Publication Type

Conference Proceeding Article

Version

publishedVersion

Publication Date

6-2018

Abstract

The emergence of augmented reality devices such as Google Glass and Microsoft Hololens has opened up a new class of vision sensing applications. Those applications often require the ability to continuously capture and analyze contextual information from video streams. They often adopt various deep learning algorithms such as convolutional neural networks (CNN) to achieve high recognition accuracy while facing severe challenges to run computationally intensive deep learning algorithms on resource-constrained mobile devices. In this paper, we propose and explore a new class of compression technique called D-Pruner to efficiently prune redundant parameters within a CNN model to run the model efficiently on mobile devices. D-Pruner removes redundancy by embedding a small additional network. This network evaluates the importance of filters and removes them during the fine-tuning phase to efficiently reduce the size of the model while maintaining the accuracy of the original model. We evaluated D-Pruner on various datasets such as CIFAR-10 and CIFAR-100 and showed that D-Pruner could reduce a significant amount of parameters up to 4.4 times on many existing models while maintaining accuracy drop less than 1%.

Keywords

Compression, Deep Learning, Continuous Vision

Discipline

Software Engineering

Research Areas

Software and Cyber-Physical Systems

Publication

EMDL'18: Proceedings of the 2nd International Workshop on Embedded and Mobile Deep Learning, Munich, Germany, June 15

First Page

7

Last Page

12

ISBN

9781450358446

Identifier

10.1145/3212725.3212730

Publisher

ACM

City or Country

New York

Additional URL

https://doi.org/10.1145/3212725.3212730

Download

Share

COinS