System and method for interactive image-based modeling of curved surfaces using single-view and multi-view feature curves

A computer-readable storage device storing program instructions executable on a computer to implement a 3D model generator configured to perform operations comprising:

• receiving images of an object captured from a plurality of different viewpoints of the object;
• receiving input that is considered to have been drawn onto features of the object and not onto portions of the images that are not part of the object, the input effective to derive a plurality of feature curves for the object, including one or more multi-view feature curves and two or more single-view feature curves, the one or more multi-view feature curves corresponding to said feature curves that are indicative of a shape of the object derived from the input that is considered to have been drawn onto the features of the object and not onto the portions of the images that are not part of the object in more than one of the images and the two or more single-view feature curves each corresponding to one said feature curve that is derived from the input that is considered to have been drawn onto the features of the object in a respective one of the images;
• generating one or more three-dimensional (3D) shape constraints for the object based on the one or more multi-view feature curves and independent of the two or more single view feature curves;
• generating an approximation of a 3D surface for the object based on the two or more single-view feature curves;
• and
• generating a 3D model of the object based on the one or more 3D shape constraints and the 3D surface approximation.

A system, comprising:

• a memory;
• and
• one or more processors coupled to the memory, wherein the memory stores program instructions executable by the one or more processors to implement a 3D model generator configured to perform operations comprising:
• receiving images of an object captured from a plurality of different viewpoints of the object;
• receiving input that is considered to have been drawn onto features of the object and not onto portions of the images that are not part of the object, the input effective to derive a plurality of feature curves for the object, including one or more multi-view feature curves and two or more single-view feature curves, the one or more multi-view feature curves corresponding to said feature curves that are indicative of a shape of the object derived from the input that is considered to have been drawn onto the features of the object and not onto the portions of the images that are not part of the object in more than one of the images and the two or more single-view feature curves each corresponding to one said feature curve that is derived from the input that is considered to have been drawn onto the features of the object in a respective one of the images;
• generating one or more three-dimensional (3D) shape constraints for the object from the one or more multi-view feature curves;
• generating, separately from generating the one or more 3D shape constraints for the object, an approximation of a 3D surface for the object from the two or more single-view feature curves and independent of the one or more multi-view feature curves;
• and
• generating a 3D model of the object based on the one or more 3D shape constraints and the 3D surface approximation.