TY  - CONF
AU  - Eduardo Aguilar
AU  - Petia Radeva
A2  - CAIP
PY  - 2019//
TI  - Class-Conditional Data Augmentation Applied to Image Classification
T2  - LNCS
BT  - 18th International Conference on Computer Analysis of Images and Patterns
SP  - 182
EP  - 192
VL  - 11679
KW  - CNNs
KW  - Data augmentation
KW  - Deep learning
KW  - Epistemic uncertainty
KW  - Image classification
KW  - Food recognition
N2  - Image classification is widely researched in the literature, where models based on Convolutional Neural Networks (CNNs) have provided better results. When data is not enough, CNN models tend to be overfitted. To deal with this, often, traditional techniques of data augmentation are applied, such as: affine transformations, adjusting the color balance, among others. However, we argue that some techniques of data augmentation may be more appropriate for some of the classes. In order to select the techniques that work best for particular class, we propose to explore the epistemic uncertainty for the samples within each class. From our experiments, we can observe that when the data augmentation is applied class-conditionally, we improve the results in terms of accuracy and also reduce the overall epistemic uncertainty. To summarize, in this paper we propose a class-conditional data augmentation procedure that allows us to obtain better results and improve robustness of the classification in the face of model uncertainty.
UR  - https://doi.org/10.1007/978-3-030-29891-3_17
N1  - MILAB; no proj
ID  - Eduardo Aguilar2019
ER  -