Lip Detection Essay

844 WordsNov 6, 20114 Pages
Lip Detection in an Image EE 301 Course Project INTRODUCTION Lip reading is an active field that receives much attention from computer scientists. Its applications take part not only in science, such as a speech recognition system, but also in social activities, such as teaching pronunciation for deaf children in order to recover their speaking ability. In this paper, we aim to solve a narrower problem, the detection lip in a colour image, which is an essential step to provide visual lip data for the lip-reading system. MOTIVATION We took up this project so as to complete one of my earlier project titled “Monophonic to Stereophonic Video Conversion”. A video demo can be seen at http://home.iitk.ac.in/~rohitpd/Demo . In that project, one had to detect the coordinates of the sound source and depending upon its relative location on the screen adjusts the sound intensities in left and the right speakers. Assuming the input monophonic video contains only human voices, we had to look for the lips coordinates on the screen. Being only a week long project, I was not able to do lip detection part and assumed in the given video only there is motion at location of lips and wrote a motion detection code. Hence to complete that project we choose to do lips detection in EE301 course project. MAJOR STEP INVOLVED • • Face Detection Lip detection from the face Face detection Skin color is the key characteristic of human face in color image. We use equations (1-3) to filter out the effect of luminance in the RGB color space: r=R/(R+G+B) g=G/(R+G+B) b=B/(R+G+B) (1) (2) (3) The normalized colors can be effectively represented by using r and g values as b can be obtained by b=l-r-g. In skin color analysis, the color histogram based on r and g shows that the cluster of face color shows Gaussian distribution in the normalized color space. This means that color

More about Lip Detection Essay

Open Document