Home // ACHI 2016, The Ninth International Conference on Advances in Computer-Human Interactions // View article

Recognizing Hand Gesture for Human-Robot Interaction

Authors:
Zuhair Zafar
Karsten Berns

Keywords: Human-Robot Interaction; Nonverbal communication; Hand Gestures; Bag-of-features approach.

Abstract:
Human-Robot Interaction is the most important aspect for the development of social service robots. Interacting with social robots via non-verbal communication makes the interaction natural and efficient for human. We present an interface that uses hand gestures to interact with humanoid robot. The major goal of this interface is to recognize gestures in dynamic environments with high accuracy and efficiency. Our proposed system enables automatic recognition of 18 different human hand gestures from RGB-D (color and depth data) device. Robot expresses different facial expressions and performs the gestures after recognizing them. We use bag-of-features approach to recognize gestures using scale invariant feature transform (SIFT) keypoints. The system is invariant to scale, slight rotation and illumination and can work in cluttered backgrounds. We use multi-class support vector machines for classification task. In order to validate our scheme, we use this interface in our humanoid robot that reports more than 94% recognition rate for 18 hand gestures.

Pages: 333 to 338

Copyright: Copyright (c) IARIA, 2016

Publication date: April 24, 2016

Published in: conference

ISSN: 2308-4138

ISBN: 978-1-61208-468-8

Location: Venice, Italy

Dates: from April 24, 2016 to April 28, 2016