Object alignment from a 2-dimensional image

A machine-readable non-transitory medium storing instructions for object alignment from a 2-dimensional (2-D) image of the object executable by a machine to:

• receive 3-dimensional (3-D) scan data obtained from scanning of the object with a scanning device;
• obtain a 3-D model of the object by determining a first set of points for a first 3-D training object and a second set of points for a second 3-D training object, wherein each of the second set of points correspond to a point of the first set of points;
• calculate an average between each of the corresponding points from the first and second set of points, wherein the average of the corresponding points include an x, y, z coordinate that corresponds to a distinguishing feature of the object;
• define a 2-D shape in the 2-D image and a 3-D shape in the 3-D model of the object wherein the 3-D model is derived from the 3-D scan data;
• map a number of points on the 2-D and 3-D shapes, wherein the points on the 2-D and 3-D shapes include the calculated average between each of the corresponding points;
• define the 2-D and 3-D shapes with a number of triangles, wherein a number of vertices of the number of triangles correspond to the calculated average between each of the corresponding points and each of the number of triangles includes a face;
• subdivide the number of triangles defining the 2-D and 3-D shapes into a plurality of subdivided triangles that include a plurality of new vertices, wherein each of the plurality of subdivided triangles includes a new face;
• assign a number of z-coordinates from the 3-D shape to the 2-D shape to create a 3-D reconstructured shape;
• and
• add a texture included within the 2-D shape to the 3-D reconstructured shape.

A system for object alignment from a 2-dimensional (2-D) image of the object:

• comprising
• a machine including processor resources;
• and
• memory resources associated with the machine, the memory resources storing machine-readable instructions that, when executed by the processor resources, cause the processor resources to:
• receive 3-dimensional (3-D) scan data obtained from scanning of the object with a scanning device;
• obtain a 3-D model of the object by determining a first set of points for a first 3-D training object and a second set of points for a second 3-D training object, wherein each of the second set of points correspond to a point of the first set of points;
• calculate an average between each of the corresponding points from the first and second set of points, wherein the average of the corresponding points include an x, y, z coordinate that corresponds to a distinguishing feature of the object;
• define a 2-D shape in the 2-D image and a 3-D shape in the 3-D model of the object, wherein the 3-D model is derived from the 3-D scan data, and wherein the object is a face of an individual;
• map the calculated average between each of the corresponding points on the 2-D and 3-D shapes;
• define the 2-D and 3-D shapes with a number of triangles, wherein a number of vertices of the number of triangles correspond to the calculated average between each of the corresponding points;
• subdivide the number of triangles defining the 2-D and 3-D shapes into a plurality of subdivided triangles that include a plurality of new vertices;
• determine whether there is a redundancy in the plurality of new vertices and remove new vertices of the plurality of vertices that are redundant;
• assign a number of z-coordinates from the 3-D shape to the 2-D shape to create a 3-D reconstructured shape;
• and
• add a texture corresponding to a point on the 2-D shape to the 3-D reconstructured shape.