Optimized 3D Gaussian Splatting using Coarse-to-Fine Image Frequency Modulation
This addresses a practical bottleneck for deploying high-quality 3D scene reconstruction on consumer-grade devices, offering an incremental improvement over existing 3DGS methods.
They tackled the high GPU memory and storage demands of 3D Gaussian Splatting for novel view synthesis by proposing Opti3DGS, which reduced Gaussians by 62%, GPU memory by 40%, and optimization time by 20% without sacrificing visual quality.
The field of Novel View Synthesis has been revolutionized by 3D Gaussian Splatting (3DGS), which enables high-quality scene reconstruction that can be rendered in real-time. 3DGS-based techniques typically suffer from high GPU memory and disk storage requirements which limits their practical application on consumer-grade devices. We propose Opti3DGS, a novel frequency-modulated coarse-to-fine optimization framework that aims to minimize the number of Gaussian primitives used to represent a scene, thus reducing memory and storage demands. Opti3DGS leverages image frequency modulation, initially enforcing a coarse scene representation and progressively refining it by modulating frequency details in the training images. On the baseline 3DGS, we demonstrate an average reduction of 62% in Gaussians, a 40% reduction in the training GPU memory requirements and a 20% reduction in optimization time without sacrificing the visual quality. Furthermore, we show that our method integrates seamlessly with many 3DGS-based techniques, consistently reducing the number of Gaussian primitives while maintaining, and often improving, visual quality. Additionally, Opti3DGS inherently produces a level-of-detail scene representation at no extra cost, a natural byproduct of the optimization pipeline. Results and code will be made publicly available.