Further Development of A Computational Vision Model for the Generation and Analysis of Photo-Realistic Scenes for Military and Civilian Virtual Reali

The further development of a Computational Vision Model (CVM) for the generation and analysis of photo-realistic scenes for military and civilian virtual reality applications is proposed. The proposed approach is to extend the existing CVM model, the TARDEC Vision Model (TVM). The main outputs of Phase I will be: 1) A specification of image fidelity needed (spatial resolution, temporal resolution, color definition) to simulate virtual reality for a selected set of military and civilian applications. 2) Identification of hardware required for real-time virtual reality implementation on a Silicon Graphics Workstation for a target acquisition application. 3) The demonstration of a baseline computational vision model on a Silicon Graphics Workstation for the generation of realistic military scenes for Army target acquisition systems. Phase II will expand on the Phase I effort to provide a flexible software tool that can be used for a broad range of military and civilian virtual reality applications. It will implement in hardware the capability to generate realistic visual imagery in real-time, using a conventional workstation. Phase II will provide for the demonstration and testing of the software for at least two applications, including automotive conspicuity and vehicle target acquisition device.