LIRIS-ACCEDE

the Annotated Creative Commons Emotional DatabasE for affective video content analysis



Request access

To acquire access to the database:

Please note, that any requests from free email addresses (hotmail, yahoo, gmail, etc.) will be refused. After submission of the form, it may take up to a week for you to receive the download link (you will receive a notification by email).

Introduction

In contrast to existing datasets with very few video resources and limited accessibility due to copyright constraints, LIRIS-ACCEDE consists of video excerpts with a large content diversity annotated along affective dimensions. All excerpts are shared under Creative Commons licenses and can thus be freely distributed without copyright issues. The dataset (the video clips, annotations, features and protocols) are publicly available.

LIRIS-ACCEDE is composed of three parts:

A complete description of the discrete section of the dataset can be found in the following journal paper:

Continuous annotations are described in the following publication:

The MediaEval 2015 affective impact of movies task is introduced in the following publication:

Other related publications are listed here.

Credits

The discrete and continuous sections of LIRIS-ACCEDE have been created by a french team of researchers:

Finally, we want to thank Léo Perrin who created the program generating the comparisons and collecting the data from CrowdFlower, Xingxian Li for his help on the modification of the GTrace program and we further would like to thank Ting Li who worked on the correlation between continuous affective ratings and physiological measurements. Of course, we also would like to thank all film-makers that shared their work under Creative Commons licenses.

The data for the MediaEval 2015 affective impact of movies task has been collected by:

Violence and affective classes could not have been collected without the effort from all the task organizers. Special thanks go to Bogdan Ionescu's team (University Politehnica of Bucharest, Romania), Hanli Wang's team (Tongji University, China), Vu Lam Quan's team (University of Science, VNU-HCMC, Vietnam), and Markus Schedl's team (Johannes Kepler University, Linz, Austria), who contributed a lot to the annotations, and of course Mats Sjöberg who organized... almost everything!