A region extraction unit extracts a region indicating distance distribution of an attachment from distance image data acquired by a distance sensor by using a posture detected by a posture detection unit. A feature amount extraction unit extracts a feature amount of the attachment from the region indicating the distance distribution of the attachment extracted by the region extraction unit. An attachment recognition unit compares the feature amount extracted by the feature amount extraction unit with a feature amount of one or more reference attachments stored in a database to recognize the attachment.