Method for manipulating 3D objects by flattened mesh
A computer-implemented method for generating a corresponding 3D mesh representing a 3D object includes transforming an initial three-dimensional mesh into a planar mesh, wherein each vertex or edge of the planar mesh is a transformation of a vertex or edge from the initial three-dimensional mesh; and sampling the planar mesh to generate a plurality of samples such that each sample comprises a three-dimensional coordinate representing a point in a three-dimensional space derived and/or taken directly from the initial three-dimensional mesh, and a coordinate representing a position of the sample relative to other samples; and generating the sampled matrix based on the plurality of samples; and representing the sampled matrix as a corresponding 3D mesh.
1 . A computer-implemented method for generating a dental crown restoration prediction, the method comprising:
training an autoencoder on a first set of training input data, the first set of training input data representative of an ideal dental restoration, the training being performed to identify a first set of latent variables and generate a first set of output data, where the autoencoder comprises a first encoder and a first decoder, wherein the first encoder converts the first set of
training input data into the first set of latent variables via a process involving encoding a sampled matrix of the first set of training input data into a vector and parameterizing an aspect of the dental restoration, wherein the first decoder decodes the parameterized aspect back into the sampled matrix by converting the first set of latent variables to the first set of output data, wherein the first set of output data has the same data structure as the first set of training input data;
training an hourglass predictor to return a second set of latent variables, wherein the hourglass predictor comprises a second encoder and the first decoder, wherein the second encoder converts a second set of training input data to the second set of latent variables, the second set of training input data representative of a dental surrounding associated with the dental restoration, the second set of latent variables having a comparable data format as the first set of latent variables, the second encoder converts the second set of training input data via a process involving encoding a sampled matrix of the second set of training data into a vector and sampling the vector to estimate the second set of latent variables, wherein the first decoder converts the second set of latent variables into a second set of output data having the same data structure as a set of training target data, the set of training target data including parameterized matchings between the ideal dental restoration and the dental surrounding that are decoded back into the sampled matrix, and wherein the second set of training input data is different from the first set of training input data;
using the hourglass predictor on a third set of input data to generate a third set of output data, the third set of input data includes newly obtained data representative of a dental surrounding associated with the dental restoration, wherein the third set of output data is in a comparable data format to that of the first set of output data, wherein the third set of output data comprises the dental restoration prediction for the dental surrounding associated with the third set of input data;
generating a three-dimensional (3D) mesh based on a matrix of the dental restoration prediction; and
converting the 3D mesh into a data format configured for manufacturing the dental restoration as a physical object.
2 . The method according to claim 1 , wherein the first set of training input data comprises sampled matrices of a plurality of crowns or teeth.
3 . The method according to claim 1 , wherein the first set of latent variables comprises crown or tooth parametrization.
4 . The method according to claim 1 , wherein the first set of output data comprises autoencoded matrices of crowns or teeth.
5 . The method according to claim 1 , wherein the second set of training input data comprises sampled matrices of matched dental surroundings.
6 . The method according to claim 1 , wherein the second set of latent variables comprises matched crown parametrization.
7 . The method according to claim 1 , wherein the second set of output data comprises matrices of matched crowns.
8 . The method according to claim 1 , wherein the training target data comprises sampled matrices of matched crowns.
9 . The method according to claim 1 , wherein the third set of input data comprises a sampled matrix of new dental surroundings.
10 . The method according to claim 1 , further comprising the steps of manufacturing a dental crown from the third set of output data by milling and 3D printing.
11 . The method according to claim 1 , further comprising the steps of manufacturing a dental crown from the third set of output data by milling or 3D printing.
12 . The method according to claim 1 , wherein the autoencoder is a variational autoencoder.
13 . The method according to claim 1 , wherein the first encoder and the first decoder are neural networks.
14 . The method according to previous claim 13 , wherein the neural networks are convolutional neural networks.
15 . The method according to claim 1 , wherein a data format of the first set of latent variables and the second set of latent variables is a vector of scalar numbers.
16 . The method according to claim 1 , wherein the first set of training input data comprises sampled matrices of a plurality of crowns and teeth.
17 . The method according to claim 1 , wherein the first set of latent variables comprises crown and tooth parametrization.
18 . The method according to claim 1 , wherein the first set of output data comprises autoencoded matrices of crowns and teeth.
19 . A computer program product in a non-transitory computer readable medium, wherein the computer program product includes computer readable program code executable by a hardware data processor to cause the hardware data processor to perform the method of claim 1 when said computer readable program code is executed by the hardware data processor.