IP Library Granted Patent US 12,731,437
Granted Patent B2
US 12,731,437 · App. 17/564,195 · Granted Sep 8, 2026

Virtual trainer on mobile/IOT devices

Inventors: Yibing Michelle Wang (Temple City, CA); Hongyu Wendy Wang (South Pasadena, CA)
Assignee: SAMSUNG ELECTRONICS CO., LTD.
G06V40/23A63B24/0062A63B24/0075A63B71/0622G06T13/40G06T13/80G06T19/006A63B2071/0625A63B2071/0636A63B2220/05A63B2244/10A63B2244/22
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,731,437
App. No.
17/564,195
Granted
Sep 8, 2026
Kind
B2
Abstract

A virtual training system and method are disclosed. The system and methods include receiving, at a processing device, a two-dimensional (2D) image of a user from a 2D image sensor; receiving, at the processing device, a three-dimensional (3D) image of the user from a 3D image sensor; receiving, at the processing device, at least one biometric characteristic of the user; generating, by the processing device, feedback information relating to movement of the user during a session compared to a standard form movement; and outputting the feedback information to the user as an image of the user captured by at least one of the 2D image sensor and the 3D image sensor overlayed on an instructor avatar.

Claims (36)

1 . A virtual training system, comprising:

a two-dimensional (2D) image sensor configured to capture a 2D image of a user;

a three-dimensional (3D) image sensor configured to sense a volume and motion of the user and generate 3D image data of the user based on the sensed volume and motion of the user;

a wireless receiver configured to receive a wireless signal that includes biometric data from a biometric data sensor worn by the user, the biometric data comprising at least one biometric characteristic of the user sensed by the biometric sensor;

a processing device configured to receive the 2D image, the 3D image data, and the at least one biometric characteristic of the user, the processing device further configured to:

generate a profile for the user based on the 2D image, the 3D image data, and the at least one biometric characteristic of the user;

generate, based on the profile of the user, a customized form movement by modifying one or more parameters of a standard form movement according to the profile of the user derived from the 2D image, the 3D image data, and the at least one biometric characteristic;

generate, based on processing the 2D image, the 3D image data, and the at least one biometric characteristic of the user, feedback information relating to movement of the user during a session compared to the customized form movement; and

an output device configured to provide the feedback information to the user as an image of the user captured by at least one of the 2D image sensor and the 3D image sensor overlayed on an instructor avatar.

2 . The virtual training system of claim 1 , wherein the processing device is further configured to generate the feedback information in real time during the session.

3 . The virtual training system of claim 1 , wherein the output device is further configured to provide the feedback information to the user as at least one of graphical and textual information.

4 . The virtual training system of claim 1 , wherein the output device further comprises an audio output device configured to provide the feedback information as audible information.

5 . The virtual training system of claim 1 , wherein the processing device is further configured to generate the instructor avatar to illustrate the standard form movement based on the profile for the user.

6 . The virtual training system of claim 5 , wherein the processing device is further configured to adapt the instructor avatar to the user based on the profile for the user.

7 . The virtual training system of claim 1 , wherein the processing device is further configured to monitor the at least one biometric characteristic of the user compared to a safe zone of the at least one biometric characteristic during the session.

8 . The virtual training system of claim 1 , wherein at least one of the 2D image sensor and the 3D image sensor comprises part of a smartphone.

9 . The virtual training system of claim 1 , wherein the standard form movement comprises one of an exercise movement, a dance movement, a martial arts movement, or a physical therapy movement.

10 . The virtual training system of claim 1 , wherein the processing device is further configured to:

compare the at least one biometric characteristic to a safe zone based on the processing device monitoring the at least one biometric characteristic during the session; and

based on a determination that the at least one biometric characteristic is outside the safe zone, modify, without user intervention, an intensity or pace of the session.

11 . A method, comprising:

receiving, at a processing device, a two-dimensional (2D) image of a user from a 2D image sensor;

receiving, at the processing device, a three-dimensional (3D) image of the user from a 3D image sensor configured to sense a volume and motion of the user and generate computer 3D image data of the user based on the sensed volume and motion of the user;

receiving, at the processing device via a wireless receiver, biometric data, the wireless receiver being configured to receive a wireless signal that includes biometric data from a biometric data sensor worn by the user, the biometric data comprising at least one biometric characteristic of the user;

generating a profile for the user based on the 2D image, the 3D image data, and the at least one biometric characteristic of the user;

generating, by the processing device and based on the profile of the user, a customized form movement by modifying one or more parameters of a standard form movement according to the profile of the user derived from the 2D image, the 3D image data, and the at least one biometric characteristic;

generating, by the processing device, feedback information relating to movement of the user during a session compared to the customized form movement; and

outputting the feedback information to the user as an image of the user captured by at least one of the 2D image sensor and the 3D image sensor overlayed on an instructor avatar.

12 . The method of claim 11 , wherein generating the feedback information comprises generating the feedback information in real time during the session.

13 . The method of claim 11 , wherein outputting the feedback information further comprises outputting the feedback information as at least one of graphical and textual information.

14 . The method of claim 11 , wherein outputting the feedback information further comprises outputting the feedback information to an audio device configured to provide the feedback information as audible information.

15 . The method of claim 11 , further comprising generating, by the processing device, the instructor avatar to illustrate the standard form movement based on the profile for the user.

16 . The method of claim 15 , further comprising adapting, by the processing device, the instructor avatar to the user based on the profile for the user.

17 . The method of claim 11 , further comprising monitoring, by the processing device, the at least one biometric characteristic of the user compared to a safe zone of the at least one biometric characteristic during the session.

18 . The method of claim 11 , wherein at least one of the 2D image sensor and the 3D image sensor comprises part of a smartphone.

19 . The method of claim 11 , wherein the standard form movement comprises one of an exercise movement, a dance movement, a martial arts movement, or a physical therapy movement.

Continuity (2)
Provisional Application 63271705 · Oct 25, 2021
Related Publication 20230130555A1 · Apr 27, 2023
References Cited (25)
US 9330239B2 · Koduri et al. · 2016 [cited by applicant]
US 10188930B2 · Winsper et al. · 2019 [cited by applicant]
US 10369412B2 · Aragones et al. · 2019 [cited by applicant]
US 10583328B2 · Aragones et al. · 2020 [cited by applicant]
US 10943407B1 · Morgan · 2021 [cited by examiner]
US 11051767B2 · Kaleal, III et al. · 2021 [cited by applicant]
US 11065527B2 · Putnam · 2021 [cited by applicant]
US 11615648B2 · Lee · 2023 [cited by examiner]
US 11990233B2 · Kaleal, III · 2024 [cited by examiner]
US 20070038153A1 · Basson · 2007 [cited by examiner]
US 20120183940A1 · Aragones · 2012 [cited by examiner]
US 20140225978A1 · Saban · 2014 [cited by examiner]
US 20200054929A1 · Ward et al. · 2020 [cited by applicant]
US 20200111384A1 · Bell et al. · 2020 [cited by applicant]
US 20210008413A1 · Asikainen et al. · 2021 [cited by applicant]
US 20210093920A1 · Poulin et al. · 2021 [cited by applicant]
US 20210197022A1 · Liu · 2021 [cited by applicant]
US 20210307650A1 · Barr · 2021 [cited by examiner]
US 20220296966A1 · Asikainen · 2022 [cited by examiner]
WO 2019231982A1 · 2019 [cited by applicant]
WO 2020132110A1 · 2020 [cited by applicant]
WO 2021014149A1 · 2021 [cited by applicant]
Fieraru, Mihai et al., “AlFit: Automatic 3D Human-Interpretable Feedback Models for Fitness Training,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 9919-9928. [cited by applicant]
Venek, Verena et al., “Towards a Live Feedback Training System: Interchangeability of Orbbec Persee and Microsoft Kinect for Exercise Monitoring,” Designs 2021, vol. 5, No. 30, 2021, 15 pages. [cited by applicant]
Xie, Haoran et al., “Visual Feedback for Core Training with 3D Human Shape and Pose,” 2019 Nicograph International (NicoInt), 2019, 9 pages. [cited by applicant]