Add spatial option which returns a spatial map of the perceptual difference (shape height x width) #6
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
added 3 commits
April 10, 2018 16:47
…he original scalar distance in the paper)
… spatial_factor (also mentioned in the example program test_network.py) to control spatial output shape and upsampling filter order. make the automatic spatial size be determined based on a multiplier (spatial_factor) of the max size of the CNN layer spatial extents.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This option is demonstrated in the test_network.py when spatial is set to True.
When the mean is taken of the spatial distance map, the resulting distance is the same as the scalar perceptual distance. However, the spatial distance map additionally allows one to visualize which spatial regions the network has determined are different.
The spatial distance map is created by using nearest neighbor upsampling on the individual layer responses. The upsampled size is the least common multiple of the associated layer sizes. This choice allows the mean of the spatial distance map to match the original scalar distance.