System and method for dynamically adjusting level of details of point clouds

1. A method comprising:

receiving point cloud data representing one or more three-dimensional objects;

tracking a viewpoint of a user relative to the point cloud data;

selecting a selected object from the one or more three-dimensional objects using the viewpoint;

for the selected object, obtaining a respective neural network model corresponding to the selected object from a plurality of available neural network models;

generating an increased level of detail for the selected object using the obtained respective neural network model; and

rendering a view of the point cloud data comprising the increased level of detail.

2. The method of claim 1 , wherein generating the increased level of detail for the selected object comprises hallucinating additional details for the selected object.

3. The method of claim 1 , wherein generating the increased level of detail for the selected object comprises using a neural network to infer details lost due to a limited sampling density for the selected object.

4. The method of claim 1 , further comprising replacing, within the point cloud data, points corresponding to the selected object with the increased level of detail.

5. The method of claim 4 , wherein the increased level of detail for the selected object has a greater resolution than the points replaced within the point cloud data.

6. The method of claim 5 , wherein selecting the selected object from the one or more three-dimensional objects using the viewpoint comprises responsive to determining that a point distance between the viewpoint and an object picked from the one or more three-dimensional objects is less than a threshold, selecting the object as the selected object.

7. The method of claim 1 , further comprising:

detecting, at a viewing client, one or more objects within the point cloud data,

wherein selecting the selected object comprises selecting the selected object from the one or more objects detected within the point cloud data.

8. The method of claim 1 , further comprising:

capturing, at a viewing client, data indicative of user movement; and

setting the viewpoint using, at least in part, the data indicative of user movement.

9. The method of claim 1 , further comprising:

capturing motion data of a head mounted display (HMD); and

setting the viewpoint using, at least in part, the motion data.

10. The method of claim 1 , wherein obtaining the respective neural network model corresponding to the selected object comprises responsive to determining a viewing client lacks the neural network model, retrieving, from a neural network server, the neural network model for the selected object.

11. The method of claim 1 , further comprising:

identifying, at a first server, a second server; and

transmitting, to a client device, an identification of the second server,

wherein obtaining the respective neural network model comprises retrieving the neural network model from the second server.

12. The method of claim 1 , wherein receiving point cloud data comprises receiving point cloud data from a sensor.

13. The method of claim 1 , wherein obtaining the respective neural network model comprises retrieving the neural network model for the selected object from a server.

14. The method of claim 1 , wherein obtaining the respective neural network model comprises retrieving the neural network model for the selected object from a local storage location of a client device.

15. The method of claim 1 , wherein obtaining the respective neural network model is based on information mapping respective objects to available neural network models.

16. The method of claim 15 , wherein information mapping respective objects to available neural network models comprises a respective object label indicating a respective object category with which the respective object most closely matches.

17. The method of claim 1 ,

wherein selecting the selected object selects an object from the one or more three-dimensional objects for which a corresponding level of detail is capable of being increased, and

wherein obtaining the respective neural network model corresponding to the selected object comprises determining whether the selected object corresponds to a new neural network model not present on a client device.

18. A method comprising:

for each object of a plurality of objects:

collecting at least one first resolution scan of the object;

producing a first resolution point cloud from the at least one first resolution scan of the object;

simulating a second resolution scan of the object based on the first resolution scan of the object, the second resolution scan having a lower resolution than the first resolution scan;

compiling a training set featuring a pair of the first and second resolution scans of the object; and

training a neural network model using the training set to hallucinate a higher level of detail from the second resolution scan of the object; and

storing, on a server, the respective neural network model for each object of the plurality of objects.

19. A system comprising:

one or more processors; and

one or more non-transitory computer-readable mediums storing instructions that are operative, when executed by the processor, to:

receive point cloud data representing one or more three-dimensional objects;

track a viewpoint of a user relative to the point cloud data;

select a selected object from the one or more three-dimensional objects using the viewpoint;

for the selected object, obtain a respective neural network model corresponding to the selected object from a plurality of available neural network models;

generate an increased level of detail for the selected object using the obtained respective neural network model; and

render a view of the point cloud data comprising the increased level of detail.

20. The method of claim 1 , wherein the increased level of detail comprises a level of detail data.