US 12,249,185 B2
System and method for exercise type recognition using wearables
Mark Fralick, San Diego, CA (US); and Brian Chen, Sunnyvale, CA (US)
Assigned to Google LLC, Mountain View, CA (US)
Filed by Google LLC, Mountain View, CA (US)
Filed on Nov. 30, 2023, as Appl. No. 18/524,615.
Application 18/524,615 is a continuation of application No. 17/419,368, granted, now 11,842,571, issued on Dec. 12, 2023, previously published as PCT/US2020/043997, filed on Jul. 29, 2020.
Prior Publication US 2024/0096136 A1, Mar. 21, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G06V 40/20 (2022.01); G06V 10/778 (2022.01); G06V 10/80 (2022.01)
CPC G06V 40/20 (2022.01) [G06V 10/778 (2022.01); G06V 10/809 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A method for detecting exercise, comprising:
receiving, by one or more processors, first sensor data from one or more first sensors of a first wearable device;
receiving, by the one or more processors, second sensor data from one or more second sensors of a second wearable device;
determining, using the one or more processors executing a machine learning model, a type of exercise performed by a user during receiving of the first sensor data and the second sensor data; and
training the machine learning model, wherein the training comprises generating one or more first training images based on data collected from the first and second wearable devices while the user performs a first type of exercise and inputting the one or more first training images into the machine learning model as training data, wherein the data collected from the first and second wearable devices comprises movement data of the first and second wearable devices in x, y, and z directions.