Perceptual-GS: Scene-adaptive Perceptual Densification for Gaussian Splatting

TL;DR

Perceptual-GS introduces a scene-adaptive framework that leverages human visual perception to optimize Gaussian primitive distribution, significantly improving 3D scene reconstruction quality and efficiency.

Contribution

It presents a novel perceptual sensitivity-aware approach for adaptive Gaussian densification in 3D Gaussian Splatting, enhancing reconstruction and robustness.

Findings

Achieves state-of-the-art reconstruction quality.

Improves efficiency and robustness in large-scale scenes.

Effectively allocates resources to visually critical regions.

Abstract

3D Gaussian Splatting (3DGS) has emerged as a powerful technique for novel view synthesis. However, existing methods struggle to adaptively optimize the distribution of Gaussian primitives based on scene characteristics, making it challenging to balance reconstruction quality and efficiency. Inspired by human perception, we propose scene-adaptive perceptual densification for Gaussian Splatting (Perceptual-GS), a novel framework that integrates perceptual sensitivity into the 3DGS training process to address this challenge. We first introduce a perception-aware representation that models human visual sensitivity while constraining the number of Gaussian primitives. Building on this foundation, we develop a perceptual sensitivity-adaptive distribution to allocate finer Gaussian granularity to visually critical regions, enhancing reconstruction quality and robustness. Extensive evaluations on multiple datasets, including BungeeNeRF for large-scale scenes, demonstrate that Perceptual-GS achieves state-of-the-art performance in reconstruction quality, efficiency, and robustness. The code is publicly available at: https://github.com/eezkni/Perceptual-GS