Computer Science Stack Exchange is a question and answer site for students, researchers and practitioners of computer science. It only takes a minute to sign up.
Sign up to join this community
I have a frame taken from a video. The frame contains a face and I have the (x, y) locations of the features (corners of lips, edge of eyebrows, etc.) and the headpose vector (pitch, yaw, roll), which shows the direction that the face is looking in degrees ((0, 0, 0) would be at the camera).
I need to calculate distances between specific points in real (3D) space. How can I map the feature locations to 3D space?
You don't have enough information to calculate 3D points for anything--all your information is relative right now. You need something that establishes a measurement, and some way to determine depth. If you know how far a point is from the camera (or can estimate it), then you can at least start calculating locations of features you know to be within the vertical plane containing that point. Or if you know how large a feature is, then you can calculate the sizes of features also assumed to be in the same vertical plane. Alternatively, if you can use information from other frames, then you can assume a rigid model for certain facial features and begin to make estimates, but it's still all scale-free until you can pin down a distance somewhere.
