Optimal control for real-time visualization and 3D rendering using neural networks

This article describes a novel approach to the real-time visualization and 3D rendering using neural networks. The present scheme is within the general framework of approximate dynamic programming where optimal/suboptimal control is achieved through learning to use pulse-coupled neural networks. The neural network model is proposed in this paper for optimal control with control constraints. The resulting 3D models can then be viewed from any angle and subsequently processed to integrately match them against 3D model data stored in a synthetic database. Our work is unique in that it supports both an online and off-tine visualization and rendering of a distributed system.