Loading...
Loading...

Go to the content (press return)

3D CNNs on distance matrices for human action recognition

Author
Hernandez, A.; Porzi, L.; Rota, S.; Moreno-Noguer, F.
Type of activity
Presentation of work at congresses
Name of edition
25th ACM Conference on Multimedia
Date of publication
2017
Presentation's date
2017
Book of congress proceedings
Proceedings of the 25th ACM Conference on Multimedia
First page
1087
Last page
1095
DOI
https://doi.org/10.1145/3123266.3123299 Open in new window
Project funding
TIN2014-58178-R Instructing robots using natural communication skills National Project
Unit of Excellence María de Maeztu
Repository
http://hdl.handle.net/2117/114317 Open in new window
URL
https://dl.acm.org/citation.cfm?doid=3123266.3123299 Open in new window
Abstract
In this paper we are interested in recognizing human actions from sequences of 3D skeleton data. For this purpose we combine a 3D Convolutional Neural Network with body representations based on Euclidean Distance Matrices (EDMs), which have been recently shown to be very effective to capture the geometric structure of the human pose. One inherent limitation of the EDMs, however, is that they are defined up to a permutation of the skeleton joints, i.e., randomly shuffling the ordering of the join...
Citation
Hernandez, A., Porzi, L., Rota, S., Moreno-Noguer, F. 3D CNNs on distance matrices for human action recognition. A: ACM Conference on Multimedia Conference. "Proceedings of the 25th ACM Conference on Multimedia". Mountain view: 2017, p. 1087-1095.
Keywords
3d convolutional neural networks, activity recognition, deep learning, human action recognition, pattern recognition
Group of research
ROBiri - IRI Robotics Group

Participants