This project, developing an instrument consisting of a large-scale tiled display, multiple video capture cameras, and a high-performance GPU cluster for large-scale data analysis and visualization, and multi-view coding and rendering, aims to enable a significant amount of research where graphics, vision, and multi-view video converge (including image-based modeling and rendering (IBMR) and MPEG-4/7). The instrument additionally includes augmentation of traditional interfaces that support research in 3D merger of graphics and video-coding, augmented reality and human perception research. It will also be used a to further advance the development of novel algorithms and techniques for large-scale data analysis and visualization, immersive virtual environments and image coding, as well as their on-going collaborations with scientists and researchers at national laboratories and industry. As evidenced by the availability of very large flat-panel displays and basic tracking systems such as Xbox Kinect, these types of coders, displays, and tracking systems should become more mature and affordable in the future.
Broader Impacts:
The proposed instrument can greatly impact the areas of telemedicine, manufacturing, and robotics. It can also be used for classes on computer graphics, virtual environments, data visualization, multi-media, and computer gaming. The PIs propose to organize summer programs to attract high-school students into science and STEM related areas.
Throughout this project, we have successfully constructed the infrastructure for multi-view, visualizations, and augmented reality, trained graduate and undergraduate students in the rare combination of high-resolution display, high-quality tracking, and high-performance computing. We have also successfully reached out to middle and high school students through display wall demos at summer youth programs and open house events, inspiring them to pursue visualization research in the future. The infrastructure has become a centerpiece for researchers across the campus to conduct interdisciplinary research and development. The project has enabled research in the areas of visualization, full-body motion tracking, and virtual reality which seeks to improve the overall usefulness of large display systems. The infrastructure has been built into the undergraduate course curriculum so that students gain hands-on experience and are capable of assisting with research projects that use the infrastructure. The computing cluster infrastructure has also enabled research and helped graduate student training in data intensive multimedia processing area. There have been several publications from this infrastructure and expected to be several more. Myounghoon Jeon, Michael Smith, James Walker, and Scott Kuhl. (2014). Constructing the Immersive Interactive Sonification Platform (iISoP). Distributed, Ambient and Pervasive Interactions (DAPI),. HCI International Conference. Jun Ma, James Walker, Chaoli Wang, Scott A. Kuhl, and Ching-Kuang Shene (2014). FlowTour: An Automatic Guide for Exploring Internal Flow Features. Proceedings of IEEE Pacific Visualization Symposium. Yokohama, Japan. Caoyang Jiang and Saeid Nooshabadi (2013). GPU Accelerated Motion and Disparity Estimations for Multiview Coding. Int Conf on Image Processing. Melbourne Australia. Yi Gu , Jun Ma , Chaoli Wang, David L. Kao, and Robert J. Nemiroff (2015). IGRAPH: Graph-Based Technique for Visual Analytics of Image and Text Collections. IS&T/SPIE Conference on Visualization and Data Analysis. San Francisco, CA. C. Jiang and S. Nooshabadi,. A scalable massively parallel motion and disparity estimation scheme for multiview video coding. IEEE Transactions on Circuits and Systems for Video Technology, UNDER REVIEW. Chaoli Wang, John P. Reese, Huan Zhang , Jun Tao , Yi Gu , Jun Ma , and Robert J. Nemiroff. (2014). Similarity-Based Visualization of Large Image Collections. Information Visualization. Information Visualization journal. ACCEPTED.
