| 摘要 |
<P>PROBLEM TO BE SOLVED: To provide an operation input device with little possibility of erroneous recognition using a relatively large region free from restriction of positional relationship with a user for operation input. <P>SOLUTION: Distance image generating means 10 generates a moving image of a distance image, which has a distance represented by a pixel value to an object representing a hand. Object extracting means 2 extracts a region corresponding to the object from the distance image. Action extracting means 3 extracts a positional change of the object from the moving image of the distance image. The action extracting means 3 includes: a feature amount extraction section 31 that extracts a feature amount representing the position and the orientation of the object extracted by the object extracting means 2; and an operation amount determination section 32 that determines an operation amount instructing a control target based on a change amount of the feature amount accompanying the movement of the object in a period from a point when the feature amount satisfies predetermined start conditions to a point when the feature amount satisfies predetermined termination conditions. <P>COPYRIGHT: (C)2013,JPO&INPIT |
