Arbitrary-Scale 3D Gaussian Super-Resolution

TL;DR

This paper introduces an integrated framework for 3D Gaussian super-resolution that enables high-quality, arbitrary-scale rendering with a single model, overcoming limitations of existing methods in resource-limited scenarios.

Contribution

The proposed method combines scale-aware rendering, generative prior-guided optimization, and progressive super-resolution to achieve flexible, high-quality 3D view synthesis at arbitrary scales.

Findings

6.59 dB PSNR gain over 3DGS

Real-time rendering at 85 FPS at 1080p

Supports both integer and non-integer scale rendering

Abstract

Existing 3D Gaussian Splatting (3DGS) super-resolution methods typically perform high-resolution (HR) rendering of fixed scale factors, making them impractical for resource-limited scenarios. Directly rendering arbitrary-scale HR views with vanilla 3DGS introduces aliasing artifacts due to the lack of scale-aware rendering ability, while adding a post-processing upsampler for 3DGS complicates the framework and reduces rendering efficiency. To tackle these issues, we build an integrated framework that incorporates scale-aware rendering, generative prior-guided optimization, and progressive super-resolving to enable 3D Gaussian super-resolution of arbitrary scale factors with a single 3D model. Notably, our approach supports both integer and non-integer scale rendering to provide more flexibility. Extensive experiments demonstrate the effectiveness of our model in rendering high-quality arbitrary-scale HR views (6.59 dB PSNR gain over 3DGS) with a single model. It preserves structural consistency with LR views and across different scales, while maintaining real-time rendering speed (85 FPS at 1080p).