, we propose CausNVS, a multi-view diffusion model in an autoregressive setting, which supports arbitrary input-output view configurations and generates views sequentially. We train CausNVS with causal masking and per-frame noise, using pairwise-relative camera pose encodings (CaPE) for precise camera control. At inference time, we combine a spatially-aware sliding-window with key-value caching and noise conditioning augmentation to mitigate drift. Our experiments demonstrate that CausNVS supports a broad range of camera trajectories, enables flexible autoregressive novel view synthesis, and achieves consistently strong visual quality across diverse settings. Project page: https://kxhit.github.io/CausNVS.html.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
