Search papers, labs, and topics across Lattice.
This paper introduces 3DGS-IEval-15K+, a comprehensive dataset designed for assessing the quality of compressed 3D Gaussian Splatting (3DGS) images, featuring 15,200 images across diverse scenes and annotated with 45,600 mean opinion scores. It also presents 3DGSI-Assessor, a novel framework that leverages a large multimodal model to evaluate overall, geometry, and color quality in a unified manner, addressing the limitations of traditional image quality assessment metrics. The framework achieves state-of-the-art performance on the dataset while demonstrating robust generalization across other novel view synthesis benchmarks.
3DGSI-Assessor reveals that traditional IQA metrics fail to capture critical distortions in 3D Gaussian Splatting, leading to a breakthrough in multi-dimensional quality assessment.
3D Gaussian Splatting (3DGS) has become a dominant representation for real-time novel view synthesis (NVS), yet its storage footprint makes compression indispensable for practical deployment. 3DGS training and compression introduce representation-specific distortions such as floating artifacts and surface scattering, which conventional image quality assessment (IQA) metrics fail to capture. Moreover, the independent compression of geometric and color attributes may lead to decoupled dimension-specific distortions that must be diagnosed separately, yet existing metrics report only a single overall score. To address these gaps, we present 3DGS-IEval-15K+, a large-scale, multi-dimensional IQA dataset for compressed 3DGS, comprising 15,200 images from 10 diverse scenes, produced by 6 representative 3DGS algorithms at systematically designed compression levels and rendered from 20 strategically selected viewpoints spanning both training views and challenging novel views, annotated with 45,600 mean opinion scores (MOSs) across overall, geometry, and color quality. Based on 3DGS-IEval-15K+, we propose 3DGSI-Assessor, an all-in-one 3DGS IQA framework that integrates global semantic and dimension-specific local features within a large multimodal model (LMM), predicting all three dimensions in a single forward pass. 3DGSI-Assessor achieves state-of-the-art performance on 3DGS-IEval-15K+, and exhibits competitive generalization on other NVS benchmarks. Dataset and code will be released at https://github.com/YukeXing/3DGSI-Assessor.
