提出自回归多视角扩散模型，可灵活生成任意视图序列。
CausNVS: Autoregressive Multi-view Diffusion for Flexible 3D Novel View Synthesis
- 采用自回归生成，逐帧输出新视角图像。
- 支持任意输入输出视图配置，推理速度更快。
- 适合需要动态相机轨迹的3D内容生成场景。
多视角扩散模型在3D新视角合成中表现良好，但现有方法多为非自回归形式，仅支持固定数量视图，且因需同时去噪所有帧导致推理缓慢。为此，我们提出CausNVS，一种基于自回归设置的多视角扩散模型，可支持任意输入输出视图配置，并逐帧生成视图。训练时使用因果掩码和每帧噪声，结合成对相对相机位姿编码（CaPE）实现精确相机控制。推理时，通过空间感知滑动窗口、键值缓存与噪声条件增强缓解生成漂移。实验表明，CausNVS支持广泛相机轨迹，实现灵活的自回归新视角合成，在多种设置下均保持优异视觉质量。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。