search

ishay2b / VanillaCNN

Implementation of the Vanilla CNN described in the paper: Yue Wu and Tal Hassner, "Facial Landmark Detection with Tweaked Convolutional Neural Networks", arXiv preprint arXiv:1511.04031, 12 Nov. 2015. See project page for more information about this project. http://www.openu.ac.il/home/hassner/projects/tcnn_landmarks/ Written by Ishay Tubi : ishay2b [at] gmail [dot] com https://www.l
188 stars 80 forks source link

Mean and STD images #5

Closed fishman2008 closed 8 years ago

fishman2008 commented 8 years ago

Do you have the code to compute the mean and std images? I know that the mean can be computed by averaging the training images in general, but just not sure if the calculation will be the same as yours. It might go through some scale transformations?

If this question is duplicated, please delete it. Thanks.

ishay2b commented 8 years ago

Please pull latest commit, edit mainLoop.py:62 with: STEPS =['trainSetHD5', 'calcTrainSetMean'] and run mainLoop.py This will recalculate the mean and std and write it to: trainMean.png and trainSTD.png