Kinect and Episodic Reasoning for Human Action Recognition
Computer Architecture and Networks Group, University of Castilla-La Mancha, Ciudad Real, Spain
Ruben.Cantarero@uclm.es
Computer Architecture and Networks Group, University of Castilla-La Mancha, Ciudad Real, Spain
MariaJose.Santofimia@uclm.es
Computer Architecture and Networks Group, University of Castilla-La Mancha, Ciudad Real, Spain
David.Villa@uclm.es
Computer Architecture and Networks Group, University of Castilla-La Mancha, Ciudad Real, Spain
JuanCarlos.Lopez@uclm.es
No disponemos de la información de contacto de este autor.
No disponemos de la información de contacto de este autor.
No disponemos de la información de contacto de este autor.
No disponemos de la información de contacto de este autor.
No disponemos de la información de contacto de este autor.
CONFERENCE — International Symposium on Distributed Computing and Artificial Intelligence
DATE — 01/06/2016 – 03/06/2016
PUBLISHER — Springer
YEAR — 2016
ISBN — 978-3-319-40162-1
DOI — 10.1007/978-3-319-40162-1_16
@inproceedings{cantarero2016kinect,
title={Kinect and episodic reasoning for human action recognition},
author={Cantarero, Ruben and Santofimia, Maria J and Villa, David and Requena, Roberto and Campos, Maria and Florez-Revuelta, Francisco and Nebel, Jean-Christophe and Martinez-del-Rincon, Jesus and Lopez, Juan C},
booktitle={Distributed Computing and Artificial Intelligence, 13th International Conference},
pages={147--154},
year={2016},
organization={Springer}
}
Abstract
This paper presents a method for rational behaviour recognition that combines vision-based pose estimation with knowledge modeling and reasoning. The proposed method consists of two stages. First, RGB-D images are used in the estimation of the body postures. Then, estimated actions are evaluated to verify that they make sense. This method requires rational behaviour to be exhibited. To comply with this requirement, this work proposes a rational RGB-D dataset with two types of sequences, some for training and some for testing. Preliminary results show the addition of knowledge modeling and reasoning leads to a significant increase of recognition accuracy when compared to a system based only on computer vision.