Ben

Ok, thank you. I am actually first trying to replicate the training of the shape_predictor_68_face_landmarks.dat file, so that I can then add to it. I have downloaded the ibug dataset, and am running the training example with labels_ibug_300W.xml and the example train_shape_predictor_ex. It runs, and gives me sp.data, but the file size is only 19mb, compared to 95 for the shape_predictor_68_face_landmarks file. And gives a much worse / more shaky result when used. What am I missing to replicate this...

Hi, I have trained a shape detector with a large dataset of annotated faces. I now want to use the detected landmark positins to check for two expressions, smile and blink. I have two datasets, annotated in imglab and exported to xml, with a couple hundred images each for these two expressions. My question is, what now? Do I train two new shape predictors and load both ? Or is there another way to train and look for landmark positions? Is there an example of this kind of thing? Thank you!

Hi, thank you for your code. I am trying to use it for sensor fusion, fusing data...

...and as a follow up, where would I port in PCL clouds, or Matrices of points? I...

Hi, I am still having the compile error that I posted in help, but have a question......

Hi all, i just built 3DTK (windows, visual studio 2013), and am getting an error...

If you could give me a way to send my own matrix data in for base station positions...

ok, so it now builds. But when i try to test the multilateration, using: using namespace...