1 code implementation • 15 Mar 2021 • Prem Nair, Rohan Doshi, Stefan Keselj
Convolutional neural networks use pooling and other downscaling operations to maintain translational invariance for detection of features, but in their architecture they do not explicitly maintain a representation of the locations of the features relative to each other.
no code implementations • 4 Apr 2019 • Eric Mitchell, Stefan Keselj, Sergiy Popovych, Davit Buniatyan, H. Sebastian Seung
We show that siamese encoding enables more accurate alignment than the image pyramids of SPyNet, a previous deep learning approach to coarse-to-fine alignment.