Capturing and quantifying loop drive ball metrics

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 historical loop drive ball return video recording data set, including a plurality of video recordings of historical loop drive ball returns, where a given video recording includes at least one instance of a historical table tennis player returning a table tennis ball with a loop drive technique,
• determining, for at least one given video recording of historical loop drive ball returns, a set of metrics and their corresponding quantitative values corresponding to properties of the table tennis ball in the given video recording of historical loop drive ball returns,
• building a machine learning model to assign weights to the set of metrics,
• receiving a live loop drive ball return video recording data set, including a video recording of a live loop drive ball return, where the video recording of the live loop drive ball return includes at least one instance of a live table tennis player returning a table tennis ball with a loop drive technique,
• determining a set of metrics and their corresponding quantitative values corresponding to properties of the table tennis ball in the video recording of the live loop drive ball return, and
• determining a quantitative loop drive ball return summary value corresponding to properties of the table tennis ball in the video recording of a live loop drive ball return based, at least in part, on the weights assigned by the machine learning model.

A computer system (CS) comprising:

• a processor(s) set;
• a machine readable storage device;
• and
• computer code stored on the machine readable storage device, with the computer code including instructions for causing the processor(s) set to perform operations including the following:
• receiving a historical loop drive ball return video recording data set, including a plurality of video recordings of historical loop drive ball returns, where a given video recording includes at least one instance of a historical table tennis player returning a table tennis ball with a loop drive technique,
• determining, for at least one given video recording of historical loop drive ball returns, a set of metrics and their corresponding quantitative values corresponding to properties of the table tennis ball in the given video recording of historical loop drive ball returns,
• building a machine learning model to assign weights to the set of metrics,
• receiving a live loop drive ball return video recording data set, including a video recording of a live loop drive ball return, where the video recording of the live loop drive ball return includes at least one instance of a live table tennis player returning a table tennis ball with a loop drive technique,
• determining a set of metrics and their corresponding quantitative values corresponding to properties of the table tennis ball in the video recording of the live loop drive ball return, and
• determining a quantitative loop drive ball return summary value corresponding to properties of the table tennis ball in the video recording of a live loop drive ball return based, at least in part, on the weights assigned by the machine learning model.