Efficient Object Detection Using Embedded Binarized Neural Networks

Memory performance is a key bottleneck for deep learning systems. Binarization of both activations and weights is one promising approach that can best scale to realize the highest energy efficient system using the lowest possible precision. In this paper, we utilize and analyze the binarized neural...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	Journal of signal processing systems 2018-06, Vol.90 (6), p.877-890
Hauptverfasser:	Kung, Jaeha, Zhang, David, van der Wal, Gooitzen, Chai, Sek, Mukhopadhyay, Saibal
Format:	Artikel
Sprache:	eng
Schlagworte:	Circuits and Systems Computation Computer Imaging Computer memory Electrical Engineering Energy consumption Engineering Floating point arithmetic Image detection Image Processing and Computer Vision Infrared analysis Infrared imagery Machine learning Neural networks Object recognition Pattern Recognition Pattern Recognition and Graphics Signal,Image and Speech Processing Vision
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Schreiben Sie den ersten Kommentar!