Mesh segmentation
Described herein is a method to segment meshes into sub-meshes based on triangle properties. The triangles are first classified according to some characteristic using their respective areas. A filtering process may change the classification according to the neighboring triangles. Then connected components are generated, and neighboring connected components are merged following a certain criteria. In some embodiments, connected components that share the most amount of edges are merged. With this technique, sub-meshes can be automatically generated without any previous knowledge of the mesh generation stage.
1 . A method programmed in a non-transitory memory of a device comprising:
classifying triangles within a mesh based on a triangle characteristic into a classification of a plurality of classifications, wherein the triangle characteristic is a triangle area, wherein the classification of the plurality of classifications comprises a first classification when the triangle area is equal to or greater than a threshold, and the classification of the plurality of classifications comprises a second classification when the triangle area is less than the threshold, wherein the threshold is generated using artificial intelligence and machine learning;
changing the classification of a triangle when the triangle is determined to be an isolated triangle, wherein the determination that the triangle is an isolated triangle is based on comparing the classification of the triangle with the classification of neighboring triangles, and if the classification of the triangle is different from the classification of all of the neighboring triangles, then the triangle is the isolated triangle;
generating connected components of the triangles that are connected and are in a same classification; and
merging neighboring connected components to generate a sub-mesh.
2 . The method of claim 1 further comprising expanding each sub-mesh by adding one or more triangles at a border of the sub-mesh such that the sub-meshes overlap.
3 . The method of claim 1 wherein the threshold is user-generated.
4 . The method of claim 1 wherein the threshold is adjusted per frame or per sequence.
5 . The method of claim 1 further comprising tracking the mesh and using mesh segmentation from a previous frame for a current frame.
6 . An apparatus comprising:
a non-transitory memory for storing an application, the application for:
classifying triangles within a mesh based on a triangle characteristic into a classification of a plurality of classifications, wherein the triangle characteristic is a triangle area, wherein the classification of the plurality of classifications comprises a first classification when the triangle area is equal to or greater than a threshold, and the classification of the plurality of classifications comprises a second classification when the triangle area is less than the threshold, wherein the threshold is generated using artificial intelligence and machine learning;
changing the classification of a triangle when the triangle is determined to be an isolated triangle, wherein the determination that the triangle is an isolated triangle is based on comparing the classification of the triangle with the classification of neighboring triangles, and if the classification of the triangle is different from the classification of all of the neighboring triangles, then the triangle is the isolated triangle;
generating connected components of the triangles that are connected and are in a same classification; and
merging neighboring connected components to generate a sub-mesh; and
a processor coupled to the memory, the processor configured for processing the application.
7 . The apparatus of claim 6 wherein the application is further configured for expanding each sub-mesh by adding one or more triangles at a border of the sub-mesh such that the sub-meshes overlap.
8 . The apparatus of claim 6 wherein the threshold is user-generated.
9 . The apparatus of claim 6 wherein the threshold is adjusted per frame or per sequence.
10 . The apparatus of claim 6 wherein the application is further configured for tracking the mesh and using mesh segmentation from a previous frame for a current frame.
11 . A system comprising:
an encoder configured for encoding content; and
a decoder configured for:
classifying triangles within a mesh based on a triangle characteristic into a classification of a plurality of classifications, wherein the triangle characteristic is a triangle area, wherein the classification of the plurality of classifications comprises a first classification when the triangle area is equal to or greater than a threshold, and the classification of the plurality of classifications comprises a second classification when the triangle area is less than the threshold, wherein the threshold is generated using artificial intelligence and machine learning;
changing the classification of a triangle when the triangle is determined to be an isolated triangle, wherein the determination that the triangle is an isolated triangle is based on comparing the classification of the triangle with the classification of neighboring triangles, and if the classification of the triangle is different from the classification of all of the neighboring triangles, then the triangle is the isolated triangle;
generating connected components of the triangles that are connected and are in a same classification; and
merging neighboring connected components to generate a sub-mesh.
12 . The system of claim 11 wherein the decoder is further configured for expanding each sub-mesh by adding one or more triangles at a border of the sub-mesh such that the sub-meshes overlap.
13 . The system of claim 11 wherein the threshold is user-generated.
14 . The system of claim 11 wherein the threshold is adjusted per frame or per sequence.
15 . The system of claim 11 wherein the decoder is further configured for tracking the mesh and using mesh segmentation from a previous frame for a current frame.