Methods and systems for determining ball shot attempt location on ball court

1. A method for determining a ball shot attempt location on a ball court, comprising:

receiving an input video captured using a camera connected to a computing device, wherein the input video comprises the ball shot attempt and one or more players;

receiving a declared ball shot attempt, wherein the declared ball shot attempt is detected by analyzing a ball trajectory in the input video using a first machine vision algorithm, and wherein the declared ball shot attempt is associated with a ball shot attempt time;

determining a potential shooter for the declared ball shot attempt by detecting a moving object flow in the input video, from one or more frames backtracked in reverse time from the ball shot attempt time;

determining a potential shooter image region in an image plane by analyzing an overlapping image area of the ball trajectory and the potential shooter;

determining a shooter foot-on-court time;

determining a most probable pair of shooter legs in the potential shooter image region around the shooter foot-on-court time;

determining a shooter foot location based on the most probable pair of shooter legs; and

determining a ball shot attempt location on the ball court by applying a camera projection to the shooter foot location.

2. The method of claim 1 , further comprising:

determining a ball shot attempt result associated with the ball shot attempt; and

updating a court-location-indexed shot performance statistics dataset based on the ball shot attempt result.

3. The method of claim 2 , further comprising:

determining a shooter identity as a player having the most probable pair of shooter legs, wherein the updating the court-location-indexed shot performance statistics dataset is further based on the shooter identity.

4. The method of claim 1 , wherein the determining the most probable pair of shooter legs comprises:

detecting one or more pairs of potential shooter legs of the potential shooter by applying a second machine vision algorithm to one or more frames around the shooter foot-on-court time, within the potential shooter image region; and

selecting the most probable pair of shooter legs based on a player score, from the one or more pairs of potential shooter legs.

5. The method of claim 4 , wherein the player score is calculated based on whether a player is in a ball moving direction, whether a player is within a projected shooter region, a distance between the ball and a top of a player bound, a distance between a center of the player bound and a center of a cropped image, and/or a historical shooting zone.

6. The method of claim 1 , wherein the determining the shooter foot-on-court time is based on the ball trajectory.

7. The method of claim 1 , wherein the computing device is a mobile computing device.

8. A non-transitory storage medium storing program code, the program code when executed by a processor, causes the processor to execute a process for determining a ball shot attempt location on a ball court, the program code comprising steps to:

receive an input video captured using a camera connected to a computing device, wherein the input video comprises the ball shot attempt and one or more players;

receive a declared ball shot attempt, wherein the declared ball shot attempt is detected by analyzing a ball trajectory in the input video using a first machine vision algorithm, and wherein the declared ball shot attempt is associated with a ball shot attempt time;

determine a potential shooter for the declared ball shot attempt by detecting a moving object flow in the input video, from one or more frames backtracked in reverse time from the ball shot attempt time;

determine a potential shooter image region in an image plane by analyzing an overlapping image area of the ball trajectory and the potential shooter;

determine a shooter foot-on-court time;

determine a most probable pair of shooter legs in the potential shooter image region around the shooter foot-on-court time;

determine a shooter foot location based on the most probable pair of shooter legs; and

determine a ball shot attempt location on the ball court by applying a camera projection to the shooter foot location.

9. The non-transitory storage medium of claim 8 , the program code further comprising steps to:

determine a ball shot attempt result associated with the ball shot attempt; and

update a court-location-indexed shot performance statistics dataset based on the ball shot attempt result.

10. The non-transitory storage medium of claim 9 , the program code further comprising steps to:

determine a shooter identity as a player having the most probable pair of shooter legs, wherein the program code to update the court-location-indexed shot performance statistics dataset further utilizes the shooter identity.

11. The non-transitory storage medium of claim 8 , wherein the program code to determine the most probable pair of shooter legs comprises program code to:

detect one or more pairs of potential shooter legs of the potential shooter by applying a second machine vision algorithm to one or more frames around the shooter foot-on-court time, within the potential shooter image region; and

select the most probable pair of shooter legs based on a player score, from the one or more pairs of potential shooter legs.

12. The non-transitory storage medium of claim 11 , wherein the player score is calculated based on whether a player is in a ball moving direction, whether a player is within a projected shooter region, a distance between the ball and a top of a player bound, a distance between a center of the player bound and a center of a cropped image, and/or a historical shooting zone.

13. The non-transitory storage medium of claim 8 , wherein the program code to determine the shooter foot-on-court time is based on the ball trajectory.

14. The non-transitory storage medium of claim 8 , wherein the computing device is a mobile computing device.

15. A system comprising a hardware processor and a non-transitory storage medium storing program code, the program code when executed by the hardware processor, causes the hardware processor to execute a process for determining a ball shot attempt location on a ball court, the program code comprising steps to:

receive an input video captured using a camera connected to a computing device, wherein the input video comprises the ball shot attempt and one or more players;

receive a declared ball shot attempt, wherein the declared ball shot attempt is detected by analyzing a ball trajectory in the input video using a first machine vision algorithm, and wherein the declared ball shot attempt is associated with a ball shot attempt time;

determine a potential shooter for the declared ball shot attempt by detecting a moving object flow in the input video, from one or more frames backtracked in reverse time from the ball shot attempt time;

determine a potential shooter image region in an image plane by analyzing an overlapping image area of the ball trajectory and the potential shooter;

determine a shooter foot-on-court time;

determine a most probable pair of shooter legs in the potential shooter image region around the shooter foot-on-court time;

determine a shooter foot location based on the most probable pair of shooter legs; and

determine a ball shot attempt location on the ball court by applying a camera projection to the shooter foot location.

16. The system of claim 15 , the program code further comprising steps to:

determine a ball shot attempt result associated with the ball shot attempt; and

update a court-location-indexed shot performance statistics dataset based on the ball shot attempt result.

17. The system of claim 16 , the program code further comprising steps to:

determine a shooter identity as a player having the most probable pair of shooter legs, wherein the program code to update the court-location-indexed shot performance statistics dataset further utilizes the shooter identity.

18. The system of claim 15 , wherein the program code to determine the most probable pair of shooter legs comprises program code to:

detect one or more pairs of potential shooter legs of the potential shooter by applying a second machine vision algorithm to one or more frames around the shooter foot-on-court time, within the potential shooter image region; and

select the most probable pair of shooter legs based on a player score, from the one or more pairs of potential shooter legs.

19. The system of claim 15 , wherein the program code to determine the shooter foot-on-court time is based on the ball trajectory.

20. The system of claim 15 , wherein the computing device is a mobile computing device.