用变分贝叶斯优化高斯点云，实现持续学习且无需重放数据。
Variational Bayes Gaussian Splatting
- 将高斯点云训练转化为参数的变分推断，利用共轭性质推导闭式更新公式。
- 在静态数据集上性能媲美当前最佳，在连续流数据下显著优于传统方法。
- 适合需要持续学习的3D场景建模任务，尤其适用于实时增量数据处理。
最近，3D高斯点云（3D Gaussian Splatting）作为一种基于高斯混合模型的3D场景建模方法受到关注。现有方法主要依赖可微渲染管道反向传播优化，但在处理连续数据流时易出现灾难性遗忘。为此，我们提出变分贝叶斯高斯点云（VBGS），将高斯点云训练建模为对模型参数的变分推断。利用多元高斯的共轭性质，推导出闭式变分更新规则，可在无需重放缓冲区的情况下，从部分、顺序观测中高效更新。实验表明，VBGS不仅在静态数据集上达到顶尖性能，还能有效实现从2D与3D数据流中持续学习，显著提升该场景下的表现。
原文摘要 · Abstract (English)
Recently, 3D Gaussian Splatting has emerged as a promising approach for modeling 3D scenes using mixtures of Gaussians. The predominant optimization method for these models relies on backpropagating gradients through a differentiable rendering pipeline, which struggles with catastrophic forgetting when dealing with continuous streams of data. To address this limitation, we propose Variational Bayes Gaussian Splatting (VBGS), a novel approach that frames training a Gaussian splat as variational inference over model parameters. By leveraging the conjugacy properties of multivariate Gaussians, we derive a closed-form variational update rule, allowing efficient updates from partial, sequential observations without the need for replay buffers. Our experiments show that VBGS not only matches state-of-the-art performance on static datasets, but also enables continual learning from sequentially streamed 2D and 3D data, drastically improving performance in this setting.
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。