cs.CVNov 3, 2026

3DGS-VBench+ and 3DGSV-Assessor: A Multi-Dimensional Subjective Benchmark and a Reference-Flexible Objective Model for 3D Gaussian Splatting Video Quality Assessment

Authors: Yuke Xing, William Gordon, Jiarui Wang, Yujie Zhang, Guangtao Zhai, Yiling Xu

Organizations: School of Information Science and Electronic Engineering, Shanghai Jiao Tong University, Shanghai 200240, China · Basis Independent Silicon Valley, San Jose, CA 95134, USA · JD.COM, Beijing, 100176, China

Abstract

3D Gaussian Splatting (3DGS) enables real-time novel view synthesis (NVS), but its heavy storage makes compression essential, and both training and compression introduce representation-specific distortions. Since generative compression in many methods treats geometric and color attributes via different parameters, it can induce dimension-specific distortions, motivating multi-dimensional assessment. Moreover, real-world captures generally lack an aligned ground-truth video for a synthesized, continuous camera trajectory, leaving only sparse captured views as references. To enable multi-dimensional quality assessment of 3DGS videos with scarce aligned references, we first present 3DGS-VBench+, a large-scale subjective video quality assessment (VQA) benchmark comprising 3{,}036 3DGS reconstructions, each paired with a rendered video, across 11 real-world and 22 synthetic scenes, generated by 6 representative compression algorithms at systematically designed levels and annotated with 9{,}108 MOSs covering overall, geometry, and color quality. Based on 3DGS-VBench+, we propose 3DGSV-Assessor, a reference-flexible, large multimodal framework that scores all three dimensions in one generation pass and fuses a ranking-based reference cue into the frozen regressor through a bounded gate, supporting both reduced-reference (RR) and no-reference (NR) assessment. Extensive experiments demonstrate state-of-the-art correlation with human perception on 3DGS-VBench+ and strong zero-shot generalization to other NVS benchmarks. The dataset and code will be released at https://github.com/YukeXing/3DGSV-Assessor.

Figures & tables

Explore similar work

CardsList
  1. 3DGSI-Assessor: A Large-Scale Dataset and An LMM-based Method for 3D Gaussian Splatting Image Quality Assessment

    Aug 4, 2026Yuke Xing, Jiarui Wang, William Gordon +33DGS Compression3D Gaussian Splatting

  2. SpatialQ: Understanding 3D Gaussian Splatting Scene Quality via Visual-based MLLM

    Jul 29, 2026Jingxuan Su, Shenglin Wang, Tiesong Zhao +23D Gaussian Splatting3D Scene Understanding

  3. GS-PQM: A Parameter-Domain Quality Metric for Compressed Gaussian Splatting

    Sep 18, 2026Pedro Martin, António Rodrigues, João Ascenso +1Gaussian Splatting Compression3DGS Compression