Gesture recognition method, computing device, and control device

Patent metadata

Field	Value
Publication number	US-2016334880-A1
Application number	US-201615151711-A
Country	US
Kind code	A1
Filing date	May 11, 2016
Priority date	May 12, 2015
Publication date	Nov 17, 2016
Grant date	—

Abstract

Disclosed are a gesture recognition method, a non-transitory computer readable recording medium having recorded thereon a program for executing the gesture recognition method, a computing device, and a control device. The gesture recognition method includes receiving signals of a plurality of motion features; detecting a variable gesture frame from the signals; extracting a feature sample from the variable gesture frame; and determining a gesture command corresponding to the extracted feature sample.

First claim

What is claimed is:

1 . A gesture recognition method comprising:
receiving signals of a plurality of motion features; detecting a variable gesture frame from the signals; extracting a feature sample from the variable gesture frame; and determining a gesture command corresponding to the extracted feature sample.

2 . The gesture recognition method of claim 1 , wherein the detecting of the variable gesture frame comprises:
determining whether each of the signals is a gesture component or a non-gesture component; and determining a length of the variable gesture frame based on a ratio of the gesture components to the non-gesture components.

3 . The gesture recognition method of claim 2 , wherein the determining of the length of the variable gesture frame based on the ratio of the gesture components to the non-gesture components comprises:
determining a threshold value based on a ratio of the gesture components to the non-gesture components which are detected in a predefined section of the signals; and setting a start point of the variable gesture frame based on point in time when the ratio of the gesture components to the non-gesture components of the signals equals the threshold value, increasing the length of the variable gesture frame based on whether the ratio of the gesture components to the non-gesture components of the signals exceeds the threshold value, and determining an end point of the variable gesture frame based on a point in time when the ratio of the gesture components to the non-gesture components of the signals decreases below the threshold value.

4 . The gesture recognition method of claim 1 , wherein the extracting of the feature sample from the variable gesture frame comprises:
extracting an intrinsic feature from the variable gesture frame; extracting a high level feature from a gesture sample of the variable gesture frame; and obtaining the feature sample based on a combination of the intrinsic feature and the high level feature.

5 . The gesture recognition method of claim 1 , wherein the receiving of the signals of the plurality of motion features comprises: receiving the signals of the plurality of motion features from an accelerometer and a gyroscope.

6 . A non-transitory computer-readable recording medium having recorded thereon a program, which when executed by a computer, performs a gesture recognition method comprising:
receiving signals of a plurality of motion features; detecting a variable gesture frame from the signals; extracting a feature sample from the variable gesture frame; and determining a gesture command corresponding to the extracted feature sample.

7 . The non-transitory computer-readable recording medium of claim 6 , wherein the detecting of the variable gesture frame comprises:
determining whether each of the signals is a gesture component or a non-gesture component; and determining a length of the variable gesture frame based on a ratio of the gesture components to the non-gesture components.

8 . The non-transitory computer-readable recording medium of claim 7 , wherein the determining of the length of the variable gesture frame based on the ratio of the gesture components to the non-gesture components comprises:
determining a threshold value based on a ratio of the gesture components to the non-gesture components which are detected in a predefined section of the signals; and setting a start point of the variable gesture frame based on a point in time when the ratio of the gesture components to the non-gesture components of the signals equals the threshold value, increasing the length of the variable gesture frame based on whether the ratio of the gesture components to the non-gesture components of the signals exceeds the threshold value, and determining an end point of the variable gesture frame based on a point in time when the ratio of the gesture components to the non-gesture components of the signals decreases below the threshold value.

9 . The non-transitory computer-readable recording medium of claim 6 , wherein the extracting of the feature sample from the variable gesture frame comprises:
extracting an intrinsic feature from the variable gesture frame; extracting a high level feature by filtering a gesture sample of the variable gesture frame; and obtaining the feature sample by combining the intrinsic feature and the high level feature.

10 . The non-transitory computer-readable recording medium of claim 6 , wherein the receiving of the signals of the plurality of motion features comprises: receiving the signals of the plurality of motion features from an accelerometer and a gyroscope.

11 . A computing device comprising:
a communicator comprising communication circuitry configured to receive, from a control device, signals of a plurality of motion features; and a controller configured to determine a gesture command corresponding to the received signals and to control the computing device to perform an action corresponding to the gesture command, and wherein the controller, in determining the gesture command, is configured to detect a variable gesture frame from the received signals, to extract a feature sample from the variable gesture frame, and to determine a gesture command corresponding to the extracted feature sample.

12 . The computing device of claim 11 , wherein the controller, in detecting the variable gesture frame, is configured to determine whether each of the signals is a gesture component or a non-gesture component and determine a length of the variable gesture frame based on a ratio of the gesture components to the non-gesture components.

13 . The computing device of claim 12 , wherein the controller is configured to determine a threshold value based on a ratio of the gesture components to the non-gesture components which are detected in a predefined section of the signals, to set a start point of the variable gesture frame based on a point in time when the ratio of the gesture components to the non-gesture components of the signals equals the threshold value, to increase the length of the variable gesture frame based on whether the ratio of the gesture components to the non-gesture components of the signals exceeds the threshold value, and to determine an end point of the variable gesture frame based on a point in time when the ratio of the gesture components to the non-gesture components of the signals decreases below the threshold value.

14 . The computing device of claim 11 , wherein the controller, in extracting the feature sample from the variable gesture frame, is configured to extract an intrinsic feature from the variable gesture frame, to extract a high level feature by filtering a gesture sample of the variable gesture frame, and to obtain the feature sample based on a combination of the intrinsic feature and the high level feature.

15 . A control device comprising:
a communicator comprising communication circuitry; a sensor configured to sense motion of the control device and signals of a plurality of motion features; and a controller configured to determine a gesture command corresponding to the signals sensed by the sensor and to control the communicator to transmit the gesture command to an external device, and wherein the controller, in determining the gesture command, is configured to detect a variable gesture frame from the signals sensed by the sensor, to extract a feature sample from the variable gesture frame, and to determine a gesture command corresponding to the extracted feature sample.

16 . The control device of claim 15 , wherein the cont

Assignees

• Samsung Electronics Co Ltd

Inventors

• Favre Philippe
• Kryukov Evgeny

Classifications

• G06K9/00335

  Physics · mapped topic

• G06F3/017 Primary

  Gesture based interaction, e.g. based on a set of recognized hand gestures (interaction based on gestures traced on a digitiser G06F3/04883) · CPC title

• G06V40/20

  Movements or behaviour, e.g. gesture recognition (recognition of facial expressions G06V40/16) · CPC title