Naslov
Effect of Data Augmentation Methods on Face Image Classification Results

Autori
Hrga, Ingrid ; Ivasic-Kos, Marina

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings of the 11th International Conference on Pattern Recognition Applications and Methods 2022 / - : SCITEPRESS, 2022, 660-667

ISBN
978-989-758-549-4

Skup
11th International Conference on Pattern Recognition Applications and Methods (ICPRAM)

Mjesto i datum
Beč, Austrija, 03.02.2022. - 05.02.2022

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Data Augmentation, Image Classification, Transfer Learning

Sažetak
Data augmentation encompasses a set of techniques to increase the size of a dataset artificially. Insufficient training data means that the network will be susceptible to the problem of overfitting, leading to a poor generalization capability of the network. Therefore, research efforts are focused on developing various augmentation strategies. Simple affine transformations are commonly used to expand a set. However, more advanced methods, such as information dropping or random mixing, are becoming increasingly popular. We analyze different data augmentation techniques suitable for the image classification task in this paper. We investigate how the choice of a particular approach affects the classification results depending on the size of the training dataset, the type of transfer learning applied, and the task's difficulty, which we determine based on the objectivity or subjectivity of the target attribute. Our results show that the choice of augmentation method becomes crucial in th e case of more challenging tasks, especially when using a pre-trained model as a feature extractor. Moreover, the methods that showed above-average results on smaller sets may not be the optimal choice on a larger set and vice versa.

Izvorni jezik
Engleski

Znanstvena područja
Računarstvo, Informacijske i komunikacijske znanosti

POVEZANOST RADA