Apparatus and method for generating skeleton model using motion data and image data

摘要

An apparatus and method for generating a skeleton model using motion data and image data. The apparatus for generating the skeleton model may synchronize the image data and the motion data, and generate a three-dimensional (3D) skeleton model of an entire body of a user using a silhouette extracted from the image data and also using a position or orientation of a joint of the user extracted from the motion data. The skeleton model may be generated using the image data and the motion data, thereby improving accuracy of the skeleton model of the entire body of the user.

主权项

1. An apparatus for generating a skeleton model, the apparatus comprising:
a data synchronization unit to perform a synchronization upon image data for a user collected by an image sensor and motion data for the user collected by a motion sensor; a silhouette extraction unit to extract a silhouette of the user from the synchronized image data; a joint information calculation unit to calculate a position or orientation of a joint of the user from the synchronized motion data; and a skeleton model generation unit to generate a three-dimensional (3D) skeleton model of the user using the silhouette and the position or orientation of the joint of the user, wherein the skeleton model generation unit sets a reference point extracted from the silhouette and a position or orientation of the joint of the user included in the motion data, initializes positions or orientations of remaining joints of the user using the a position or orientation of the joint of the user included in the motion data, and determines positions or orientations of remaining joints of the user to be in a range of the silhouette.