Anysize GAN: A solution to the image-warping problem
This addresses a common issue in deep learning for image generation, enabling training on datasets with varied image sizes without preprocessing, which is incremental as it builds on existing GAN structures.
The paper tackles the problem of GANs requiring uniform image sizes by proposing a novel architecture that generates images of any size without resizing, validated on the ISIC 2019 dataset to produce realistic images while preserving spatial and feature relationships.
We propose a new type of General Adversarial Network (GAN) to resolve a common issue with Deep Learning. We develop a novel architecture that can be applied to existing latent vector based GAN structures that allows them to generate on-the-fly images of any size. Existing GAN for image generation requires uniform images of matching dimensions. However, publicly available datasets, such as ImageNet contain thousands of different sizes. Resizing image causes deformations and changing the image data, whereas as our network does not require this preprocessing step. We make significant changes to the standard data loading techniques to enable any size image to be loaded for training. We also modify the network in two ways, by adding multiple inputs and a novel dynamic resizing layer. Finally we make adjustments to the discriminator to work on multiple resolutions. These changes can allow multiple resolution datasets to be trained on without any resizing, if memory allows. We validate our results on the ISIC 2019 skin lesion dataset. We demonstrate our method can successfully generate realistic images at different sizes without issue, preserving and understanding spatial relationships, while maintaining feature relationships. We will release the source codes upon paper acceptance.