Tutorial - Efficient "pixel-wise" deep learning on large images

[Andrew Janowczyk]

All,

I’ve written a blog post on how to perform efficient output generation for large images using Caffe using on the net surgery example as a starting point:

 

http://www.andrewjanowczyk.com/efficient-pixel-wise-deep-learning-on-large-images/

Its based off of a digital pathology application, but should be universally applicable, regardless of network architecture or application.

Cheers,

Andrew