Author : Giridhar N R 1
Date of Publication :30th November 2017
Abstract: A meeting outcome is a desired result or product that will be achieved by the end of the meeting. It is also necessary to take note of key points made in the meeting. In this paper, we achieve this goal using an autonomous system. The entire system is built on both hardware and software components. The system is developed on an embedded board containing a video recording camera, audio sensors and a Dynamixel motor. The camera rotates in the direction the sound originates and starts to record. The recorded video is fed to the software part where face detection is done. Face detection is a computer technology that is used to identify human faces in digital images. We make use of three colour models like RGB, HSV and YCbCr to perform skin segmentation to obtain probable regions where the human face is present. We determine the best colour model which is suitable for skin segmentation. From the segmented image obtained from the previous step, Lucas-Kanade Optical Flow Method [6] is used in determining the location of the face in the frame.
Reference :
-
- Initializing and configuring the embedded board, http://https: // www.udoo.org/udoo-dual-and-quad
- Understanding the working of audio sensors, http://www.electronics-tutorials.ws/io/io_8.html
- Guidelines on usage and handling of the camera. http://www.logitech.com/en-us/product/hd-pro-webcamc920
- Bash shell scripting. http://https: //www .tutorialspoint.com/ unix/unix-what-is-shell.html
- Dynamixel information page, http://http: //www .robotis .us/ dynamixel
- Barron, J. L., D. J. Fleet, S. S. Beauchemin, and T. A. Burkitt. “Performance of optical flow techniques”. CVPR, 1992