摘要 |
This orientation estimation device is provided with a processor. The processor inputs an image to be analysed, and sets, with respect to the inputted image to be analysed, a plurality of reference positions including a head position and a waist position of a person. Furthermore, candidate areas for part areas in the image to be analysed are determined on the basis of the plurality of set reference positions, and a joint base link model in which the orientation of the person is defined by the arrangement of the plurality of part areas and a plurality of point positions including the head position and the waist position. Moreover, it is determined whether the person included in the image to be analysed has the orientation on the basis of image features of the determined candidate areas, and part image features, i.e. image features of the part areas in images of the person. |