Published May 24, 2024 | Version 1.0.3
Dataset Open

Data and code for 3D-ARM-Gaze: a public dataset of 3D Arm Reaching Movements with Gaze information in virtual reality

  • 1. Univ. Bordeaux, CNRS, INCIA, UMR 5287, F-33000 Bordeaux, France
  • 2. Univ. Poitiers, Univ. Tours, CNRS, CeRCA, F-86000 Poitiers, France
  • 3. Univ. La Rochelle, MIA, F-17000 La Rochelle, France
  • 4. Univ. de Bordeaux, CNRS, LaBRI, UMR 5800, F-33400, Talence, France

Description

This repository contains data and code for

Lento B., Segas E., Leconte V., Doat E., Danion F., Péteri R., Benois-Pineau J., de Rugy A. (2024). 3D-ARM-Gaze: a public dataset of 3D Arm Reaching Movements with Gaze information in virtual reality. doi:

It contains a dataset (DBAS22_DataOnline folder) of natural arm movements together with visual and gaze information when reaching objects in a wide reachable space from a precisely controlled, comfortably seated posture. More details could be find in the link publication (see Related identifiers section).

The DBAS22_DocOnline folder contains all the documentation files. The MainDataExplained file lists and describes the variables recorded during the experimental phases. In the SummaryOfFiles document, you will find descriptions for all the files within the DBAS22_DataOnline folder, and at the bottom, there is also a file tree that illustrates the file structure. The DBAS22FilesWorkflow document offers an overview of the workflow of experimental file creation during the experiment.

The DBAS22_CodeOnline folder contains all the scripts to perform data analysis, listed and described in the files CodeExplanations and DependenciesRelations. The GuideInstall file contains information needed to run the Python code files.

The DBAS22_CodeOnline folder also contains the DataPlayer Unity project. Instructions for running the project are provided in the DataPlayerGuide file and SupplementaryVideo2 (see Related identifiers section for more details). The folder DBAS22_DataPlayer_StandAloneApp contains the standalone version of the DataPlayer, which doesn't require any software installation.

The DBAS22_VideoOnline folder contains all the videos. 

Notes

This work was supported by a PhD grant from the French Direction Générale de l’Armement (DGA) awarded to B.L., and the ANR-PRCE grant I-Wrist (ANR-23-CE19-0031-01) awarded to A.d.R., R.P. and J.B.-P.

Technical info

Version 1.0.1:

  • Fix bug of target set visualization parameters not displayed in the DataPlayer Unity project

Version 1.0.2:

  • Correction of the Return Neutral Posture cubes not displayed for the head and the opposite shoulder
  • Correction of transform parent position according to children position that was reversed

Version 1.0.3:

  • Add folder containing videos

Other

Videos:

Video 1 : https://youtu.be/4ZRYN6ljeCQ
A participant performing the pick-and-place task with the Return to Neutral Posture (RNP) procedure between each pair of pick-and-place movements.

Video 2 : https://youtu.be/2lSEX_VLK4A
Tutorial on how to launch and use the DataPlayer to replay and visualize data in 3D.

Video 3 : https://youtu.be/RZhN5IR34_c
Examples of new video data generated with the DataPlayer by combining real data about arm and gaze with various fictitious visual backgrounds and objects.

Files

DBAS22_CodeOnline.zip

Files (5.0 GB)

Name Size Download all
md5:d459c3d0be8d56169e06f317486d81f7
33.4 MB Preview Download
md5:58c0b186d26ab2fd14e4ca1976b159db
4.7 GB Preview Download
md5:4df2896aaa9727e9992e1d7fafc54ab5
38.8 MB Preview Download
md5:2b290da9301b6210f0db82bcdca89e14
801.4 kB Preview Download
md5:61ee15647d37859ddd71eddefff05dee
176.0 MB Preview Download