Synthetic Images for Fine Grain Texture Analysis

This dataset consists of 300 images, each in PNG format with a size of 256x256 pixels, designed for fine-grain texture analysis. It serves as a valuable resource for researchers and professionals working in the field of image quality assessment.

Current metrics such as SSIM and PSNR, while commonly used, often fail to capture key aspects of an image's texture properties. This dataset aims to facilitate the discovery of new texture-based quality evaluation metrics that are not correlated with existing metrics, enabling a more comprehensive assessment of image quality by incorporating overlooked texture characteristics.