Improving Deep Learning with Generic Data Augmentation

Nitschke, Geoff and Taylor, Luke (2018) Improving Deep Learning with Generic Data Augmentation, Proceedings of IEEE Symposium Series on Computational Intelligence (IEEE SSCI 2018), Bengaluru, India.

PDF 2018-Improving_Deep_Learning_with_Generic_Data_Augmentation.pdf Download (436kB)

Abstract

Deep artificial neural networks require a large corpus of training data in order to effectively learn, where collection of such training data is often expensive and laborious. Data augmentation overcomes this issue by artificially inflating the training set with label preserving transformations. Recently there has been extensive use of generic data augmentation to improve Convolutional Neural Network (CNN) task performance. This study benchmarks various popular data augmentation schemes to allow researchers to make informed decisions as to which training methods are most appropriate for their data sets. Various geometric and photometric schemes are evaluated on a coarse grained data set using a relatively simple CNN. Experimental results, run using 4-fold cross-validation and reported in terms of Top-1 and Top-5 accuracy, indicate that cropping in geometric augmentation significantly increases CNN task performance.

Item Type:	Conference paper
Subjects:	Computing methodologies > Artificial intelligence
Date Deposited:	09 Nov 2018
Last Modified:	10 Oct 2019 15:31
URI:	http://pubs.cs.uct.ac.za/id/eprint/1281

Actions (login required)

View Item