Multimodal Datasets for Assessment of Quality of Experience in Immersive Multimedia

Dataset

Multimedia technologies aim at providing higher Quality of Experience (QoE), through combination of sensory, in particular audio and visual information.
 
The Sense of Presence (SoP), also called Immersiveness Levels (ILs) in these research work, is a desired quality metric for immersive environments. The SoP is expected to be highly correlated to the QoE. 
 
 
The Dataset1 and Dataset2 are multimodal datasets for the analysis of Quality of Experience (QoE) in emerging immersive multimedia applications.
Dataset1 was created to investigate the perceived SoP induced by one-minute long video stimuli with respect to content, quality, resolution, and sound reproduction.
Dataset2 was created to investigate the typical consumption of audiovisual contents by multimedia services end-user.
Detailed descriptions are provided in the following sections.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
These datasets comprises EEG, ECG, and respiration signals, as well as subjective ratings recorded during the stimuli.
 
   Dataset 1

This multimodal dataset for the analysis of Quality of Experience (QoE) in emerging immersive multimedia applications investigates the influence of the content, the resolution, the quality and the sound reproduction. In particular, the perceived Sense of Presence (SoP) induced by one-minute long video stimuli is explored in this study.

A UHD screen was used for the multimedia content consumption.

Annotated subjective scores and recorded physiological signals, including EEG, ECG, and respiration are stored in the folder Dataset1 of the ftp link below.

   Dataset 2

This multimodal dataset for the analysis of Quality of Experience (QoE) in emerging immersive multimedia applications investigates the typical experience of  multimedia services end-user. In particular, the perceived Sense of Presence (SoP), highly correlated with the QoE is explored in this study.

Three devices, namely an iPhone5, an iPad4 and a UHD screen, were used to induce the typical experiences.

Annotated subjective scores and recorded physiological signals, including EEG, ECG, and respiration are stored in the folder Dataset2 of the ftp link below.

   Permission is hereby granted, without written agreement and without license or royalty fees, to use, copy, modify, and distribute the data provided and its documentation for research purpose only. The data provided may not be commercially distributed. In no event shall the Ecole Polytechnique Fédérale de Lausanne (EPFL) be liable to any party for direct, indirect, special, incidental, or consequential damages arising out of the use of the data and its documentation. The Ecole Polytechnique Fédérale de Lausanne (EPFL) specifically disclaims any warranties. The data provided hereunder is on an “as is” basis and the Ecole Polytechnique Fédérale de Lausanne (EPFL) has no obligation to provide maintenance, support, updates, enhancements, or modifications.

 

If you use these database in your research we kindly ask you to reference the associated paper: 

Database1 : A.-F.  Perrin, H. Xu, E. Kroupi, M. Rerabek and T. Ebrahimi. Multimodal Dataset for Assessment of Quality of Experience in Immersive Multimedia. ACMMULTIMEDIA 2015 (ACM MM), Brisbane, Australia, October 26-30, 2015.

 Database2 : A.-F. Perrin, M. Rerabek and T. Ebrahimi. Towards prediction of Sense of Presence in immersive audiovisual communications. Human Vision and Electronic Imaging (HVEI) 2016, San Francisco, February 14-18, 2016.

Contact

 If you have any questions regarding this research please contact Anne-Flore Perrin (anne-flore.perrin (at) epfl.ch)

Download

   You can download all the dataset (EEG, ECG, respiration signals, subjective scores as well as complementary information) from the following ftp (please use dedicated FTP clients, such as fileZilla or FireFTP).

 

FTP address: ftp://tremplin.epfl.ch

User name: SoPMD@grebvm2.epfl.ch

Password: oREFH2SaQXlNsiJT

 

The total size of the Dataset1 is about 47.3 GB.

The total size of the Dataset2 is about 40.4 GB.