Automated capacity management using artificial intelligence techniques

1. A computer-implemented method comprising:

obtaining image data pertaining to occupancy of a confined space;

determining a level of occupancy in the confined space and one or more types of entities occupying the confined space by processing the image data using a first set of one or more artificial intelligence techniques comprising at least a first machine learning model, wherein processing the image data using the first set of one or more artificial intelligence techniques comprising at least the first machine learning model comprises performing at least one object identification task and at least one instance segmentation task using a mask regional-convolutional neural network (R-CNN);

automatically determining one or more capacity management parameters with respect to the confined space by analyzing the determined level of occupancy and the one or more determined types of entities using a second set of one or more artificial intelligence techniques comprising at least a second machine learning model; and

performing one or more automated actions based at least in part on the one or more determined capacity management parameters;

wherein the method is performed by at least one processing device comprising a processor coupled to a memory.

2. The computer-implemented method of claim 1 , wherein processing the image data using the mask R-CNN comprises:

identifying one or more areas within the image data likely to contain one or more entities by scanning the image data using at least one feature pyramid network;

generating one or more bounding boxes within the image data based at least in part on the one or more identified areas; and

generating at least one mask at pixel-level within at least a portion of the one or more identified areas.

3. The computer-implemented method of claim 2 , wherein the at least one feature pyramid network comprises information pertaining to three-dimensional geometry and information pertaining to one or more colors.

4. The computer-implemented method of claim 2 , wherein the image data comprise three-dimensional image data, and wherein using the mask R-CNN comprises linking one or more voxels in the three-dimensional image data to at least one class label.

5. The computer-implemented method of claim 1 , wherein the second machine learning model comprises at least one random forest model, and wherein analyzing the determined level of occupancy and the one or more determined types of entities using the at least one random forest model comprises performing one of more classifications across multiple capacity management parameters using multiple decision trees corresponding thereto.

6. The computer-implemented method of claim 1 , wherein the one or more capacity management parameters comprises at least one parameter pertaining to entity entry into the confined space.

7. The computer-implemented method of claim 1 , wherein the one or more capacity management parameters comprises at least one parameter pertaining to energy utilization in the confined space.

8. The computer-implemented method of claim 1 , wherein performing the one or more automated actions comprises automatically outputting instructions to at least one capacity management-related controller associated with the confined space.

9. The computer-implemented method of claim 8 , wherein the at least one capacity management-related controller comprises at least one Internet of Things device which controls one or more variables within the confined space.

10. The computer-implemented method of claim 1 , wherein performing the one or more automated actions comprises automatically notifying at least one user with the one or more determined capacity management parameters.

11. A non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes the at least one processing device:

to obtain image data pertaining to occupancy of a confined space;

to determine a level of occupancy in the confined space and one or more types of entities occupying the confined space by processing the image data using a first set of one or more artificial intelligence techniques comprising at least a first machine learning model, wherein processing the image data using the first set of one or more artificial intelligence techniques comprising at least the first machine learning model comprises performing at least one object identification task and at least one instance segmentation task using a mask R-CNN;

to automatically determine one or more capacity management parameters with respect to the confined space by analyzing the determined level of occupancy and the one or more determined types of entities using a second set of one or more artificial intelligence techniques comprising at least a second machine learning model; and

to perform one or more automated actions based at least in part on the one or more determined capacity management parameters.

12. The non-transitory processor-readable storage medium of claim 11 , wherein processing the image data using the mask R-CNN comprises:

identifying one or more areas within the image data likely to contain one or more entities by scanning the image data using at least one feature pyramid network;

generating one or more bounding boxes within the image data based at least in part on the one or more identified areas; and

generating at least one mask at pixel-level within at least a portion of the one or more identified areas.

13. The non-transitory processor-readable storage medium of claim 12 , wherein the at least one feature pyramid network comprises information pertaining to three-dimensional geometry and information pertaining to one or more colors.

14. The non-transitory processor-readable storage medium of claim 11 , wherein the second machine learning model comprises at least one random forest model, and wherein analyzing the determined level of occupancy and the one or more determined types of entities using the at least one random forest model comprises performing one of more classifications across multiple capacity management parameters using multiple decision trees corresponding thereto.

15. An apparatus comprising:

at least one processing device comprising a processor coupled to a memory;

the at least one processing device being configured:

to obtain image data pertaining to occupancy of a confined space;

to determine a level of occupancy in the confined space and one or more types of entities occupying the confined space by processing the image data using a first set of one or more artificial intelligence techniques comprising at least a first machine learning model, wherein processing the image data using the first set of one or more artificial intelligence techniques comprising at least the first machine learning model comprises performing at least one object identification task and at least one instance segmentation task using a mask R-CNN;

to automatically determine one or more capacity management parameters with respect to the confined space by analyzing the determined level of occupancy and the one or more determined types of entities using a second set of one or more artificial intelligence techniques comprising at least a second machine learning model; and

to perform one or more automated actions based at least in part on the one or more determined capacity management parameters.

16. The apparatus of claim 15 , wherein processing the image data using the mask R-CNN comprises:

identifying one or more areas within the image data likely to contain one or more entities by scanning the image data using at least one feature pyramid network;

generating one or more bounding boxes within the image data based at least in part on the one or more identified areas; and

generating at least one mask at pixel-level within at least a portion of the one or more identified areas.

17. The apparatus of claim 15 , wherein the second machine learning model comprises at least one random forest model, and wherein analyzing the determined level of occupancy and the one or more determined types of entities using the at least one random forest model comprises performing one of more classifications across multiple capacity management parameters using multiple decision trees corresponding thereto.

18. The apparatus of claim 15 , wherein performing the one or more automated actions comprises automatically outputting instructions to at least one capacity management-related controller associated with the confined space.

19. The apparatus of claim 15 , wherein the at least one capacity management-related controller comprises at least one Internet of Things device which controls one or more variables within the confined space.

20. The apparatus of claim 15 , wherein performing the one or more automated actions comprises automatically notifying at least one user with the one or more determined capacity management parameters.