System and method for machine learning-driven object detection

1. A system for game object identification, comprising:

a non-transitory computer-readable memory storing computer-readable instructions; and

at least one processor configured to execute the computer-readable instructions to cause the system to execute,

a wager object region proposal network (RPN) to receive image data from captured images of a gaming table, and

a wager object detection network to receive an output of the wager object RPN,

wherein the wager object detection network is configured to detect at least one wager object in the captured images based on the output of the wager object RPN.

2. The system of claim 1 , wherein the at least one processor is configured to execute the computer-readable instructions to cause the system to execute,

a gaming table region proposal network (RPN) to receive the image data from the captured images of the gaming table, and

a gaming table object detection network to receive an output of the gaming table region RPN,

wherein the gaming table object detection network is configured to detect at least one gaming object in the captured images based on an output of the gaming table object detection network, wherein the at least one gaming object is different from the at least one wager object.

3. The system of claim 2 , wherein the at least one processor is configured to execute the computer-readable instructions to cause the system to,

identify at least one first region of interest in the captured images that relates to the at least one gaming object,

identify a subset of the at least one first region of interest that relates to a single stack of the at least one wager object,

identify at least one second region of interest that relates to a part of an edge pattern on each of the at least one wager object that forms part of the single stack in the identified subset;

identify a value pattern in each of the at least one second region of interest; and

estimate a total wager value of the single stack using the identified value pattern and a lookup table.

4. The system of claim 3 , wherein the at least one processor is configured to execute the computer-readable instructions to cause the system to,

associate each of the at least one first region of interest with a wager area identifier.

5. The system of claim 1 , wherein the at least one processor is configured to execute the computer-readable instructions to cause the system to,

detect an edge pattern of the at least one wager object,

detect an end of the detected edge pattern, and

determine a value of the at least one wager object based on the detected edge pattern, wherein the detected edge pattern is around a circumference of the at least one wager object.

6. The system of claim 5 , wherein the at least one processor is configured to execute the computer-readable instructions to cause the system to,

identify a region of interest based on the edge pattern, the region of interest bounding at least a part of the edge pattern.

7. A computing apparatus comprising:

at least one camera configured to capture images of a gaming surface;

a non-transitory computer-readable memory storing computer-readable instructions; and

at least one processor configured to execute the computer-readable instructions to cause the computing apparatus to,

apply machine learning processes to,

identify a region of interest in at least one of the captured images, the region of interest including at least one wager object,

detect an end of an edge pattern of the at least one wager object, and

determine a value of the at least one wager object based on the detected end.

8. The computing apparatus of claim 7 , wherein the at least one processor is configured to execute the computer-readable instructions to cause the computing apparatus to,

identify at least one first region of interest in the captured images that relates to at least one gaming object, the at least one gaming object being different from the at least one wager object

identify a subset of the at least one first region of interest that relates to a single stack of the at least one wager object,

identify at least one second region of interest that relates to a part of an edge pattern on each of the at least one wager object that forms part of the single stack in the identified subset;

identify a value pattern in each of the at least one second region of interest, the value pattern including the edge pattern; and

estimate a total wager value of the single stack using the identified value pattern and a lookup table.

9. The computing apparatus of claim 8 , wherein the at least one processor is configured to execute the computer-readable instructions to cause the computing apparatus to,

associate each of the at least one first region of interest with a wager area identifier.

10. The computing apparatus of claim 9 , wherein the at least one processor is configured to execute the computer-readable instructions to cause the computing apparatus to,

identify a start and end of a game based on a game start and end trigger configuration.

11. The computing apparatus of claim 7 , wherein the at least one processor is configured to execute the computer-readable instructions to cause the computing apparatus to,

detect a plurality of wager objects in the captured images, and

detect edge patterns of the plurality of wager objects, the detected edge patterns being around circumferences of the plurality of wager objects,

detect ends of the detected edge patterns, and

determine a value pattern based on the detected ends.

12. The computing apparatus of claim 11 , wherein the at least one processor is configured to execute the computer-readable instructions to cause the computing apparatus to,

identify a top wager object, and

identify a base region, the detected plurality of wager objects being in an area between and including the top wager object and the base region.

13. A computing apparatus comprising:

at least one camera configured to capture images of a gaming surface;

a non-transitory computer-readable memory storing computer-readable instructions; and

at least one processor configured to execute the computer-readable instructions to cause the computing apparatus to,

use a region proposal network (RPN) to identify a region of interest in at least one of the captured images,

use a wager object detection network to receive an output of the RPN and detect at least one wager object in the captured images based on the output of the RPN, and

determine a wager amount based on the identified region of interest and the detected at least one wager object.

14. The computing apparatus of claim 13 , wherein the at least one processor is configured to execute the computer-readable instructions to cause the computing apparatus to execute,

a wager object region proposal network (RPN) to receive image data from captured images of a gaming table, the wager object RPN being the RPN to identify the region of interest, and

the wager object detection network to receive an output of the wager object RPN,

wherein the wager object detection network is configured to detect at least one wager object in the region of interest.

15. The computing apparatus of claim 14 , wherein the at least one processor is configured to execute the computer-readable instructions to cause the computing apparatus to execute,

a gaming table region proposal network (RPN) to receive the image data from the captured images of the gaming table, and

a gaming table object detection network to receive an output of the gaming table region RPN,

wherein the gaming table object detection network is configured to detect at least one gaming object in the captured images based on an output of the gaming table object detection network, wherein the at least one gaming object is different from the at least one wager object.

16. The computing apparatus of claim 15 , wherein the at least one processor is configured to execute the computer-readable instructions to cause the computing apparatus to,

identify at least one first region of interest in the captured images that relates to the at least one gaming object,

identify a subset of the at least one first region of interest that relates to a single stack of the at least one wager object,

identify at least one second region of interest that relates to a part of an edge pattern on each of the at least one wager object that forms part of the single stack in the identified subset;

identify a value pattern in each of the at least one second region of interest; and

estimate a total wager value of the single stack using the identified value pattern and a lookup table.

17. A computing apparatus comprising:

at least one camera configured to capture images of a gaming surface of a table;

a non-transitory computer-readable memory storing computer-readable instructions, the memory being at the table; and

at least one processor, at the table, configured to execute the computer-readable instructions to cause the computing apparatus to,

apply machine learning processes to analyze the captured images of the gaming surface to identify at least one of a gaming object, a game event and a player in the captured images, identify a region of interest in at least one of the captured images, the region of interest including at least one wager object, detect an edge pattern of the at least one wager object, and determine a value of the at least one wager object based on the detected edge pattern.

18. The computing apparatus of claim 17 , wherein the at least one processor is configured to execute the computer-readable instructions to cause the computing apparatus to,

identify at least one first region of interest in the captured images that relates to at least one gaming object,

identify a subset of the at least one first region of interest that relates to a single stack of at least one wager object,

identify at least one second region of interest that relates to a part of an edge pattern on each of the at least one wager object that forms part of the single stack in the identified subset;

identify a value pattern in each of the at least one second region of interest; and

estimate a total wager value of the single stack using the identified value pattern and a lookup table.

19. The computing apparatus of claim 17 , wherein the at least one processor is configured to execute the computer-readable instructions to cause the computing apparatus to,

detect an end of the detected edge pattern, and

determine the value based on the detected end.