Data Augmentation by Selecting Mixed Classes Considering Distance Between Classes

TL;DR

This paper introduces a data augmentation technique that dynamically selects class pairs for mixing based on class probability distances, enhancing deep learning object recognition accuracy especially in long-tailed datasets.

Contribution

It proposes a novel class-distance-based data selection method for mixup augmentation that adapts during training to improve recognition performance.

Findings

Improves recognition accuracy on general datasets.

Enhances performance on long-tailed datasets.

Outperforms conventional mixup methods.

Abstract

Data augmentation is an essential technique for improving recognition accuracy in object recognition using deep learning. Methods that generate mixed data from multiple data sets, such as mixup, can acquire new diversity that is not included in the training data, and thus contribute significantly to accuracy improvement. However, since the data selected for mixing are randomly sampled throughout the training process, there are cases where appropriate classes or data are not selected. In this study, we propose a data augmentation method that calculates the distance between classes based on class probabilities and can select data from suitable classes to be mixed in the training process. Mixture data is dynamically adjusted according to the training trend of each class to facilitate training. The proposed method is applied in combination with conventional methods for generating mixed data. Evaluation experiments show that the proposed method improves recognition performance on general and long-tailed image recognition datasets.