EVENT-CLASS

Description

EVENT-CLASS is a dataset of high-quality 360-degree videos with associated characteristics in the context of tele-education.

The sequences (video and audio) have been captured at the Universidad Politécnica de Madrid considering several environments (e.g., conventional classroom, laboratory, auditorium), lighting conditions, acquisition perspectives, and cameras (RICOH Theta V, Vuze VR, Samsung Gear 360, and Insta360), enriching the dataset.

The dataset is publicly available for research purposes, containing the videos and their characteristics (e.g., resolution,framerate, bitrate, semantic descriptions, spatial and temporalinformation, etc.).

More materials will be added soon: more videos, ground-truth annotations for training andtesting machine-learning algorithms for person detection, etc.

Applications:

EVENT-CLASS will be helpful for numerous applications related to tele-education, including quality assessment tests, and with the aim of improving the immersive experience of remote users thanks to the detection of relevant events that happen in the class (e.g., students raising their hands, changes of the slides in a presentation, person entering the room, etc.).

Database Generation:

• Number of videos: more than 50 (more to be added).
• Technical properties:

Download 

Click here to download the database (2.8 GB).

Citation

Please, cite the following paper in your publications if you use this database in your work: 

M. Orduna, J. Gutiérrez, C. Manzano, D. Ruiz, J. Cabrera, C. Díaz, P. Pérez, N. García, "EVENT-CLASS: Dataset of events in the classroom", 13th Int. Conf. on Quality of Multimedia Experience, QoMEX 2021, Montreal, Canada, pp. 1-4, 14-17 June 2021.

For questions about this database, please contact Marta Orduna at moc@gti.ssr.upm.es.