Title :
Refresh Enabled Video Analytics (REVA): Implications on power and performance of DRAM supported embedded visual systems
Author :
Advani, Siddharth ; Chandramoorthy, Nandhini ; Swaminathan, Karthik ; Irick, Kevin ; Cho, Yong Cheol Peter ; Sampson, Jack ; Narayanan, Vijaykrishnan
Author_Institution :
Pennsylvania State Univ., University Park, PA, USA
Abstract :
Video applications are becoming ubiquitous in mobile and embedded systems. Wearable video systems such as Google Glasses require capabilities for real-time video analytics and prolonged battery lifetimes. Further, the increasing resolution of image sensors in these mobile systems places an increasing demand on both the memory storage as well as the computational power. In this work, we present the Refresh Enabled Video Analytics (REVA) system, an embedded architecture for multi-object scene understanding and tackle the unique opportunities provided by real-time embedded video analytics applications for reducing the DRAM memory refresh energy. We compare our design with the existing design space and show savings of 88% in refresh power and 15% in total power, as compared to a standard DRAM refresh scheme.
Keywords :
DRAM chips; embedded systems; image sensors; low-power electronics; video signal processing; DRAM memory refresh energy; DRAM refresh scheme; Google Glasses; REVA system; battery lifetimes; computational power; embedded architecture; embedded systems; embedded video analytics applications; embedded visual systems; image sensors; memory storage; mobile systems; multiobject scene; refresh enabled video analytics; ubiquitous; video applications; wearable video systems; Computational modeling; Computer architecture; Object recognition; Random access memory; Standards; Streaming media; Visualization;
Conference_Titel :
Computer Design (ICCD), 2014 32nd IEEE International Conference on
Conference_Location :
Seoul
DOI :
10.1109/ICCD.2014.6974727
