NEAR-REALISTIC SPORTS MOTION ANALYSIS AND ACTIVITY MONITORING

Patent №

US 10,201,746

Granted

2019-02-12

Filed 2014

Owner

THE REGENTS OF THE UNIVERSITY OF CALIFORNIA

AI components

3

ml · vision · hardware

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

14273413

A method and apparatus for the detection, analysis, and identification of realistic human body and limb movements during sports actions for multiple uses comprises collecting the health information of the move, using the moves as a training and monitoring techniques in live usage settings, and as inputs to simulators and video game environments. At least one body-worn sensor transmits values wirelessly to a computing device that records the moves and calculates the health statistics then uses a machine-learning algorithm to analyze and identify the moves. These algorithms then output information and can simulate realistic moves, identify how far from these realistic moves the actual move is, and apply these moves to an electronic game. These algorithms, through various techniques, can distinguish movements, identify many different fine-grain movements, and enforce the full completion of these moves before registering them, thus preventing cheating in a realistic gaming scenario.

Machine learningVisionAI hardwareA63B 24/00A61B 5/11A61B 5/1123A61B 5/7264A63F 13/212A63F 13/428A61B 5/6824A61B 5/6829+3 more

AI classification

Vision1.00
AI hardware1.00
Machine learning1.00
Planning0.04
Knowledge representation0.00
Natural language0.00
Evolutionary computation0.00
Speech0.00

Ownership

THE REGENTS OF THE UNIVERSITY OF CALIFORNIA

assignment · 328540068

Assignors

SARRAFZADEH, MAJID, MORTAZAVI, JACK BOBAK

On an employer assignment, the assignors are typically the inventors.

© 2026 NYSGPT2525 LLC