Cross-modal images and 3D Gaussian Splats dataset for 10 household objects

DOI:10.4121/3eabd3f5-d814-48be-bbff-b440f2d48a2b.v1
The DOI displayed above is for this specific version of this dataset, which is currently the latest. Newer versions may be published in the future. For a link that will always point to the latest version, please use
DOI: 10.4121/3eabd3f5-d814-48be-bbff-b440f2d48a2b

Datacite citation style

Simionescu, Andrei (2024): Cross-modal images and 3D Gaussian Splats dataset for 10 household objects. Version 1. 4TU.ResearchData. dataset. https://doi.org/10.4121/3eabd3f5-d814-48be-bbff-b440f2d48a2b.v1
Other citation styles (APA, Harvard, MLA, Vancouver, Chicago, IEEE) available at Datacite

Dataset

A dataset consisting of 64 views and the resulting 3D Gaussian Splat taken around 10 household objects. The objects used correspond to the ModelNet10 dataset. The attached files contain the generated images using a Blender script, and the optimized 3D Gaussian point clouds for each model. The Blender scene, and python script used to generate the views are also attached.


This dataset was generated as part of the Bachelor Thesis project at TU Delft. The associated paper can be digitally accessed through the faculty's repository: https://repository.tudelft.nl, and the published code using this dataset can be found at: https://github.com/SimiOy/Self-Supervised-Learning-for-3DGS.

History

  • 2024-06-26 first online, published, posted

Publisher

4TU.ResearchData

Format

The rendered views are in the .png format, and the generated 3D Gaussian point clouds are in .ply format.

Organizations

TU Delft, Faculty of Electrical Engineering, Mathematics and Computer Science;
Princeton University

DATA

Files (12)