Semantic-Guided Generative Image Augmentation Method with Diffusion Models for Image Classification

TL;DR

SGID is a novel image augmentation method using diffusion models guided by labels and captions to enhance image diversity while preserving semantic integrity, improving classification performance.

Contribution

This paper introduces SGID, a diffusion-based augmentation technique that balances image diversity and semantic consistency using label and caption guidance.

Findings

SGID outperforms baseline augmentation methods on multiple models and datasets.

SGID can be combined with other augmentations for further performance gains.

Quantitative and qualitative evaluations confirm semantic preservation and diversity.

Abstract

Existing image augmentation methods consist of two categories: perturbation-based methods and generative methods. Perturbation-based methods apply pre-defined perturbations to augment an original image, but only locally vary the image, thus lacking image diversity. In contrast, generative methods bring more image diversity in the augmented images but may not preserve semantic consistency, thus incorrectly changing the essential semantics of the original image. To balance image diversity and semantic consistency in augmented images, we propose SGID, a Semantic-guided Generative Image augmentation method with Diffusion models for image classification. Specifically, SGID employs diffusion models to generate augmented images with good image diversity. More importantly, SGID takes image labels and captions as guidance to maintain semantic consistency between the augmented and original images. Experimental results show that SGID outperforms the best augmentation baseline by 1.72% on ResNet-50 (from scratch), 0.33% on ViT (ImageNet-21k), and 0.14% on CLIP-ViT (LAION-2B). Moreover, SGID can be combined with other image augmentation baselines and further improves the overall performance. We demonstrate the semantic consistency and image diversity of SGID through quantitative human and automated evaluations, as well as qualitative case studies.