US 12,243,150 B1
Systems and methods for reducing point cloud and texture data using adapted splatting techniques
Sean Looper, Flagstaff, AZ (US); and Thomas Matterson, Breaker Bay (NZ)
Assigned to Illuscio, Inc., Culver City, CA (US)
Filed by Illuscio, Inc., Culver City, CA (US)
Filed on Jun. 20, 2024, as Appl. No. 18/748,830.
Int. Cl. G06T 15/04 (2011.01); G06T 15/20 (2011.01); G06T 17/00 (2006.01)
CPC G06T 15/04 (2013.01) [G06T 15/20 (2013.01); G06T 17/00 (2013.01); G06T 2210/56 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A method comprising:
receiving a point cloud comprising a plurality of points that are distributed across a three-dimensional (3D) space and that generate a 3D model, wherein each point of the plurality of points is defined with a position in the 3D space and a plurality of visual characteristics that are presented at the position;
generating a first optimized splat for a first visual characteristic of the plurality of visual characteristics, wherein generating the first optimized splat comprises replacing a first set of points from the plurality of points with a first replacement primitive based on the first set of points being defined with a first common value for the first visual characteristic, and wherein the first replacement primitive is defined with a first shape that spans the position of each point from the first set of points and with the first common value for the first visual characteristic;
generating a second optimized splat for a second visual characteristic of the plurality of visual characteristics, wherein generating the second optimized splat comprises replacing a second set of points from the plurality of points with a second replacement primitive based on the second set of points being defined with a second common value for the second visual characteristic, and wherein the second replacement primitive is defined with a second shape that spans the position of each point from the second set of points and with the second common value for the second visual characteristic;
generating a third optimized splat for at least a third visual characteristic and a fourth visual characteristic of the plurality of visual characteristics, wherein generating the third optimized splat comprises replacing a third set of points from the plurality of points with a third replacement primitive based on the third set of points being defined with a third common value for the third visual characteristic and a fourth common value for the fourth visual characteristic, and wherein the third replacement primitive is defined with a third shape that spans the position of each point from the third set of points, with the third common value for the third visual characteristic, and with the fourth common value for the fourth visual characteristic; and
providing the first optimized splat, the second optimized splat, and the third optimized splat in response to a request to access the point cloud, wherein the first optimized splat, the second optimized splat, and the third optimized splat generate a lossy representation of the 3D model.