3 datasets found

The rotMNIST dataset contains 28 × 28 grayscale images of digits from 0 to 9, randomly rotated by an angle uniformly distributed from 0 to 2π.

The paper proposes a decomposition of the convolutional filters over joint steerable bases across the space and the group geometry simultaneously, namely a rotation-equivariant...

The dataset used in the paper is not explicitly described. However, it is mentioned that the authors used the rotated MNIST, CIFAR-10, and PatchCamelyon datasets.