DepthInSpace

Description

The dataset contains different 3D scenes figuring various objects captured using a 3D scanner Artec Eva and a Seres featuring a Belago 1.1.

Specifically, it contains a 3D mesh of each scene created by fusing together partial scans of the scenes obtained with the Artec Eva.

Secondly, it contains 150 frames (~10 seconds) of dots and flood image pairs of the same scenes with the Seres using structured light and the corresponding depth frames.

Finally, it contains the depth maps obtained by rendering the mesh mentioned earlier in the view of each Seres depth frame.

Reference

If you use this dataset, please cite the following publication:

@InProceedings{Johari_2021_ICCV,
author = {Johari, Mohammad Mahdi and Carta, Camilla and Fleuret, Fran\c{c}ois},
title = {DepthInSpace: Exploitation and Fusion of Multiple Video Frames for Structured-Light Depth Estimation},
booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
month = {October},
year = {2021},
pages = {6039-6048}
}

© 2021 ams International AG