An RGB-D based social behavior interpretation system for a humanoid social robot

Aolfazl Zaraki; Manuel Giuliani; Maryam Banitalebi Dehkordi; Daniele Mazzei; Annamaria D'Ursi; Danilo De Rossi

doi:10.1109/ICRoM.2014.6990898

An RGB-D based social behavior interpretation system for a humanoid social robot

Aolfazl Zaraki, Manuel Giuliani, Maryam Banitalebi Dehkordi, Daniele Mazzei, Annamaria D'Ursi, Danilo De Rossi

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

6 Citations (Scopus)

Abstract

Humanoid social robots that interact with people need to be capable of interpreting the social behavior of their interaction partners in order to respond in a socially appropriate way. In this paper, we present a social behavior interpretation system that enables a humanoid robot to recognize human social behavior by analyzing communicative signals. The system receives the constructed RGB-D scene from a Kinect sensor, extracts information about body gesture and head pose from the scene using Microsoft Kinect SDK, and recognizes eight human social behaviors using a Hidden Markov Model (HMM). We trained the eight-state HMM with a corpus of 35 recorded human-human interaction scenes. The evaluation of the system shows a weighted average recognition rate of 81% for all states.

Original language	English
Title of host publication	2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	185-190
Number of pages	6
ISBN (Electronic)	9781479967438
DOIs	https://doi.org/10.1109/ICRoM.2014.6990898
Publication status	Published - 17 Dec 2014
Event	2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014 - Tehran, Iran, Islamic Republic of Duration: 15 Oct 2014 → 17 Oct 2014

Publication series

Name	2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014

Conference

Conference	2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014
Country/Territory	Iran, Islamic Republic of
City	Tehran
Period	15/10/14 → 17/10/14

Keywords

hidden Markov model
Human-robot interaction
humanlike robot
social behavior recognition

Access to Document

10.1109/ICRoM.2014.6990898

Cite this

Zaraki, A., Giuliani, M., Dehkordi, M. B., Mazzei, D., D'Ursi, A., & De Rossi, D. (2014). An RGB-D based social behavior interpretation system for a humanoid social robot. In 2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014 (pp. 185-190). Article 6990898 (2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014). Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/ICRoM.2014.6990898

Zaraki, Aolfazl ; Giuliani, Manuel ; Dehkordi, Maryam Banitalebi et al. / An RGB-D based social behavior interpretation system for a humanoid social robot. 2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014. Institute of Electrical and Electronics Engineers (IEEE), 2014. pp. 185-190 (2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014).

@inproceedings{741c8a10aaae4b6ab9991d5ebcc57177,

title = "An RGB-D based social behavior interpretation system for a humanoid social robot",

abstract = "Humanoid social robots that interact with people need to be capable of interpreting the social behavior of their interaction partners in order to respond in a socially appropriate way. In this paper, we present a social behavior interpretation system that enables a humanoid robot to recognize human social behavior by analyzing communicative signals. The system receives the constructed RGB-D scene from a Kinect sensor, extracts information about body gesture and head pose from the scene using Microsoft Kinect SDK, and recognizes eight human social behaviors using a Hidden Markov Model (HMM). We trained the eight-state HMM with a corpus of 35 recorded human-human interaction scenes. The evaluation of the system shows a weighted average recognition rate of 81% for all states.",

keywords = "hidden Markov model, Human-robot interaction, humanlike robot, social behavior recognition",

author = "Aolfazl Zaraki and Manuel Giuliani and Dehkordi, {Maryam Banitalebi} and Daniele Mazzei and Annamaria D'Ursi and {De Rossi}, Danilo",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE.; 2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014 ; Conference date: 15-10-2014 Through 17-10-2014",

year = "2014",

month = dec,

day = "17",

doi = "10.1109/ICRoM.2014.6990898",

language = "English",

series = "2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

pages = "185--190",

booktitle = "2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014",

address = "United States",

}

Zaraki, A, Giuliani, M, Dehkordi, MB, Mazzei, D, D'Ursi, A & De Rossi, D 2014, An RGB-D based social behavior interpretation system for a humanoid social robot. in 2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014., 6990898, 2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014, Institute of Electrical and Electronics Engineers (IEEE), pp. 185-190, 2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014, Tehran, Iran, Islamic Republic of, 15/10/14. https://doi.org/10.1109/ICRoM.2014.6990898

An RGB-D based social behavior interpretation system for a humanoid social robot. / Zaraki, Aolfazl; Giuliani, Manuel; Dehkordi, Maryam Banitalebi et al.
2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014. Institute of Electrical and Electronics Engineers (IEEE), 2014. p. 185-190 6990898 (2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - An RGB-D based social behavior interpretation system for a humanoid social robot

AU - Zaraki, Aolfazl

AU - Giuliani, Manuel

AU - Dehkordi, Maryam Banitalebi

AU - Mazzei, Daniele

AU - D'Ursi, Annamaria

AU - De Rossi, Danilo

PY - 2014/12/17

Y1 - 2014/12/17

N2 - Humanoid social robots that interact with people need to be capable of interpreting the social behavior of their interaction partners in order to respond in a socially appropriate way. In this paper, we present a social behavior interpretation system that enables a humanoid robot to recognize human social behavior by analyzing communicative signals. The system receives the constructed RGB-D scene from a Kinect sensor, extracts information about body gesture and head pose from the scene using Microsoft Kinect SDK, and recognizes eight human social behaviors using a Hidden Markov Model (HMM). We trained the eight-state HMM with a corpus of 35 recorded human-human interaction scenes. The evaluation of the system shows a weighted average recognition rate of 81% for all states.

AB - Humanoid social robots that interact with people need to be capable of interpreting the social behavior of their interaction partners in order to respond in a socially appropriate way. In this paper, we present a social behavior interpretation system that enables a humanoid robot to recognize human social behavior by analyzing communicative signals. The system receives the constructed RGB-D scene from a Kinect sensor, extracts information about body gesture and head pose from the scene using Microsoft Kinect SDK, and recognizes eight human social behaviors using a Hidden Markov Model (HMM). We trained the eight-state HMM with a corpus of 35 recorded human-human interaction scenes. The evaluation of the system shows a weighted average recognition rate of 81% for all states.

KW - hidden Markov model

KW - Human-robot interaction

KW - humanlike robot

KW - social behavior recognition

UR - http://www.scopus.com/inward/record.url?scp=84922618934&partnerID=8YFLogxK

U2 - 10.1109/ICRoM.2014.6990898

DO - 10.1109/ICRoM.2014.6990898

M3 - Conference contribution

AN - SCOPUS:84922618934

T3 - 2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014

SP - 185

EP - 190

BT - 2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - 2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014

Y2 - 15 October 2014 through 17 October 2014

ER -

Zaraki A, Giuliani M, Dehkordi MB, Mazzei D, D'Ursi A, De Rossi D. An RGB-D based social behavior interpretation system for a humanoid social robot. In 2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014. Institute of Electrical and Electronics Engineers (IEEE). 2014. p. 185-190. 6990898. (2014 2nd RSI/ISM International Conference on Robotics and Mechatronics, ICRoM 2014). doi: 10.1109/ICRoM.2014.6990898

An RGB-D based social behavior interpretation system for a humanoid social robot

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this