GAN を用いた生成AI によるデータ拡張手法

抄録

Recently, generative AI is proposed and promising for many areas, such as movie making and image recognition. By using a generative AI model, we generate images and use them as training data to improve both recognition accuracy and generalization performance. In this study, we conduct a comparative evaluation with conventional data augmentation methods to verify the effectiveness of using generated images as part of the training dataset.