This is a multi-modal image + text dataset based on Open Images.
For reading the multi-modal data, refer to ./dataset.py
.
For accessing images, fetch images from the Open Images store. Authors of ImageD do not distribute nor are liable for
images listed in these indexes. For simpler access, refer to ./remote_images.py