BROWSE
Detail
Conference Paper
Refresh Enabled Video Analytics (REVA): Implications on Power and Performance of DRAM Supported Embedded Visual Systems
Cited 9 time in 
Share 
Share
- Authors
- Siddharth Advani, Nandhini Chandramoorthy, Karthik Swaminathan, Kevin Irick, Yong Cheol Peter Cho, Jack Sampson, Vijaykrishnan Narayanan
- Issue Date
- 2014-10
- Citation
- International Conference on Computer Design (ICCD) 2014, pp.501-504
- Language
- English
- Type
- Conference Paper
- 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.
- KSP Keywords
- Computational Power, DRAM Refresh, Design space, Google glasses, Memory Storage, Mobile system, Multi-object, Real-time video, Refresh Power, Scene Understanding, Video applications