Template creation apparatus, object recognition processing apparatus, template creation method, and program

An object recognition processing apparatus that performs object recognition using a template, the apparatus comprising:

• an image acquisition unit configured to acquire an input image;
• a normal vector calculation unit configured to calculate a normal vector at a feature point in the input image;
• a normal vector quantization unit configured to quantize the calculated normal vector by mapping the calculated normal vector to a reference region on a plane orthogonal to an optical axis of a camera that acquired the input image, so as to acquire a quantized normal direction feature amount, the reference region including a central reference region corresponding to the vicinity of the optical axis and a reference region in the periphery of the central reference region;
• a template matching unit configured to search for a position of the object in the input image based on the template and the quantized normal direction feature amount acquired by the normal vector quantization unit, and acquire a collation result;
• and
• a recognition result output unit configured to output a recognition result that is based on the collation result.

A template creation method that is executed by a computer, comprising:

• acquiring three-dimensional data indicating a three-dimensional shape of an object that is a recognition target;
• calculating, based on the three-dimensional data, a normal vector at a feature point of the object viewed from a predetermined viewpoint that is set for the object;
• quantizing the calculated normal vector by mapping the calculated normal vector to a reference region on a plane orthogonal to an axis that passes though the viewpoint, so as to acquire a quantized normal direction feature amount, the reference region including a central reference region corresponding to the vicinity of the axis and a reference region in the periphery of the central reference region;
• creating, for each viewpoint, based on the acquired quantized normal direction feature amount, a template to be used for object recognition that is performed through template matching;
• and
• outputting the created template.

A non-transitory computer-readable recording medium storing a program for causing a computer to perform operations comprising:

• acquiring three-dimensional data indicating a three-dimensional shape of an object that is a recognition target;
• calculating, based on the three-dimensional data, a normal vector at a feature point of the object viewed from a predetermined viewpoint that is set for the object;
• quantizing the calculated normal vector by mapping the calculated normal vector to a reference region on a plane orthogonal to an axis that passes though the viewpoint, so as to acquire a quantized normal direction feature amount, the reference region including a central reference region corresponding to the vicinity of the axis and a reference region in the periphery of the central reference region;
• creating, for each viewpoint, based on the acquired quantized normal direction feature amount, a template to be used for object recognition that is performed through template matching;
• and
• outputting the created template.