Hand shape recognition using a mean-shift embedded active contour (MEAC)

Authors:
Eun Yi Kim
Affiliations:
Dept. Of Internet and Multimedia Eng., Konkuk Univ., Seoul, Republic of Korea
Venue:
ICAT'06 Proceedings of the 16th international conference on Advances in Artificial Reality and Tele-Existence
Year:
2006

Citing 5
Cited 0

Region Competition: Unifying Snakes, Region Growing, and Bayes/MDL for Multiband Image Segmentation

IEEE Transactions on Pattern Analysis and Machine Intelligence
An HMM-Based Threshold Model Approach for Gesture Recognition

IEEE Transactions on Pattern Analysis and Machine Intelligence
Digital Image Processing

Digital Image Processing
Texture-Based Approach for Text Detection in Images Using Support Vector Machines and Continuously Adaptive Mean Shift Algorithm

IEEE Transactions on Pattern Analysis and Machine Intelligence
Active contours for tracking distributions

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a hand shape recognition system using an active contour model (ACM) and applies it to an HCI to control a mobile robot. For the recognition of hand shapes, the technique should be developed to accurately track variously changing hands in real-time. For this, we develop a mean-shift embedded active contour (MEAC) which can improve the convergence speed and the tracking accurracy than the standard ACM. The proposed recognition system consists of four modules: a hand detector, a hand tracker, a hand shape recognizer and a robot controller. The hand detector locates a skin color region with a specific shape as a hand in the first frame. Thereafter, the detected region is accurately tracked through the whole video sequence by the hand tracker using a MEAC, and its shape is recognized using Hue moments. To assess the validity of the proposed system, we tested the proposed system to a walking robot, RCB-1. The experimental results show the effectiveness of the proposed system.