| 摘要 |
<p><P>PROBLEM TO BE SOLVED: To realize a configuration which performs a user discrimination process with high precision through information analysis based on uncertain and asynchronous input information. <P>SOLUTION: Based on the image information and sound information obtained with a camera and a microphone, analysis information containing discrimination information as well as presence and position of a user in a real space is generated. For each of a plurality of targets corresponding to a virtual user, (1) target presence hypothesis information applied to calculating a presence probability of a target, (2) probability distribution information on a presence position of the target, and (3) user confirmation degree information representing who is the target are set. The target presence hypothesis information is applied to calculate the presence probability of each of the targets, for setting a target anew or deleting it. So, a wrongly generated target due to erroneous detection, for example, is reduced to attain a user discrimination process with high precision and high efficiency. <P>COPYRIGHT: (C)2010,JPO&INPIT</p> |
