MVL Face Datasets

MVL Face Datasets are not your typical collection of datasets. Instead, this collection consists of image processing pipelines that generate masked face images from regular face images. Out current collection includes pipelines that work with Chicago Face Database and Karolinska Directed Emotional Faces. Below, you can see a comparison of original face images downloaded from Unsplah and the corresponding masked face images created using our pipeline. View on GitHub

MVL Random Stimulus Datasets

The MVL Random Stimulus Datasets are a collection of images and codes that allow for the generation of random shapes and textures. These datasets come in two types: splash shape and zebra blob, and each type has two datasets, categorical and continuous. The splash shapes are simple silhouettes whose contours are defined by different amplitudes at varying radial frequencies. The zebra blobs are made from band-pass filtered white noise patterns that have been windowed and saturated. The categorical datasets consist of images sampled randomly from an object space and then categorized into six subsets based on visual similarity, while the continuous datasets consist of images sampled continuously from a two-dimensional feature space. Thumbnail images from the categorical datasets and sample images from the continuous datasets are shown below for reference. View on GitHub