SplatFormer
Point Transformer for Robust 3D Gaussian Splatting

1ETH Zürich
   2University of Maryland, College Park
   3ROCS, University Hospital Balgrist, University of Zürich

TL;DR

We analyze the performance of novel view synthesis methods in challenging out-of-distribution (OOD) camera views and introduce SplatFormer, a data-driven 3D transformer designed to refine 3D Gaussian splatting primitives for improved quality in extreme camera scenarios.

Overview Video



Check out our paper for more results and comparisons.

We also released our code and dataset here.

Citation 

@misc{chen2024splatformer,
    title={SplatFormer: Point Transformer for Robust 3D Gaussian Splatting}, 
    author={Yutong Chen and Marko Mihajlovic and Xiyi Chen and Yiming Wang and Sergey Prokudin and Siyu Tang},
    year={2024},
    eprint={2411.06390},
    archivePrefix={arXiv},
    primaryClass={cs.CV},
    url={https://arxiv.org/abs/2411.06390}, 
}

Funding

This study was conducted within the national Proficiency research project funded by the Swiss Innovation Agency Innosuisse in 2021 as one of 15 flagship initiatives.

Acknowledge

The website template was borrowed from Michaël Gharbi.