Skip to main content
. Author manuscript; available in PMC: 2009 Oct 15.
Published in final edited form as: Int J Comput Vis. 2008 Feb 1;76(2):183–204. doi: 10.1007/s11263-007-0050-3

Fig. 8.

Fig. 8

An example of using our calibrated multi-view fitting algorithm to fit a single 2D+3D AAM to three simultaneously captured images of a face. Each image is overlaid with the corresponding 2D shape for that image in dark dots. The single 3D shape for the current triple of images is displayed in the top right of the center image. This 3D shape is also projected into each image using the corresponding Pn, and displayed as a white mesh. The single head pose (extracted from the rotation matrix R) is displayed in the top left of the center image as roll, pitch, and yaw. This should be compared with the algorithm in Sect. 3 in which there is a separate head pose for each camera. See the movie calib_fitting.mpg for the complete fitting sequence