Method for accelerating three-dimensional object segmentation with point cloud simplifications

1. A device, comprising:

a processing system including a processor; and

a memory that stores executable instructions that, when executed by the processing system, facilitate performance of operations, the operations comprising:

selecting model parameters for a machine-learning model, wherein the model parameters comprise: a number of feature planes for each layer of a plurality of layers, a style of submanifold block, whether to use optional submanifold blocks, or any combination thereof; and

using the machine-learning model to assign labels for segmentation and object identification to points in a downsampled point cloud, wherein the machine-learning model corresponds to a submanifold sparse convolutional network.

2. The device of claim 1 , wherein the downsampled point cloud is based on a downsampling of a full point cloud using grid simplification.

3. The device of claim 2 , wherein the downsampled point cloud comprises 20 to 60 percent of points in the full point cloud to reduce memory usage and computational overhead.

4. The device of claim 2 , wherein the downsampled point cloud comprises 63 to 87 percent of points in the full point cloud to improve segmentation accuracy.

5. The device of claim 1 , wherein the operations further comprise:

selecting a downsampling technique and a downsampling reduction percentage to reduce memory usage while meeting a required segmentation accuracy in obtaining the downsampled point cloud.

6. The device of claim 1 , wherein the operations further comprise:

selecting a downsampling technique and a downsampling reduction percentage to reduce computational overhead while meeting a required segmentation accuracy in obtaining the downsampled point cloud.

7. The device of claim 1 , wherein the processing system comprises a plurality of processors operating in a distributed computing environment, the plurality of processors including the processor.

8. The device of claim 1 , wherein the labels comprise a brand associated with an object, and wherein the brand is used for targeted advertising.

9. A non-transitory machine-readable medium, comprising executable instructions that, when executed by a processing system including a processor, facilitate performance of operations, the operations comprising:

selecting model parameters for a machine-learning model, wherein the model parameters comprise: a number of feature planes for each layer of a plurality of layers, a style of submanifold block, whether to use optional submanifold blocks, or any combination thereof, and wherein the machine-learning model corresponds to a submanifold sparse convolutional network; and

using the machine-learning model to assign labels to points in a downsampled point cloud obtained from a full point cloud.

10. The non-transitory machine-readable medium of claim 9 , wherein the downsampled point cloud is obtained using a downsampling technique, the downsampling technique including grid simplification.

11. The non-transitory machine-readable medium of claim 10 , wherein the operations further comprise:

selecting the downsampling technique and a downsampling reduction percentage to reduce memory usage while meeting a required segmentation accuracy.

12. The non-transitory machine-readable medium of claim 10 , wherein the operations further comprise:

selecting the downsampling technique and a downsampling reduction percentage to reduce computational overhead while meeting a required segmentation accuracy.

13. The non-transitory machine-readable medium of claim 9 , wherein the downsampled point cloud comprises 20 to 60 percent of points in the full point cloud to reduce memory usage and computational overhead.

14. The non-transitory machine-readable medium of claim 9 , wherein the downsampled point cloud comprises 63 to 87 percent of points in the full point cloud to improve segmenting accuracy.

15. The non-transitory machine-readable medium of claim 9 , wherein the model parameters comprise the number of feature planes for each layer, the style of submanifold block, and whether to use the optional submanifold blocks.

16. A method, comprising:

selecting, by a processing system including a processor, model parameters for a machine-learning model, wherein the model parameters comprise: a number of feature planes for each layer of a plurality of layers, a style of submanifold block, whether to use optional submanifold blocks, or any combination thereof, and wherein the machine-learning model corresponds to a submanifold sparse convolutional network; and

segmenting, by the processing system, a downsampled point cloud using the machine-learning model to assign labels to points in the downsampled point cloud.

17. The method of claim 16 , comprising:

selecting, by the processing system, a downsampling technique for obtaining the downsampled point cloud from a full point cloud.

18. The method of claim 17 , wherein the downsampling technique is applied in stages, the method further comprising:

receiving, by the processing system, a user input that identifies regions of the downsampled point cloud that should receive more detail in a first stage of the stages; and

enhancing, by the processing system and based on the receiving of the user input, detail of the regions in a second stage of the stages.

19. The method of claim 17 , comprising:

selecting, by the processing system, a downsampling reduction percentage to achieve a required segmentation accuracy in the downsampled point cloud relative to the full point cloud.

20. The method of claim 16 , comprising:

applying, by the processing system, parameter quantization to reduce a number of bits associated with the downsampled point cloud; and

applying, by the processing system, network pruning and sharing techniques to reduce a complexity of a network.