Pelacakan dan Estimasi Pose Video Wajah 3 Dimensi

Resmana Lim, Davina , Silvia R




Abstract


The paper describes a face tracking and pose estimation system by using a webcamera. A 3D wireframe face model was used in conjunction with a feature tracking method of Lucas-Kanade. A set of face feature points was tracked in every video frame by using Lucas-Kanade method. A fully projective method was depployed for face pose estimation. This application was built using Microsoft Visual C 6.0®%2C Microsoft® DirectShow®%2C Intel Performance Library and Open Source Computer Vision (OpenCV) Library. The application has been implemented and able to track the face movement in real time (30 frames/second)%2C using webcam with the resolution of 320?240 pixel/frame on PC Pentium III/533 MHz. The system is prospective to be used for human-computer interaction applications as well as face expression recognition system.
Abstract in Bahasa Indonesia :

Paper ini membahas sistem pelacakan dan estimasi pose wajah dengan menggunakan webcamera. Model wajah wireframe 3D digunakan disini bersama dengan metode feature tracking Lucas-Kanade. Titik fiture wajah dilacak pada setiap frame video yang masuk dan pose wajah diestimasi menggunakan metode Fully projective. Sistem ini dibangun menggunakan Microsoft Visual C 6.0®%2C Microsoft® DirectShow®%2C Intel Image Processing dan Open Source Computer Vision Library. Aplikasi ini telah diimplementasikan dan dapat melacak pergerakan wajah secara real-time yaitu 30 frames/s dengan menggunakan webcam resolusi 320?240 pixel/frame pada PC Pentium III/533 MHz. Sistem cukup prospektif digunakan sebagai salah satu metode interaksi manusia komputer maupun pada sistem pengenalan ekspresi wajah. 3D face tracking%2C pose estimation%2C Lucas-Kanade tracking%2C human computer interaction.


Full Text: PDF

The Journal is published by The Institute of Research & Community Outreach - Petra Christian University. It available online supported by Directorate General of Higher Education - Ministry of National Education - Republic of Indonesia.

©All right reserved 2016.Jurnal Teknik Elektro, ISSN: 1411-870X

 

shopify traffic stats
View My Stats




Copyright © Research Center Web-Dev Team