Generalized and Efficient 2D Gaussian Splatting for Arbitrary-scale Super-Resolution

TL;DR

This paper introduces GSASR, a novel method that employs generalized Gaussian Splatting for efficient and arbitrary-scale super-resolution, achieving superior quality and speed compared to traditional INR-based approaches.

Contribution

We develop a new architecture and GPU-based rasterization technique to adapt Gaussian Splatting for generalizable, efficient super-resolution across arbitrary scales.

Findings

Outperforms INR-based methods in quality and speed

Supports arbitrary scaling factors in a single model

Achieves high-quality super-resolution on diverse images

Abstract

Implicit Neural Representations (INR) have been successfully employed for Arbitrary-scale Super-Resolution (ASR). However, INR-based models need to query the multi-layer perceptron module numerous times and render a pixel in each query, resulting in insufficient representation capability and low computational efficiency. Recently, Gaussian Splatting (GS) has shown its advantages over INR in both visual quality and rendering speed in 3D tasks, which motivates us to explore whether GS can be employed for the ASR task. However, directly applying GS to ASR is exceptionally challenging because the original GS is an optimization-based method through overfitting each single scene, while in ASR we aim to learn a single model that can generalize to different images and scaling factors. We overcome these challenges by developing two novel techniques. Firstly, to generalize GS for ASR, we elaborately design an architecture to predict the corresponding image-conditioned Gaussians of the input low-resolution image in a feed-forward manner. Each Gaussian can fit the shape and direction of an area of complex textures, showing powerful representation capability. Secondly, we implement an efficient differentiable 2D GPU/CUDA-based scale-aware rasterization to render super-resolved images by sampling discrete RGB values from the predicted continuous Gaussians. Via end-to-end training, our optimized network, namely GSASR, can perform ASR for any image and unseen scaling factors. Extensive experiments validate the effectiveness of our proposed method. The code and models are available at https://github.com/ChrisDud0257/GSASR.