Large Images are Gaussians: High-Quality Large Image Representation with Levels of 2D Gaussian Splatting

TL;DR

This paper introduces LIG, a novel method that uses levels of 2D Gaussian splatting to efficiently and accurately represent large images, overcoming previous limitations in fitting and detail reconstruction.

Contribution

LIG advances large image representation by employing a level-of-Gaussian approach and optimized Gaussian fitting strategies, enabling high-quality results with large images.

Findings

Successfully represents large images with Gaussian points

Achieves high-quality image reconstruction with detailed features

Demonstrates effectiveness across various large image types

Abstract

While Implicit Neural Representations (INRs) have demonstrated significant success in image representation, they are often hindered by large training memory and slow decoding speed. Recently, Gaussian Splatting (GS) has emerged as a promising solution in 3D reconstruction due to its high-quality novel view synthesis and rapid rendering capabilities, positioning it as a valuable tool for a broad spectrum of applications. In particular, a GS-based representation, 2DGS, has shown potential for image fitting. In our work, we present \textbf{L}arge \textbf{I}mages are \textbf{G}aussians (\textbf{LIG}), which delves deeper into the application of 2DGS for image representations, addressing the challenge of fitting large images with 2DGS in the situation of numerous Gaussian points, through two distinct modifications: 1) we adopt a variant of representation and optimization strategy, facilitating the fitting of a large number of Gaussian points; 2) we propose a Level-of-Gaussian approach for reconstructing both coarse low-frequency initialization and fine high-frequency details. Consequently, we successfully represent large images as Gaussian points and achieve high-quality large image representation, demonstrating its efficacy across various types of large images. Code is available at {\href{https://github.com/HKU-MedAI/LIG}{https://github.com/HKU-MedAI/LIG}}.