A Survey on Compression and Quality Assessment Techniques for 3D Gaussian Splatting

D Gaussian splatting (3DGS) has emerged as a prevalent paradigm for 3D scene construction, employing 3D Gaussians to efficiently represent complex scenes. Despite its significant advantages in rendering quality and speed, 3DGS faces considerable limitations due to the unaffordable storage requirements, as the representation necessitates a substantial number of 3D Gaussians. This constraint has catalyzed research in two complementary domains: compression to reduce model footprints and quality assessment to evaluate the perceptual impact of compression. This survey provides a comprehensive overview of recent advancements in these two fields. Specifically, we review various compression techniques by systematically analyzing their theoretical foundations, performance, and limitations. Additionally, we investigate quality assessment studies tailored for 3DGS, with particular attention to subjective databases. This survey aims to provide researchers with a comprehensive understanding of the current landscape in 3D Gaussian compression and quality assessment, highlighting the accomplishments and key challenges in this rapidly evolving research field.