Carnot Research (@CarnotResearch) header
Carnot Research (@CarnotResearch) profile picture
Joined February 2021
Photo by Carnot Research (@CarnotResearch)Photo by Carnot Research (@CarnotResearch)
Shapes can be described using a set of landmarks. A landmark is a point of correspondence on each object that matches between and within populations. - Dryden, 2016.
form=size+shape is the geometrical information that remains when location and rotational effects are removed from an object. - Dryden, 2016.
Shape is the geometric information that remains when location, scale and rotational effects are removed from an object. - Kendall, 1977
1x1 convolutional layers are great at dimensionality reduction. A peek inside:
A bird's eye view of VGG network with 11 trainable layers.