发表机构
Indian Institute of Science; Samsung R&D Institute India - Bangalore（印度科学学院; 三星印度班加罗尔研发研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对动态场景4D表示的混叠问题，提出运动感知3D平滑滤波器，适配各类4D表示，在标准数据集上性能优于现有最优方法。
AI 中文摘要
动态场景的新视角合成对增强现实/虚拟现实（AR/VR）应用至关重要，但仍是一个具有挑战性的问题。近期方法通过将时间作为第四维（4D表示），对3D高斯溅射（3DGS）、神经辐射场（NeRF）等表示进行适配以处理动态场景。这些4D表示仍存在混叠伪影，尤其是在从发散视点（缩放操作）生成新视角时。尽管使用Mip-Splatting中提出的3D平滑滤波器看似可行，但该方法未考虑局部运动，仍会出现混叠。为解决此问题，我们提出一种专为4D表示设计的运动感知3D平滑滤波器。我们的方法基于局部运动信息调整滤波器强度，在不损害渲染质量的情况下有效缓解混叠，具体通过非参数估计方法估计时间与焦距-深度比的联合概率密度函数实现；推理阶段，我们从该联合分布中采样以确定合适的平滑滤波器。这种灵活策略可集成到各类4D表示中。我们在标准数据集上的评估表明，与现有最优方法相比，我们的方法性能更优。
英文摘要
Novel-view synthesis of dynamic scenes, crucial for AR/VR applications, remains a challenging problem. Recent methods adapt representations like 3D Gaussian Splatting (3DGS) and Neural Radiance Fields (NeRF) for dynamic scenes by incorporating time as the fourth dimension (4D representations). These 4D representations still suffer from aliasing artifacts, especially when generating novel views from divergent viewpoints (zoom-in/zoom-out operations). While using 3D smoothing filters like those proposed in Mip-Splatting might seem like a possible solution, they fail to account for local motion and also exhibit aliasing. To address this, we propose a motion-aware 3D smoothing filter specifically designed for 4D representations. Our approach adapts the filter strength based on local motion information, effectively mitigating aliasing without compromising rendering quality. This is achieved by estimating the joint density function of time and focal-to-depth ratio using a non-parametric estimation method. During inference, we sample from this joint distribution to determine the appropriate smoothing filter. This flexible strategy can be integrated with various 4D representations. Our evaluations on standard datasets demonstrate superior performance compared to state-of-the-art methods.
CommentsAccepted to ECCV 2026. Project Page: https://maaf-4dgs.github.io/
