Abstract: The present disclosure provides methods, apparatuses, and computer-readable mediums for performing data augmentation. In some embodiments, a method of performing data augmentation by a device includes obtaining a plurality of images from a dataset. The method further includes computing, for each image of the plurality of images, a corresponding saliency map based on a gradient of a full loss function of that image. The method further includes selecting, from a subset of arrangements of a plurality of possible arrangements, a rearrangement offset that maximizes an overall saliency of a resulting image combining the plurality of images. The method further includes generating, using the rearrangement offset and a plurality of mixing ratios, a new mixed image from the plurality of images and a new mixed label from corresponding labels of the plurality of images. The method further includes augmenting the dataset with the new mixed image and the new mixed label.

Abstract: A projection system for projecting an image with an increased apparent resolution is provided. The projection system includes one or more projectors, a resampler module and a deconvolution module. The resampler module is configured to upsample an incoming high-resolution signal, perform an integer shift operation on a signal, and downsample to two or more low-resolution signals. The deconvolution module is configured to filter the upsampled high-resolution signal using a spatial domain deconvolution operation, the spatial domain deconvolution operation approximating frequency domain optical corrections based on characteristics of the one or more projectors. Preferably, the spatial domain deconvolution operation uses an N×N spatial kernel extracted from a spatial domain approximation of a Wiener filter.