NoiseCutMix: A Novel Data Augmentation Approach by Mixing Estimated Noise in Diffusion Models
This is an incremental improvement for computer vision researchers and practitioners seeking more natural and effective data augmentation in image classification tasks.
The authors tackled the problem of unnatural boundaries in multi-class image augmentation by proposing NoiseCutMix, which fuses estimated noise from two classes in diffusion models, achieving improved classification performance compared to conventional methods like CutMix and MixUp.
In this study, we propose a novel data augmentation method that introduces the concept of CutMix into the generation process of diffusion models, thereby exploiting both the ability of diffusion models to generate natural and high-resolution images and the characteristic of CutMix, which combines features from two classes to create diverse augmented data. Representative data augmentation methods for combining images from multiple classes include CutMix and MixUp. However, techniques like CutMix often result in unnatural boundaries between the two images due to contextual differences. Therefore, in this study, we propose a method, called NoiseCutMix, to achieve natural, high-resolution image generation featuring the fused characteristics of two classes by partially combining the estimated noise corresponding to two different classes in a diffusion model. In the classification experiments, we verified the effectiveness of the proposed method by comparing it with conventional data augmentation techniques that combine multiple classes, random image generation using Stable Diffusion, and combinations of these methods. Our codes are available at: https://github.com/shumpei-takezaki/NoiseCutMix