TY - CONF
T1 - Low Resource Lip Finding and Tracking Algorithm for Embedded Devices
AU - Guitarte Pérez, Jesús F.
AU - Lukas, Klaus
AU - Frangi, Alejandro F.
N1 - Funding Information:
The authors wish to express special thanks to Markus Simon, Ewald Frensch, Nikos Paragios, Visvanathan Ramesh and Eduardo Lleida for their good advises and interesting discussions. Siemens AG has supported the work of J. F. Guitarte under a PhD. contract. The work of A. F. Frangi was supported by Grants TIC2002-04495-C02 and FIT-070000-2002-935 of Spanish Ministry of Science and Technology and a Ramón y Cajal Research Fellowship.
Publisher Copyright:
© 2003 AVSP 2003 - International Conference on Audio-Visual Speech Processing. All rights reserved.
PY - 2003
Y1 - 2003
N2 - One of the best challenges in Lip Reading is to apply this technology in embedded devices. In current solutions the high use of resources, especially in reference to visual processing, makes the implementation and integration into a small device very difficult. In this article a new and efficient algorithm for detection and tracking of lips is presented. Lip Finding and Tracking are customary first steps in visual processing for Lip Reading. In our approach Lips are found among a small number of blobs, which should fulfill geometric constraints. The proposed algorithm runs on an ARM920T embedded device using on average less than 4 MHz1 (2,7% of CPU load). This algorithm shows promising results in a realistic environment accomplishing successful lip finding and tracking in 94.2% of more than 4900 image frames.
AB - One of the best challenges in Lip Reading is to apply this technology in embedded devices. In current solutions the high use of resources, especially in reference to visual processing, makes the implementation and integration into a small device very difficult. In this article a new and efficient algorithm for detection and tracking of lips is presented. Lip Finding and Tracking are customary first steps in visual processing for Lip Reading. In our approach Lips are found among a small number of blobs, which should fulfill geometric constraints. The proposed algorithm runs on an ARM920T embedded device using on average less than 4 MHz1 (2,7% of CPU load). This algorithm shows promising results in a realistic environment accomplishing successful lip finding and tracking in 94.2% of more than 4900 image frames.
UR - http://www.scopus.com/inward/record.url?scp=79953722699&partnerID=8YFLogxK
M3 - Paper
AN - SCOPUS:79953722699
SP - 111
EP - 116
T2 - 2003 International Conference on Audio-Visual Speech Processing, AVSP 2003
Y2 - 4 September 2003 through 7 September 2003
ER -