Reproduce “3DGS-IEval-15K: A Large-scale Image Quality Evaluation Database for 3D Gaussian-Splatting” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yuke Xing, Jiarui Wang, Peizhi Niu, Wenjie Huang, Guangtao Zhai, Yiling Xu
The authors introduce 3DGS-IEval-15K, a human-rated benchmark for perceptual quality in compressed 3D Gaussian Splatting (3DGS), not a compression algorithm. It contains 15,200 images from 10 scenes, six algorithms, and 20 selected views, rated by 60 viewers. Test-view MOS peaks around 4–5, versus 5–6 for training views; challenging test-view selection limits this comparison.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.