Hardware Acceleration on Raspberry Pi3 Using OpenVINO Deployed on Intel Neural Compute Stick 2 for Face Recognition

  • Kollu Nimshi

Abstract

The main challenge of this research is to implement Deep Learning on Raspberry Pi3 and improve the Performance Metrics on Pi3 like: Average Prediction Time, Frames Per Second, and Accuracy , though Custom  Face Recognition Application can be implemented on powerful machines like CPU, GPU etc. , yet it is not the best solution as it consumes large size, cost, and more power. Thus, bringing this application into Embedded single board computers is very important. This led to the idea of using Intel Neural Compute Stick2 as an Edge Inference for accelerating the performance on Raspberry Pi3.

Published
2020-05-20
How to Cite
Kollu Nimshi. (2020). Hardware Acceleration on Raspberry Pi3 Using OpenVINO Deployed on Intel Neural Compute Stick 2 for Face Recognition. International Journal of Advanced Science and Technology, 29(9s), 4828 - 4842. Retrieved from https://sersc.org/journals/index.php/IJAST/article/view/17311