Personality in a nonsocial context Dataset


We introduce a novel dataset for behavior understanding and personality recognition in a nonsocial context. Forty-six participants were recorded in an unconstrained indoor space, related to a smart home environment, performing six tasks resembling Activities of Daily Living (ADL) .  During the experiment, personality scores were collected using self-assessment questionnaire (BFI-10).

Tasks Description

1) Find keys of two cabinets present in the room.

2) Find an item, the experimenter hid beforehand.

3) Search for an item that was not in the room. 

4) memorize the content in all drawers of the two cabinets.

5) sit at the table to complete two questionnaires.

6) make tea and eat cookies.

No time limit nor know-how was given to participate in the experiment,  once every task was completed, participants were asked to exit the room to record the task completion through magnetic sensors positioned on the door.


This Dataset is published with the paper:

  • D, Dotti; M. Popa; S. Asteriadis, Behavior and Personality Analysis in a nonsocial context Dataset, 2018 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Salt Lake City, US, 22 June, 2018

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. The data is released for academic research only, and you must attribute this work by citing our paper.


For privacy reasons, we did not record RGB images. The recorded data contains: 

    • 3D joints coordinates (30 fps): joints
    • Depth images(6 fps): set1, set2, set3
    • Big 5 personality score collected with the BFI-10: bfi_score
    • README

Please fill the form to obtain the download link