Scheduled System Maintenance:
On May 6th, system maintenance will take place from 8:00 AM - 12:00 PM ET (12:00 - 16:00 UTC). During this time, there may be intermittent impact on performance. We apologize for the inconvenience.
By Topic

Max-pooling convolutional neural networks for vision-based hand gesture recognition

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

The purchase and pricing options are temporarily unavailable. Please try again later.
9 Author(s)
Nagi, J. ; Dalle Molle Inst. for Artificial Intell. (IDSIA), Univ. of Lugano & SUPSI, Manno-Lugano, Switzerland ; Ducatelle, F. ; Di Caro, G.A. ; Ciresan, D.
more authors

Automatic recognition of gestures using computer vision is important for many real-world applications such as sign language recognition and human-robot interaction (HRI). Our goal is a real-time hand gesture-based HRI interface for mobile robots. We use a state-of-the-art big and deep neural network (NN) combining convolution and max-pooling (MPCNN) for supervised feature learning and classification of hand gestures given by humans to mobile robots using colored gloves. The hand contour is retrieved by color segmentation, then smoothened by morphological image processing which eliminates noisy edges. Our big and deep MPCNN classifies 6 gesture classes with 96% accuracy, nearly three times better than the nearest competitor. Experiments with mobile robots using an ARM 11 533MHz processor achieve real-time gesture recognition performance.

Published in:

Signal and Image Processing Applications (ICSIPA), 2011 IEEE International Conference on

Date of Conference:

16-18 Nov. 2011