3DGS.zip: A survey on 3D Gaussian Splatting Compression Methods(1)

Abstract

We present a work-in-progress survey on 3DGS compression methods, focusing on their statistical performance across various benchmarks. This survey aims to facilitate comparability by summarizing key statistics of different compression approaches in a tabulated format.

The datasets evaluated include TanksAndTemples, MipNeRF360, DeepBlending, and SyntheticNeRF.

For each method, we report the Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), Learned Perceptual Image Patch Similarity (LPIPS), and the resultant size in megabytes (MB), as provided by the respective authors.

project website

Figure

The best methods in each category are highlighted ( fist , second , third ). The ranks represent the average rankings of the methods across all available datasets.

The quality metrics PSNR, SSIM, and LPIPS are equally weighted with the model size, meaning they each contribute one-sixth to the ranks, while the size contributes half.

Figure 1

Reducing the Memory Footprint of 3D Gaussian Splatting

This approach addresses three main issues contributing to large storage sizes in 3DGS.

To reduce the number