通过聚类纠错与分组4D高斯点云，提升动态场景重建精度。
Clustered Error Correction with Grouped 4D Gaussian Splatting
- 按渲染误差类型聚类，针对性修复缺失颜色与遮挡问题。
- 在Technicolor数据集上提升0.39dB的PSNR，显著改善时序一致性。
- 适合需要高质量动态3D重建的研究者与开发者。
现有4D高斯点云（4DGS）方法在动态场景重建中常因像素对应模糊及动态区域稀疏化而表现不佳。本文提出新方法，包含两大核心组件：(1) 椭圆误差聚类与误差校正点添加，用于定位动态区域并优化初始化点；(2) 分组4D高斯点云，增强点与动态物体间的映射一致性。我们区分渲染误差为缺色与遮挡两类，通过跨视角颜色一致性引导的反投影或前景分割实现精准修正。在Neural 3D Video与Technicolor数据集上的评估表明，本方法显著提升时序一致性，达到当前最优感知渲染质量，在Technicolor Light Field数据集上提升0.39dB PSNR。可视化显示点云与动态物体对齐更准确，且误差纠正机制能有效识别错误并正确初始化新点。代码与实现细节见https://github.com/tho-kn/cem-4dgs。
原文摘要 · Abstract (English)
Existing 4D Gaussian Splatting (4DGS) methods struggle to accurately reconstruct dynamic scenes, often failing to resolve ambiguous pixel correspondences and inadequate densification in dynamic regions. We address these issues by introducing a novel method composed of two key components: (1) Elliptical Error Clustering and Error Correcting Splat Addition that pinpoints dynamic areas to improve and initialize fitting splats, and (2) Grouped 4D Gaussian Splatting that improves consistency of mapping between splats and represented dynamic objects. Specifically, we classify rendering errors into missing-color and occlusion types, then apply targeted corrections via backprojection or foreground splitting guided by cross-view color consistency. Evaluations on Neural 3D Video and Technicolor datasets demonstrate that our approach significantly improves temporal consistency and achieves state-of-the-art perceptual rendering quality, improving 0.39dB of PSNR on the Technicolor Light Field dataset. Our visualization shows improved alignment between splats and dynamic objects, and the error correction method's capability to identify errors and properly initialize new splats. Our implementation details and source code are available at https://github.com/tho-kn/cem-4dgs.
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。