Patent US9275470

Computer vision system for tracking ball movement and analyzing user skill
[+] Show Details

US9275470

1. A method, comprising: (17) (9)

overlaying, by a computer processor of the mobile device, a shape graphic on a fixed area of a screen of the mobile device;

searching, by a camera of the mobile device, the fixed area for a ball with a marking authorized by the mobile device for enabling calibration and tracking;

recognizing, by the computer processor, the authorized marking on the ball;

calibrating, by the computer processor, characteristics of the ball in the overlayed shape graphic for tracking of the ball in view of the authorized marking on the ball;

tracking and recording, by the camera, data pertaining to the movement of the ball;

calculating, by the computer processor, one or more metrics relating to ball controlling abilities from the data; and

displaying, by the computer processor, the one or more metrics on the screen, wherein the one or more metrics are calculated and displayed while ball tracking pertaining to the movement of the ball is taking place;

displaying, by the processing device, a training video of a character or person executing movements of the ball on the screen of the mobile device; and

comparing, by the processing device, the movements of the ball in the training video to the movements of the ball while ball tracking is taking place.

2. The method of claim 1, wherein the computer processor calculates the one or more metrics and displays the one or more metrics responsive to the computer processor detecting that the ball is no longer in motion or a specified time is reached. (0)

3. The method of claim 1, wherein the marking is a logo. (0)

4. The method of claim 1, wherein the camera is a front-facing camera of the mobile device. (0)

5. The method of claim 1, wherein the one or more metrics are calculated and displayed in real-time. (0)

6. The method of claim 1, wherein recognizing applies to balls with specific brands or markings. (0)

7. The method of claim 1, wherein recognizing the marking on the ball further comprises applying feature recognition computer vision algorithms to data captured by the computer processor pertaining to the marking. (1) (0)

9. The method of claim 1, wherein the marking is recognized based on at least one of image, text, or shape. (0)

10. The method of claim 1, wherein calibrating further comprises recognizing one or more colors of the marking and the ball, lighting conditions, or ball pixel diameter. (0)

11. The method of claim 1, further comprising determining, by the computer processor, ball color spectrum for color blob detection while tracking the movement of the calibrated ball. (0)

12. The method of claim 1, wherein tracking the movement of the calibrated ball comprises employing color blob detection, shape recognition, or movement filtering. (0)

13. The method of claim 1, further comprising displaying, by the screen, a live video feed while the computer processor is tracking the ball. (0)

14. The method of claim 1, wherein responsive to the camera tracking position of the ball on the screen, the computer processor calculates and stores the centroid of the ball, a frame timestamp, and a ball pixel diameter in a memory. (0)

15. The method of claim 1, further comprising estimating, by the computer processor, depth movement of the ball by examining ball pixel area in the recorded tracking data. (1) (0)

17. The method of claim 1, further comprising, looking up, by the computer processor in a memory of the mobile device, preloaded diameter of the ball corresponding to the marking detected; and (1) (1)

dynamically determining physical scale for different balls with different markings.

18. The method of claim 17, converting pixel tracking to physical tracking of the ball in a three-dimensional space in view of determining physical scale. (1) (0)

20. The method of claim 1, wherein the ball is a basketball and the one or more metrics comprises at least one of dribble rate, crossover-width, location consistency, maximum dribble height, dribble rate variance, dribble speed, or dribble fatigue. (0)

21. The method of claim 1, wherein the ball is a soccer ball and the one or more metrics comprises at least one of juggle location consistency, maximum height, and number of juggles. (0)

22. The method of claim 1, further comprising providing a training program that employs the data. (0)

23. A mobile device, comprising: (2) (5)

a memory;

a computer processor, operatively coupled to the memory;

a screen, coupled to the computer processor; and

a camera, coupled to the computer processor, the camera to: search a fixed area of the screen for a ball with a marking authorized by the mobile device for enabling calibration and tracking; and

track and record data pertaining to the movement of the ball;

the computer processor to: overlay a shape graphic on the fixed area of the screen; recognize the authorized marking on the ball;

calibrate characteristics of the ball in the overlayed shape graphic for tracking of the ball in view of the authorized marking on the ball;

calculate one or more metrics relating to ball controlling abilities from the data;

display the one or more metrics on the screen, wherein the one or more metrics are calculated and displayed while ball tracking pertaining to the movement of the ball is taking place;

display a training video of a character or person executing movements of the ball on the screen of the mobile device; and

compare the movements of the ball in the training video to the movements of the ball while ball tracking is taking place.

24. The mobile device claim 23, wherein the one or more metrics are calculated and displayed in real-time. (0)

25. The mobile device of claim 23, further comprising a positioning apparatus, wherein the mobile device is configured to be placed in the positioning apparatus. (1) (0)

View Abstract and Specification Size

PDF with Images and Document Face >

Full Text Publication >

Patent US9275470

Computer vision system for tracking ball movement and analyzing user skill

Patent Matrix® Search