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

Data augmentation #7

Closed rexnxiaobai closed 8 years ago

rexnxiaobai commented 8 years ago

Dear @ishay2b , sorry to trouble you again, two issues here.

  1. may you tell me how to do the data augmentation, if there are 100 images in one cluster,and each image has 5 landmarks. it seems the transformation increase the number to x_(x-1)/2, 100_99/2=4950, less than 5000....
  2. And also what tool did you use to compute H(non-reflective similarity transform), are there some python tools? may i use "affine" to replace it?