How to demarcate the "Roll","Yaw"and "Pitch" in the training set?

qiexing / face-landmark-localization

cnn network predict face landmarks (68 points) and head pose (3d pose, yaw,roll,pitch).

410 stars 184 forks source link

How to demarcate the "Roll","Yaw"and "Pitch" in the training set? #2

Closed kylinly closed 8 years ago

kylinly commented 8 years ago

How to demarcate the "Roll","Yaw"and "Pitch" in the training set? How many samples in your training set? Thank you!

qiexing commented 8 years ago

It's responsible for 3d head pose, you can google to know the difference between. I use the data set from ibug300w. About 3800+ image for training .

kylinly commented 8 years ago

Thank you for the reply. The image set of "ibug300w" gives 68 key points of the face, but not offer the angle of "Roll,Yaw, and Pitch".So, How to get the three angles from "ibug300w" and using the three angles for training?

kylinly commented 8 years ago

Closed?

Benuraj commented 8 years ago

@kylinly were you able to figure out how to get angles with landmark points?