Face pose correction in videoconferencing using RGBD cameras

• Grant number: 15/26951-7
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: April 01, 2016
• End date: December 31, 2016
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: Carlos Hitoshi Morimoto
• Grantee: Rafael Pereira Lima
• Host Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

Visual stumili provided in videoconference enriches distance communication, but the user's experience is much different from a face-to-face contact. One of the limiting factors is the lack of eye contact, because the camera must be placed at a position distinct from the display showing the face.In this undergraduate research project, we propose the study of computer vision techniques to estimate the face pose using a RGBD camera and to generate a new video with the corrected pose to make the interaction more natural. To do that, the candidate will study the Intel RealSense technology to detect and estimate face poses in RGBD images, and he will also study computer graphics techniques to model faces and render them in a new orientation.