Model prediction
Examples of methods for model prediction are described herein. In some examples, a method includes predicting a compensated model. In some examples, the compensated model is predicted based on a three-dimensional (3D) object model. In some examples, a method includes predicting a deformed model. In some examples, the deformed mode is predicted based on the compensated model.
1. A method, comprising:
predicting a compensated model based on a three-dimensional (3D) object model having global values and local values for each of a plurality of edges, wherein the compensated model comprises of a training object model point cloud representing the 3D object model with modifications that compensate for a predicted manufacturing deformation, wherein the global values provide global information for simulating a thermal mass effect feature of the plurality of edges, and wherein the local values provide local neighborhood information for simulating a thermal diffusion effect feature of the plurality of edges;
predicting a deformed model based on convolution of features of the plurality of edges for the compensated model predicted;
comparing the deformed model to the 3D object model; and
based on comparing the deformed model to the 3D object model, adjusting one or more of the 3D object model, the compensation machine learning model, the compensated model, or a printing variable to reduce geometric inaccuracy in a 3D object to be printed.
2. The method of claim 1 , wherein the compensated model is predicted further based on an object model point cloud of the 3D object model.
3. The method of claim 1 , wherein predicting the compensated model is further based on a compensation machine learning model.
4. The method of claim 3 , wherein the compensation machine learning model is trained with a training scanned object point cloud as input and a training object model point cloud as output.
5. The method of claim 1 , wherein predicting the deformed model is further based on a deformation machine learning model.
6. The method of claim 5 , wherein the deformation machine learning model is trained with a training object model point cloud as input and a training scanned object point cloud as output.
7. The method of claim 1 , further comprising comparing the deformed model predicted from the compensated model with the 3D object model.
8. The method of claim 1 , further comprising:
printing a 3D object based on the compensated model; and
comparing a scanned object point cloud of the 3D object with an object model point cloud of the 3D object model and with a deformed point cloud of the deformed model.
9. The method of claim 8 , wherein printing the 3D object is performed in an online stage and predicting the deformed model is performed in an offline stage.
10. The method of claim 8 , further comprising presenting an image of the scanned object point cloud with the object model point cloud and the deformed point cloud.
11. An apparatus comprising:
a memory; and
a processor in electronic communication with the memory, wherein the processor is to:
based on a compensation machine learning model and a three-dimensional (3D) object model having global values and local values for each of a plurality of edges, predict a compensated model that comprises a training object model point cloud representing the 3D object model with modifications that compensate for a predicted manufacturing deformation, wherein the global values provide global information for simulating a thermal mass effect feature of the plurality of edges, and wherein the local values provide local neighborhood information for simulating a thermal diffusion effect feature of the plurality of edges;
predict a deformed model that comprises a deformed object model point cloud based on convolution of features of the plurality of edges for the compensated model that is predicted;
compare the deformed object model point cloud to a 3D object point cloud of the 3D object model; and
based on comparing the deformed object model point cloud to the 3D object point cloud of the 3D object model, adjust one or more of the 3D object model, the compensation machine learning model, the compensated model, or a printing variable to reduce geometric inaccuracy in a 3D object to be printed.
12. The apparatus of claim 11 , wherein the processor is to:
generate a graph by determining the plurality of edges for each point of the object model point cloud; and
determine an edge feature for each of the plurality of edges of the graph, wherein the compensated point cloud prediction is based on the edge features.
13. The apparatus of claim 12 , wherein the processor is to compare the deformed point cloud with the object model point cloud.
14. A non-transitory tangible computer-readable medium storing executable code, comprising:
code to cause a processor to:
based on a compensation machine learning model and a three-dimensional (3D) object model having global values and local values for each of a plurality of edges, predict a compensated model that comprises a training object model point cloud representing the 3D object model with modifications that compensate for a predicted manufacturing deformation, wherein the global values provide global information for simulating a thermal mass effect feature of the plurality of edges, and wherein the local values provide local neighborhood information for simulating a thermal diffusion effect feature of the plurality of edges;
predict a deformed model based on convolution of features of the plurality of edges for the compensated model predicted;
compare the deformed model to the 3D object model; and
based on comparing the deformed model to the 3D object model, adjust one or more of the 3D object model, the compensation machine learning model, the compensated model, or a printing variable to reduce geometric inaccuracy in a 3D object to be printed.
15. The computer-readable medium of claim 14 , further comprising code to cause the processor to compare a scanned object point cloud from the 3D object with the deformed point cloud and with the object model point cloud.