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

Input expected by caffe model #24

Open marcobarnobi opened 3 years ago

marcobarnobi commented 3 years ago

I am trying to use this caffe` model in a c++ framework, with OpenCV.dnn. To do this I need to know what is the input format expected by the model.

Right now what I do is resizing my faces to cv::Mat 40x40, subtracting mean, dividing by std. At this point I need to know what datatype the input should be, is it float32? Finally I need to know if the input is RGB or BGR.

Could you provide this information and let me know if I am missing/doing wrong some preprocessing step? Many thanks

@ishay2b