A team of researchers including Profs. Jia Deng (PI), Jason Mars, Kevin Pipe, Lingjia Tang, Thomas Wenisch and CSE Chair Marios Papaefthymiou have been awarded a $1.4M joint NSF and Intel grant for their research project, “Large-Scale Visual Recognition: From Cloud Data Centers to Wearable Devices”.
The researchers are finding a solution to implement state-of-the-art vision systems in wearable devices where there is little heat dissipation. They will do this by developing artificial intelligence systems that efficiently manage the resources most crucial for high-performance wearable-based visual recognition. The systems will put out computations that are thermally managed by materials in the wearable device that will melt during heaving heating and solidify between bursts.
The goal of the project is to enable encyclopedic, real-time visual recognition through seamless integration of visual computing on wearable devices and in the cloud. The researchers hope to create this wearable visual recognition system that captures live video input while providing intelligent, real-time assistance through automatic or on-demand visual recognition. They will address the challenges of creating such a systems through interdisciplinary approach integrating computer vision, hardware architecture, VLSI design, and heat transfer.
The main concepts of this research will be integrated in undergraduate and graduate coursework, and a demonstration system will be made available to the research community and used in educational modules for high school students. More information about the project is available in the grant posting by NSF.
Posted: December 3, 2015