25/02/2020 Séminaire MLMDA
Lieu : campus de Cachan, C103, 12h
Learning to describe and capture local affine maps in images.
We present a CNN-based patch descriptor which captures affine invariance without the need for viewpoint simulations. We also present a CNN-based method for retrieving local affine maps that drastically improves the accuracy of geometry estimation between images when compared to the state of the art.