Using Wavelets to Analyze Similarities in Image-Classification Datasets

TL;DR

This paper introduces a fast, wavelet-based method to analyze similarities in image classification datasets, providing insights into dataset structure and model generalization without requiring pre-trained models.

Contribution

The authors develop a novel, efficient approach using wavelet decomposition for dataset analysis that bypasses the need for pre-trained models, enabling rapid similarity detection.

Findings

Similar images can be identified in seconds

Insights into dataset structure can inform model generalization

Method corroborates previous CNN-based similarity analyses

Abstract

Deep learning image classifiers usually rely on huge training sets and their training process can be described as learning the similarities and differences among training images. But, images in large training sets are not usually studied from this perspective and fine-level similarities and differences among images is usually overlooked. This is due to lack of fast and efficient computational methods to analyze the contents of these datasets. Some studies aim to identify the influential and redundant training images, but such methods require a model that is already trained on the entire training set. Here, using image processing and numerical analysis tools we develop a practical and fast method to analyze the similarities in image classification datasets. We show that such analysis can provide valuable insights about the datasets and the classification task at hand, prior to training a model. Our method uses wavelet decomposition of images and other numerical analysis tools, with no need for a pre-trained model. Interestingly, the results we obtain corroborate the previous results in the literature that analyzed the similarities using pre-trained CNNs. We show that similar images in standard datasets (such as CIFAR) can be identified in a few seconds, a significant speed-up compared to alternative methods in the literature. By removing the computational speed obstacle, it becomes practical to gain new insights about the contents of datasets and the models trained on them. We show that similarities between training and testing images may provide insights about the generalization of models. Finally, we investigate the similarities between images in relation to decision boundaries of a trained model.