Semantic Content Filtering Using Self-Organizing Neural Networks

Abstract

Human figures synthesis has been one of the most complicated problems in computer graphics. In particular, facial animation is the most significant issue. Two decades have passed since Parke's pioneering work in animating faces (Parke, 1972). In the span of time, significant effort has been devoted to the development of computational models of human face for applications in diverse areas such as entertainment, low bandwidth teleconferencing, surgical facial planning and virtual reality. However, the task of accurately modeling the expressive human face by computer remains a major challenge. The main aspect of this research is to construct a new modeling and rendering technique. There will be brief analysis of IBMR (image based modeling and rendering) techniques for human face synthesis in computer graphics, research results obtained to date and future aims with discussions. The outcome of this research is to provide a faster, low cost and accurate 3D model that enables realistic facial expressions for animation purposes.