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.
@article{arxiv.2410.03592,
title = {Variational Bayes Gaussian Splatting},
author = {Toon Van de Maele and Ozan Catal and Alexander Tschantz and Christopher L. Buckley and Tim Verbelen},
journal= {arXiv preprint arXiv:2410.03592},
year = {2025}
}