Method and system for head tracking and pose estimation

摘要

Techniques for performing accurate and automatic head pose estimation are disclosed. According to one aspect of the techniques, head pose estimation is integrated with a scale-invariant head tracking method along with facial features detected from a located head in images. Thus the head pose estimation works efficiently even when there are large translational movements resulting from the head motion. Various computation techniques are used to optimize the process of estimation so that the head pose estimation can be applied to control one or more objects in a virtual environment and virtual character gaze control.

主权项

1. A method for determining pose of a head, the method comprising:
generating a sequence of images from a camera provided to look at a user, wherein the user is not required to wear any marker or any special device; estimating a position and a size of the head of the user in at least one image from a plurality of images generated from the camera, wherein the position of the head and the size of the head in a region of interest are further provided to get frontal facial features of the head or to define the region of interest, the frontal facial features include eye centers, nose, and mouth corners of the head, and a head pose tracker unit is configured to work with large and fast translation movements of the head, said estimating a position and a size of the head of the user is performed using a scale-invariant head tracking technique designed to scan the at least one image regardless of whatever scale the at least one image is in and further comprising:
tracking in the at least one image to determine whether the head is present or not in the at least one image;defining the region of interest when the head is detected in the at least one image;locating the position of the head in the at least one image; andmeasuring the size of the head in the region of interest; determining the pose of the head from the position and size of the head, wherein said determining the pose of the head from the position and size of the head is performed based on tracked facial features in the at least one image and projected corresponding features obtained from a 3D head model.