Perceptual Quality Assessment of 3D Gaussian Splatting: A Subjective Dataset and Prediction Metric

TL;DR

This paper introduces the first subjective quality assessment dataset and a no-reference prediction model for 3D Gaussian Splatting, addressing the perceptual impact of common distortions in 3D visualization.

Contribution

It provides a novel dataset for 3DGS quality evaluation and proposes a structure-aware, no-reference quality prediction model operating directly on Gaussian primitives.

Findings

The proposed model outperforms existing quality assessment methods.

Common distortions significantly affect perceptual quality in 3DGS.

The dataset enables systematic study of perceptual factors in 3D visualization.

Abstract

With the rapid advancement of 3D visualization, 3D Gaussian Splatting (3DGS) has emerged as a leading technique for real-time, high-fidelity rendering. While prior research has emphasized algorithmic performance and visual fidelity, the perceptual quality of 3DGS-rendered content, especially under varying reconstruction conditions, remains largely underexplored. In practice, factors such as viewpoint sparsity, limited training iterations, point downsampling, noise, and color distortions can significantly degrade visual quality, yet their perceptual impact has not been systematically studied. To bridge this gap, we present 3DGS-QA, the first subjective quality assessment dataset for 3DGS. It comprises 225 degraded reconstructions across 15 object types, enabling a controlled investigation of common distortion factors. Based on this dataset, we introduce a no-reference quality prediction model that directly operates on native 3D Gaussian primitives, without requiring rendered images or ground-truth references. Our model extracts spatial and photometric cues from the Gaussian representation to estimate perceived quality in a structure-aware manner. We further benchmark existing quality assessment methods, spanning both traditional and learning-based approaches. Experimental results show that our method consistently achieves superior performance, highlighting its robustness and effectiveness for 3DGS content evaluation. The dataset and code are made publicly available at https://github.com/diaoyn/3DGSQA to facilitate future research in 3DGS quality assessment.