The AFLW dataset you downloaded in the code was not AFLW dataset.

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

The AFLW dataset you downloaded in the code was not AFLW dataset. #1

Open mariolew opened 8 years ago

mariolew commented 8 years ago

It's the dataset Yi Sun used in CVPR 13, which is easier to get high performance since most of the faces are frontal.

ishay2b commented 8 years ago

Thanks @mariolew, this is my proxy to the implementation Yue Wu and Tal Hassner used in the article and probably differs slightly in details, but I will recheck it soon when possible.