| CPC G06V 20/647 (2022.01) [G06T 7/73 (2017.01); G06T 17/205 (2013.01); G06T 19/20 (2013.01); G06V 10/44 (2022.01); G06V 10/7715 (2022.01); G06T 2210/36 (2013.01); G06T 2219/2012 (2013.01); G06T 2219/2016 (2013.01)] | 20 Claims |
|
1. A system for increasing a resolution of a three-dimension (3D) image, comprising:
a memory configured to store:
a baseline dataset that comprises:
a first set of feature points that are known for a first category of objects, wherein the first set of feature points indicates physical attributes that are common among the first category of objects; and
a color code associated with each of the first set of feature points;
a first 3D image of a first object, wherein the first object belongs to the first category of objects;
a processor, operably coupled to the memory, and configured to:
determine a set of contours from the first 3D image, wherein each of the set of contours represents a boundary around the first object in a different two-dimension (2D) plane;
determine a mesh image vector that indicates a set of location coordinates of a second set of feature points on a surface of the first object, wherein the second set of feature points indicates physical attributes of the first object;
for at least a first contour from among the set of contours:
compare the mesh image vector with the first contour;
determine an intersecting feature point where the mesh image vector meets the first contour;
determine that the baseline dataset comprises a first feature point that corresponds to the intersecting feature point;
in response to determining that the baseline dataset comprises the first feature point that corresponds to the intersecting feature point, generate a structural vector by populating the structural vector with the intersecting feature point;
determine a first color code associated with the intersecting feature point based at least in part upon determining that the first feature point is associated with the first color code;
generate a textural vector by populating the textural vector with the first color code;
generate an image vector of the first object by combining the structural vector with the textural vector;
access a test 3D image of a second object;
extract a set of features from the test 3D image, wherein the set of features represents physical attributes of the second object shown in the test 3D image;
compare the image vector of the first object to the extracted set of features; and
determine that the first object is the second object in response to determining that more than a threshold percentage of feature points of the image vector corresponds to counterpart features of the extracted set of features.
|