Systems and methods for reducing point cloud and texture data using adapted splatting techniques
An optimization system reduces the data encoded within a point cloud for streaming and/or rendering of a lossy representation of the point cloud. The optimization system generates a first optimized splat for a first visual characteristic of the point cloud by replacing a first set of points having a first common value for the first visual characteristic with a first replacement primitive, and generates a second optimized splat for a second visual characteristic of the point cloud by replacing a second set of points having a second common value for the second visual characteristic with a second replacement primitive. The optimization system provides the first optimized splat and the second optimized splat instead of the original points of the point cloud in response to a request to access the point cloud.
1 . A method comprising:
receiving a three-dimensional (3D) model comprising a plurality of primitives with each primitive of the plurality of primitives being defined with a plurality of visual characteristics;
generating a first splat with a first set of splat primitives with each splat primitive of the first splat replacing a neighboring group of two or more primitives from the plurality of primitives that are defined with a common value for one or more first visual characteristics from the plurality of visual characteristics;
generating a second splat with a second set of splat primitives with each splat primitive of the second splat replacing a neighboring group of two or more primitives from the plurality of primitives that are defined with a common value for one or more second visual characteristics from the plurality of visual characteristics, wherein the second set of splat primitives are generated without modifying any of the first set of splat primitives; and
presenting a visualization of the 3D model by rendering the first set of splat primitives with the second set of splat primitives and combining rendered visual characteristics of the first set of splat primitives with rendered visual characteristics of the second set of splat primitives in the visualization.
2 . The method of claim 1 further comprising:
receiving optimization criteria for reducing a file size of the 3D model; and
setting a first range for the common value for the one or more first visual characteristics and a second range for the common value for the one or more second visual characteristics that result in the first set of splat primitives and the second set of splat primitives satisfying the optimization criteria.
3 . The method of claim 1 further comprising:
comparing the visualization of the 3D model produced from the first set of splat primitives and the second set of splat primitives against a visualization of the 3D model produced from rendering the plurality of primitives; and
wherein presenting the visualization of the 3D model comprises:
providing the first set of splat primitives and the second set of splat primitives to a requesting device in response to the visualization of the 3D model generated from the first and second sets of splat primitives having an acceptable amount of loss relative to the visualization of the 3D model produced from rendering the plurality of primitives; and
regenerating the first set of splat primitives and the second set of splat primitives to replace different groups of the plurality of primitives in response to the visualization of the 3D model generated from the first and second sets of splat primitives having more than the acceptable amount of loss relative to the visualization of the 3D model produced from rendering the plurality of primitives.
4 . The method of claim 1 further comprising:
analyzing variation in values defined for the plurality of visual characteristics of the plurality of primitives; and
wherein generating the second splat comprises generating fewer splat primitives for the second set of splat primitives than the first set of splat primitives in response to the variation in the values defined for the one or more first visual characteristics being greater than the variation in the values defined for the one or more second visual characteristics.
5 . The method of claim 1 further comprising:
selecting a first set of neighboring groups for replacement by the first set of splat primitives based on subsets of primitives from the plurality of primitives that are positioned next to one another and that have values defined for the one or more first visual characteristics that are within a specified threshold of one another; and
selecting a second set of neighboring groups for replacement by the second set of splat primitives based on subsets of primitives from the plurality of primitives that are positioned next to one another and that have values defined for the one or more second visual characteristics that are within a specified threshold of one another.
6 . The method of claim 1 further comprising:
interpolating values defined for the one or more first visual characteristics of the first set of splat primitives to the second set of splat primitives based on the second set of splat primitives being defined with more splat primitives than the first set of splat primitives; and
wherein presenting the visualization comprises rendering the second set of splat primitives based on values defined for the one or more second visual characteristics of the second set of splat primitives and interpolated values defined for the one or more first visual characteristics.
7 . The method of claim 1 further comprising:
performing a nearest neighbor matching between splat primitives of the first set of splat primitives and splat primitives of the second set of splat primitives;
mapping different visual characteristics defined for the first set of splat primitives and the second set of splat primitives to a single set of splat primitives based on the nearest neighbor matching; and
wherein presenting the visualization comprises rendering the single set of splat primitives.
8 . The method of claim 1 , wherein each primitive of the plurality of primitives is defined in a different 3D format than each splat primitive of the first and second sets of splat primitives.
9 . The method of claim 1 , wherein each splat primitive is an ellipsoid that approximates a shape formed by two or more primitives of the plurality of primitives that are replaced by that splat primitive.
10 . The method of claim 1 , wherein presenting the visualization comprises:
streaming at least one set of splat primitives derived from the first set of splat primitives and the second set of splat primitives in response to a request for the 3D model.
11 . The method of claim 1 , wherein each splat of the first set of splat primitives and the second set of splat primitives is defined with a different ellipsoidal shape, different positional coordinates, one or more radii, and values for one or more visual characteristics.
12 . The method of claim 1 , wherein presenting the visualization comprises:
generating a first visualization from rendering the first set of splat primitives;
generating a second visualization from rendering the second set of splat primitives; and
generating the visualization of the 3D model by combining the first visualization and the second visualization.
13 . A system comprising:
one or more hardware processors configured to:
receive a three-dimensional (3D) model comprising a plurality of primitives with each primitive of the plurality of primitives being defined with a plurality of visual characteristics;
generate a first splat with a first set of splat primitives with each splat primitive of the first splat replacing a neighboring group of two or more primitives from the plurality of primitives that are defined with a common value for one or more first visual characteristics from the plurality of visual characteristics;
generate a second splat with a second set of splat primitives with each splat primitive of the second splat replacing a neighboring group of two or more primitives from the plurality of primitives that are defined with a common value for one or more second visual characteristics from the plurality of visual characteristics, wherein the second set of splat primitives are generated without modifying any of the first set of splat primitives; and
present a visualization of the 3D model by rendering the first set of splat primitives with the second set of splat primitives and combining rendered visual characteristics of the first set of splat primitives with rendered visual characteristics of the second set of splat primitives in the visualization.
14 . The system of claim 13 , wherein the one or more hardware processors are further configured to:
receive optimization criteria for reducing a file size of the 3D model; and
set a first range for the common value for the one or more first visual characteristics and a second range for the common value for the one or more second visual characteristics that result in the first set of splat primitives and the second set of splat primitives satisfying the optimization criteria.
15 . The system of claim 13 , wherein the one or more hardware processors are further configured to:
compare the visualization of the 3D model produced from the first set of splat primitives and the second set of splat primitives against a visualization of the 3D model produced from rendering the plurality of primitives; and
wherein presenting the visualization of the 3D model comprises:
providing the first set of splat primitives and the second set of splat primitives to a requesting device in response to the visualization of the 3D model generated from the first and second sets of splat primitives having an acceptable amount of loss relative to the visualization of the 3D model produced from rendering the plurality of primitives; and
regenerating the first set of splat primitives and the second set of splat primitives to replace different groups of the plurality of primitives in response to the visualization of the 3D model generated from the first and second sets of splat primitives having more than the acceptable amount of loss relative to the visualization of the 3D model produced from rendering the plurality of primitives.
16 . The system of claim 13 , wherein the one or more hardware processors are further configured to:
analyze variation in values defined for the plurality of visual characteristics of the plurality of primitives; and
wherein generating the second set of splat comprises generating fewer splat primitives for the second set of splat primitives than the first set of splat primitives in response to the variation in the values defined for the one or more first visual characteristics being greater than the variation in the values defined for the one or more second visual characteristics.
17 . The system of claim 13 , wherein the one or more hardware processors are further configured to:
select a first set of neighboring groups for replacement by the first set of splat primitives based on subsets of primitives from the plurality of primitives that are positioned next to one another and that have values defined for the one or more first visual characteristics that are within a specified threshold of one another; and
select a second set of neighboring groups for replacement by the second set of splat primitives based on subsets of primitives from the plurality of primitives that are positioned next to one another and that have values defined for the one or more second visual characteristics that are within a specified threshold of one another.
18 . The system of claim 13 , wherein the one or more hardware processors are further configured to:
interpolate values defined for the one or more first visual characteristics of the first set of splat primitives to the second set of splat primitives based on the second set of splat primitives being defined with more splat primitives than the first set of splat primitives; and
wherein presenting the visualization comprises rendering the second set of splat primitives based on values defined for the one or more second visual characteristics of the second set of splat primitives and interpolated values defined for the one or more first visual characteristics.
19 . The system of claim 13 , wherein the one or more hardware processors are further configured to:
perform a nearest neighbor matching between splat primitives of the first set of splat primitives and splat primitives of the second set of splat primitives;
map different visual characteristics defined for the first set of splat primitives and the second set of splat primitives to a single set of splat primitives based on the nearest neighbor matching; and
wherein presenting the visualization comprises rendering the single set of splat primitives.
20 . A non-transitory computer-readable medium storing instructions that, when executed by one or more processors of a system, cause the system to perform operations comprising:
receiving a three-dimensional (3D) model comprising a plurality of primitives with each primitive of the plurality of primitives being defined with a plurality of visual characteristics;
generating a first splat with a first set of splat primitives with each splat primitive of the first splat replacing a neighboring group of two or more primitives from the plurality of primitives that are defined with a common value for one or more first visual characteristics from the plurality of visual characteristics;
generating a second splat with a second set of splat primitives with each splat primitive of the second splat replacing a neighboring group of two or more primitives from the plurality of primitives that are defined with a common value for one or more second visual characteristics from the plurality of visual characteristics, wherein the second set of splat primitives are generated without modifying any of the first set of splat primitives; and
presenting a visualization of the 3D model by rendering the first set of splat primitives with visual characteristics defined for the second set of splat primitives and combining rendered visual characteristics of the first set of splat primitives with rendered visual characteristics of the second set of splat primitives in the visualization.