O'Keefe, Simon and Villing, Rudi (2018) Evaluating Extended Pruning on Object Detection Neural Networks. In: 2018 29th Irish Signals and Systems Conference (ISSC). IEEE, pp. 1-6. ISBN 9781538660461

Preview

Download (788kB) | Preview

Abstract

CNNs are the state-of-the-art for many computer vision problems, including object detection. However, reducing the computational complexity of a CNN is a key prerequisite to deploying state-of-the-art deep learning networks in many low power embedded real-time robotic applications. Pruning has been shown to be an effective method to reduce the computational complexity of a Convolutional Neural Network (CNN) while maintaining accuracy. In the literature, accuracy lost through pruning is recovered with extended fine-tuning of the pruned network at the end of the pruning procedure, but further pruning is not conducted after extended fine-tuning. In this work we modify the pruning procedure to incorporate extended fine-tuning at intervals during the procedure to maintain network accuracy while pruning further than would otherwise be possible. We evaluate this procedure on a small scale custom object detection dataset and the more challenging standard PASCAL VOC dataset. On the former the new procedure achieves a 19.6× reduction in FLOPS for a drop of only 0.4% mean Average Precision (mAP) while the latter achieves only a 1.8× reduction in FLOPS for a drop of 0.8% mAP. The results indicate differing levels of parameter redundancy in the initial networks.

Item Type:	Book Section
Additional Information:	The authors would like to gratefully acknowledge funding provided by the Irish Research Council under their Government of Ireland Postgraduate Scholarship 2013. Cite as: S. O'Keeffe and R. Villing, "Evaluating Extended Pruning on Object Detection Neural Networks," 2018 29th Irish Signals and Systems Conference (ISSC), Belfast, 2018, pp. 1-6, doi: 10.1109/ISSC.2018.8585345.
Keywords:	Convolutional Neural Networks; pruning; real time; embedded; multi-class object detection;
Academic Unit:	Faculty of Science and Engineering > Electronic Engineering Faculty of Science and Engineering > Research Institutes > Hamilton Institute
Item ID:	13385
Identification Number:	https://doi.org/10.1109/ISSC.2018.8585345
Depositing User:	Rudi Villing
Date Deposited:	02 Oct 2020 16:58
Publisher:	IEEE
Refereed:	Yes
Funders:	Irish Research Council (IRC)
URI:	Publisher
Use Licence:	This item is available under a Creative Commons Attribution Non Commercial Share Alike Licence (CC BY-NC-SA). Details of this licence are available here

Repository Staff Only(login required)

Item control page

Downloads

Origin of downloads