Capturing and quantifying body postures

Patent metadata

Field	Value
Publication number	US-11417136-B2
Application number	US-202017038568-A
Country	US
Kind code	B2
Filing date	Sep 30, 2020
Priority date	Sep 30, 2020
Publication date	Aug 16, 2022
Grant date	Aug 16, 2022

Abstract

Disclosed are techniques for quantifying body postures of a player employing a loop drive technique to strike a ball, such as performed in table tennis activities. A video recording of a player striking a ball with a loop drive technique is received and divided, using image processing techniques, into two segments: the first concerning player body postures before the ball is hit, and the second concerning body postures from the moment of impact between the ball and racket and the subsequent follow-through body postures. Then, image processing techniques are again leveraged to isolate and quantify specific body postures contributing to a loop drive technique in a given segment.

First claim

What is claimed is:

1. A computer-implemented method (CIM) comprising:
receiving a table tennis video recording data set, with the table tennis video recording data set including a video recording observing a table tennis player and their body as the table tennis player performs a sequence of motions leading up to and inclusive of follow-through motions after hitting a table tennis ball with a table tennis racket;
determining, automatically, at least two segments of video recordings from the table tennis video recording data set, including a first segment corresponding to motions leading up to hitting the table tennis ball with the table tennis racket, and a second segment corresponding to hitting the table tennis ball with the table tennis racket and subsequent follow-through motions;
determining, automatically, for the first segment, a first set of data points corresponding to quantitative values for changes in body posture of the player as they perform the sequence of motions leading up to hitting the table tennis ball with the table tennis racket; and
the first set of data points includes at least: (i) an angle value corresponding to total rotation as the player rotates at their waist, (ii) an angle value corresponding to an angle formed by a lower leg and upper leg of the table tennis player as the table tennis player bends their legs at their respective knees at maximum flexion, (iii) a height value corresponding to the difference between a maximum and minimum distance between a ground surface and shoulders of the table tennis player as they bend and extend their legs, and (iv) a distance value corresponding to a total distance travelled by the table tennis racket beginning with the table tennis player swinging the table tennis racket towards the table tennis ball and ending immediately upon physical contact between the table tennis racket and the table tennis ball.

2. A computer-implemented method (CIM) comprising:
receiving a table tennis video recording data set, with the table tennis video recording data set including a video recording observing a table tennis player and their body as the table tennis player performs a sequence of motions leading up to and inclusive of follow-through motions after hitting a table tennis ball with a table tennis racket;
determining, automatically, at least two segments of video recordings from the table tennis video recording data set, including a first segment corresponding to motions leading up to hitting the table tennis ball with the table tennis racket, and a second segment corresponding to hitting the table tennis ball with the table tennis racket and subsequent follow-through motions;
determining, automatically, for the first segment, a first set of data points corresponding to quantitative values for changes in body posture of the player as they perform the sequence of motions leading up to hitting the table tennis ball with the table tennis racket;
determining, automatically, for the second segment, a second set of data points corresponding to quantitative values for changes in body posture of the player as the player hits the table tennis ball with the table tennis racket and subsequent follow-through motions; and
the second set of data points includes at least: (i) an angle value corresponding to an angle formed between a forearm and upper arm of the table tennis player where the forearm and upper arm correspond to an arm holding the table tennis racket, (ii) an angle between the upper arm and a torso of the table tennis player, (iii) a height value corresponding to a distance between a table tennis table and an elbow of the table tennis player where the elbow corresponds to the arm holding the table tennis racket, (iv) a velocity value corresponding to average velocity of the table tennis racket from beginning to end of physical contact with the table tennis ball, (v) a height value corresponding to a distance between the table tennis table and the table tennis racket upon beginning contact between the table tennis ball and the table tennis racket, and (vi) a pair of coordinates corresponding to a location upon a striking surface of the table tennis racket where contact between the table tennis racket and table tennis ball occurred.

3. A computer program product (CPP) comprising:
a machine readable storage device; and
computer code stored on the machine readable storage device, with the computer code including instructions for causing a processor(s) set to perform operations including the following:
receiving a table tennis video recording data set, with the table tennis video recording data set including a video recording observing a table tennis player and their body as the table tennis player performs a sequence of motions leading up to and inclusive of follow-through motions after hitting a table tennis ball with a table tennis racket,

determining, automatically, at least two segments of video recordings from the table tennis video recording data set, including a first segment corresponding to motions leading up to hitting the table tennis ball with the table tennis racket, and a second segment corresponding to hitting the table tennis ball with the table tennis racket and subsequent follow-through motions,
determining, automatically, for the first segment, a first set of data points corresponding to quantitative values for changes in body posture of the player as they perform the sequence of motions leading up to hitting the table tennis ball with the table tennis racket, and
the first set of data points includes at least: (i) an angle value corresponding to total rotation as the player rotates at their waist, (ii) an angle value corresponding to an angle formed by a lower leg and upper leg of the table tennis player as the table tennis player bends their legs at their respective knees at maximum flexion, (iii) a height value corresponding to the difference between a maximum and minimum distance between a ground surface and shoulders of the table tennis player as they bend and extend their legs, and (iv) a distance value corresponding to a total distance travelled by the table tennis racket beginning with the table tennis player swinging the table tennis racket towards the table tennis ball and ending immediately upon physical contact between the table tennis racket and the table tennis ball.

Assignees

• IBM

Inventors

• Ran Chuan
• Wang Junhui
• Zhu Guandong
• Cui Yongchang
• Lu Shuyan
• Yang Pu

Classifications

• A63B69/38

  for tennis {(A63B61/006, A63B69/0073 and A63B69/0097 take precedence)} · CPC title

• G06V40/23

  Recognition of whole body movements, e.g. for sport training · CPC title

• G06V20/49 Primary

  Segmenting video sequences, i.e. computational techniques such as parsing or cutting the sequence, low-level clustering or determining units such as shots or scenes · CPC title

• G06V40/103 Primary

  Static body considered as a whole, e.g. static pedestrian or occupant recognition · CPC title

• G06V40/113

  Recognition of static hand signs · CPC title