SAPPHIRE: An Always-on Context-aware Computer Vision System for Portable Devices

IEEE/ACM Conf. Design Automation and Test in Europe (DATE) |

Published by ACM - Association for Computing Machinery

Publication

Being aware of objects in the ambient provides a new dimension of context awareness. Towards this goal, we present a system that exploits powerful computer vision algorithms in the cloud by collecting data through always-on cameras on portable devices. To reduce comunication-energy costs, our system allows client devices to continually analyze streams of video and distill out frames that contain objects of interest. Through a dedicated image-classification engine SAPPHIRE, we show that if an object is found in 5% of all frames, we end up selecting 30% of them to be able to detect the object 90% of the time: 70% data reduction on the client device at a cost of 60mW of power (45nm ASIC). By doing so, we demonstrate system-level energy reductions of 2 . Thanks to multiple levels of pipelining and parallel vector-reduction stages, SAPPHIRE consumes only 3.0 mJ/frame and 38 pJ/OP – estimated to be lower by 11.4 than a 45 nm GPU – and a slightly higher level of peak performance (29 vs. 20 GFLOPS). Further, compared to a parallelized sofware implementation on a mobile CPU, it provides a processing speed up of up to 235 (1.81 s vs. 7.7 ms/frame), which is necessary to meet the real-time processing needs of an always-on context-aware system.