GS-QA: Comprehensive Quality Assessment Benchmark for Gaussian Splatting View Synthesis
This work addresses the problem of evaluating rendering quality for researchers and practitioners in 3D computer vision, though it is incremental as it applies existing assessment methods to a new domain.
The paper tackles the lack of in-depth quality assessment for Gaussian Splatting (GS) view synthesis by conducting a subjective study to evaluate videos from state-of-the-art GS methods across diverse scenes, and it analyzes 18 objective metrics against human scores to provide a benchmark database.
Gaussian Splatting (GS) offers a promising alternative to Neural Radiance Fields (NeRF) for real-time 3D scene rendering. Using a set of 3D Gaussians to represent complex geometry and appearance, GS achieves faster rendering times and reduced memory consumption compared to the neural network approach used in NeRF. However, quality assessment of GS-generated static content is not yet explored in-depth. This paper describes a subjective quality assessment study that aims to evaluate synthesized videos obtained with several static GS state-of-the-art methods. The methods were applied to diverse visual scenes, covering both 360-degree and forward-facing (FF) camera trajectories. Moreover, the performance of 18 objective quality metrics was analyzed using the scores resulting from the subjective study, providing insights into their strengths, limitations, and alignment with human perception. All videos and scores are made available providing a comprehensive database that can be used as benchmark on GS view synthesis and objective quality metrics.