Computer-implemented methods and system for localizing an object

A computer-implemented method of localizing an object, comprising:

• capturing a first image of the object from a first camera position;
• performing edge detection on the first image to form an edge-detected second image;
• performing a distance transform on the second image to form a distance transformed third image, wherein performing the distance transform comprises, for each non-edge pixel in the edge-detected second image, assigning a brightness value to a corresponding pixel in the third image that is equal to a distance from the non-edge pixel to a closest edge pixel in the edge-detected second image;
• positioning a virtual camera at a virtual camera position in virtual space relative to a virtual model of the object;
• repeating operations of:
• generating an edge-detected fourth image of the virtual model of the object based on the virtual camera position;
• generating an alignment cost of the edge-detected fourth image of the virtual model relative to the edge-detected second image of the object based on the distance transformed third image;
• and
• transforming a position of the virtual camera from the virtual camera position to a transformed virtual camera position having a lower alignment cost than the alignment cost of the edge-detected fourth image;
• until the alignment cost of the edge-detected fourth image relative to the edge-detected second image converges;
• and
• estimating a physical location of the object based on the transformed virtual camera position.

A system localizing an object, comprising:

• a processing circuit;
• and
• a memory circuit coupled to the processing circuit, wherein the memory circuit comprises computer-readable program instructions that, when executed by the processing circuit, cause the system to perform operations of:
• receiving a first image of the object from a first camera position;
• performing edge detection on the first image to form an edge-detected second image;
• performing a distance transform on the second image to form a distance transformed third image;
• positioning a virtual camera at a virtual camera position in virtual space relative to a virtual model of the object;
• generating an edge-detected fourth image of the virtual model of the object based on the virtual camera position;
• generating an alignment cost of the edge-detected fourth image of the virtual model relative to the edge-detected second image of the object based on the distance transformed third image;
• transforming a position of the virtual camera from the virtual camera position to a transformed virtual camera position having a lower alignment cost than the alignment cost of the edge-detected fourth image;
• and
• determining a physical location of the object based on the transformed virtual camera position.