Matryoshka Gaussian Splatting
2026-03-19 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionGraphics
AI summaryⓘ
The authors address the challenge of smoothly adjusting the detail level in 3D Gaussian Splatting (3DGS) models, which is important for practical use. They propose Matryoshka Gaussian Splatting (MGS), a training method that allows rendering from low to high detail smoothly using a single model without losing quality at full detail. Their approach uses a clever training trick where the model learns to create good images even when only part of its data is used. Tests show that MGS keeps the best image quality while letting users trade off speed and quality seamlessly.
3D Gaussian SplattingLevel of Detail (LoD)Continuous Level of DetailRenderingStochastic Budget TrainingModel CapacityTraining FrameworkImage FidelitySpeed-Quality Trade-offModel Optimization
Authors
Zhilin Guo, Boqiao Zhang, Hakan Aktas, Kyle Fogarty, Jeffrey Hu, Nursena Koprucu Aslan, Wenzhao Li, Canberk Baykal, Albert Miao, Josef Bengtson, Chenliang Zhou, Weihao Xia, Cristina Nader Vasconcelos. Cengiz Oztireli
Abstract
The ability to render scenes at adjustable fidelity from a single model, known as level of detail (LoD), is crucial for practical deployment of 3D Gaussian Splatting (3DGS). Existing discrete LoD methods expose only a limited set of operating points, while concurrent continuous LoD approaches enable smoother scaling but often suffer noticeable quality degradation at full capacity, making LoD a costly design decision. We introduce Matryoshka Gaussian Splatting (MGS), a training framework that enables continuous LoD for standard 3DGS pipelines without sacrificing full-capacity rendering quality. MGS learns a single ordered set of Gaussians such that rendering any prefix, the first k splats, produces a coherent reconstruction whose fidelity improves smoothly with increasing budget. Our key idea is stochastic budget training: each iteration samples a random splat budget and optimises both the corresponding prefix and the full set. This strategy requires only two forward passes and introduces no architectural modifications. Experiments across four benchmarks and six baselines show that MGS matches the full-capacity performance of its backbone while enabling a continuous speed-quality trade-off from a single model. Extensive ablations on ordering strategies, training objectives, and model capacity further validate the designs.