# US 2019/0377938 A1 · Device and method for recognizing gesture

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2019/0377938 A1
- Document ID: US20190377938A1
- Status: Published application
- Kind: A1
- Application number: 16/435,085
- Filed: 2019-06-07
- Published: 2019-12-12
- Earliest claimed: 2018-06-08
- Assignees: Korea Electronics Technology Institute
- Applicants: Korea Electronics Technology Institute
- Inventors: Yang Keun AHN; Young Choong PARK
- Web page: https://patentgrep.com/patent/US20190377938A1

## CPC

- G06F 3/017: {Gesture based interaction, e.g. based on a set of recognized hand gestures (interaction based on gestures traced on a digitiser G06F3/04883)}
- G06T 7/55: from multiple images
- G06V 10/28: Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
- G06V 40/20: Movements or behaviour, e.g. gesture recognition (recognition of facial expressions G06V40/16)
- G06V 40/28: {Recognition of hand or arm movements, e.g. recognition of deaf sign language (static hand signs G06V40/113)}
- G06T 2207/10016: Video; Image sequence
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

Provided are a device and method for recognizing a gesture. The device includes a video image acquirer configured to acquire a video including a gesture object by photographing the gesture object, a center point detector configured to detect a center point of the gesture object in frames of the acquired video, a velocity calculator configured to calculate a velocity of the gesture object by calculating a difference between motion coordinates on each axis on the basis of the center point of the gesture object, and a gesture recognizer configured to recognize a gesture of the gesture object by comparing the calculated velocity of the gesture object with a preset velocity.

## Claims

### Claim 1

A device for recognizing a gesture, the device comprising:
a video image acquirer configured to acquire a video including a gesture object by photographing the gesture object;
a center point detector configured to detect a center point of the gesture object in frames of the acquired video;
a velocity calculator configured to calculate a velocity of the gesture object by calculating a difference between motion coordinates on each axis on the basis of the center point of the gesture object; and
a gesture recognizer configured to recognize a gesture of the gesture object by comparing the calculated velocity of the gesture object with a preset velocity.

### Claim 2 (depends on 1)

The device of claim 1, wherein the video image acquirer is a three-dimensional (3D) depth camera.

### Claim 3 (depends on 1)

The device of claim 1, wherein the center point detector binarizes a frame of the acquired video, determines whether a blob is detected in the binarized frame, and when a blob is detected in the frame, detects a center point of the blob by separately calculating average values of three-dimensional (3D) pixels of the detected blob.

### Claim 4 (depends on 1)

The device of claim 1, wherein the velocity calculator sets up a frame in which a center point of a blob has been initially detected as a reference frame among detected frames, initializes the center point of the blob, and then when the center point of the blob is detected in a subsequent frame, calculates axis-specific center point velocities of the blob using coordinate differences between the center point of the blob detected in the reference frame and the center point of the blob detected in the subsequent frame and the number of frames therebetween.

### Claim 5 (depends on 4)

The device of claim 4, wherein the gesture recognizer recognizes an axis on which the center point velocity of the blob is higher than the preset velocity as a gesture of the gesture object.

### Claim 6 (depends on 4)

The device of claim 4, wherein the gesture recognizer compares the axis-specific center point velocities of the blob with each other and recognizes an axis on which the center point velocity of the blob is the highest as a gesture of the gesture object.

### Claim 7

A method of recognizing a gesture, the method comprising:
acquiring a video including a gesture object by photographing the gesture object;
detecting a center point of the gesture object in frames of the acquired video;
calculating a velocity of the gesture object by calculating a difference between motion coordinates on each axis on the basis of the center point of the gesture object; and
recognizing a gesture of the gesture object by comparing the calculated velocity of the gesture object with a preset velocity.

### Claim 8 (depends on 7)

The method of claim 7, wherein the acquiring of the video including the gesture object comprises acquiring the video including the gesture object through a three dimensional (3D) depth camera.

### Claim 9 (depends on 7)

The method of claim 7, wherein the detecting of the center point of the gesture object comprises:
binarizing a frame of the acquired video;
determining whether a blob of the gesture object is detected in the binarized frame; and
when a blob is detected in the frame, detecting a center point of the blob by separately calculating average values of three-dimensional (3D) pixels of the detected blob.

### Claim 10 (depends on 7)

The method of claim 7, wherein the calculating of the velocity of the gesture object comprises:
setting a frame in which a center point of a blob has been initially detected as a reference frame among detected frames and initializing the center point of the blob;
after the reference frame is set, determining whether the center point of the blob has been detected in a subsequent frame;
when it is determined that the center point of the blob has been detected, determining whether there are a preset minimum number of required frames between the reference frame and the frame in which the center point of the blob has been detected; and
when it is determined that there are not the preset minimum number of required frames, determining whether the center point of the blob has been detected in a subsequent frame, and when it is determined that there are the preset minimum number of required frames, calculating axis-specific center point velocities of the blob using coordinate differences between the center point of the blob detected in the reference frame and the center point of the blob detected in the subsequent frame.

### Claim 11 (depends on 10)

The method of claim 10, wherein the recognizing of the gesture of the gesture object comprises recognizing an axis on which the center point velocity of the blob is higher than the preset velocity as a gesture of the gesture object.

### Claim 12 (depends on 10)

The method of claim 10, wherein the recognizing of the gesture of the gesture object comprises comparing the axis-specific center point velocities of the blob with each other and recognizing an axis on which the center point velocity of the blob is the highest as a gesture of the gesture object.
