RAVE: Rate-Adaptive Visual Encoding for 3D Gaussian Splatting

TL;DR

This paper introduces RAVE, a flexible, lightweight compression method for 3D Gaussian Splatting that allows dynamic rate adjustment without retraining, maintaining high rendering quality across various bandwidths.

Contribution

It presents a novel, rate-adaptive compression scheme for 3D Gaussian Splatting that supports interpolation at any rate, enhancing flexibility and efficiency in neural scene representations.

Findings

Supports interpolation at any rate within bounds

Requires no retraining for different rates

Maintains high rendering quality across various operating points

Abstract

Recent advances in neural scene representations have transformed immersive multimedia, with 3D Gaussian Splatting (3DGS) enabling real-time photorealistic rendering. Despite its efficiency, 3DGS suffers from large memory requirements and costly training procedures, motivating efforts toward compression. Existing approaches, however, operate at fixed rates, limiting adaptability to varying bandwidth and device constraints. In this work, we propose a flexible compression scheme for 3DGS that supports interpolation at any rate between predefined bounds. Our method is computationally lightweight, requires no retraining for any rate, and preserves rendering quality across a broad range of operating points. Experiments demonstrate that the approach achieves efficient, high-quality compression while offering dynamic rate control, making it suitable for practical deployment in immersive applications. The code is available at https://github.com/inspiros/RAVE.