# US 2021/0343042 A1 · Audio acquisition device positioning method and apparatus, and speaker recognition method and system

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

- Number: US 2021/0343042 A1
- Document ID: US20210343042A1
- Status: Published application
- Kind: A1
- Application number: 17/377,316
- Filed: 2021-07-15
- Published: 2021-11-04
- Earliest claimed: 2019-06-17
- Assignees: Tencent Technology (Shenzhen) Company Limited
- Applicants: Tencent Technology (Shenzhen) Company Limited
- Inventors: Zequn JIE; Zheng GE; Wei LIU
- Web page: https://patentgrep.com/patent/US20210343042A1

## CPC

- G06T 7/246: using feature-based methods, e.g. the tracking of corners or segments
- G06T 7/74: {involving reference images or patches}
- G06T 7/13: Edge detection
- G06T 7/44: using image operators, e.g. filters, edge density metrics or local histograms
- G06T 7/85: {Stereo camera calibration}
- G06T 2207/20084: Artificial neural networks [ANN]
- G06T 2207/30201: Face
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

An audio acquisition device positioning method is provided. In the method, a first image that includes an audio acquisition device is obtained. The audio acquisition device in the first image is identified. First coordinate data of the identified audio acquisition device in the first image is obtained. First displacement data is determined according to the first coordinate data and historical coordinate data of the audio acquisition device. First coordinates of the audio acquisition device are determined according to the first displacement data.

## Claims

### Claim 1

An audio acquisition device positioning method, comprising:
obtaining a first image that includes an audio acquisition device;
identifying the audio acquisition device in the first image;
obtaining first coordinate data of the identified audio acquisition device in the first image;
determining first displacement data according to the first coordinate data and historical coordinate data of the audio acquisition device; and
determining, by processing circuitry, first coordinates of the audio acquisition device according to the first displacement data.

### Claim 2 (depends on 1)

The audio acquisition device positioning method according to claim 1, wherein
the identifying includes identifying a recognition target conforming to the audio acquisition device in the first image; and
the determining the first coordinates includes determining, when the recognition target is identified, the first coordinates of the audio acquisition device based on target coordinate data corresponding to the recognition target.

### Claim 3 (depends on 2)

The audio acquisition device positioning method according to claim 2, wherein the determining the first coordinates comprises:
performing edge detection on the first image when a plurality of recognition targets are identified, to determine the recognition target disposed on a support apparatus among the plurality of recognition targets; and
determining the first coordinates of the audio acquisition device based on the target coordinate data corresponding to the recognition target disposed on the support apparatus.

### Claim 4 (depends on 1)

The audio acquisition device positioning method according to claim 1, wherein the determining the first displacement data comprises:
obtaining first historical coordinate data based on a weighted average processing performed on historical coordinate data of the audio acquisition device in a historical coordinate database; and
obtaining the first displacement data based on a comparison of the first coordinate data of the audio acquisition device and the first historical coordinate data.

### Claim 5 (depends on 4)

The audio acquisition device positioning method according to claim 4, wherein the determining the first coordinates of the audio acquisition device comprises:
storing historical coordinate data corresponding to the first coordinate data of the audio acquisition device in the historical coordinate database when the first displacement data is less than a first threshold;
obtaining second coordinate data based on a weighted average processing performed on the historical coordinate data in the historical coordinate database; and
determining the first coordinates of the audio acquisition device based on the second coordinate data.

### Claim 6 (depends on 4)

The audio acquisition device positioning method according to claim 4, wherein the first image is an nth image of a plurality of images; and
the determining the first coordinates of the audio acquisition device includes determining the first coordinates of the audio acquisition device in the first image based on the first historical coordinate data when the first displacement data is greater than or equal to a first threshold, n being a positive integer.

### Claim 7 (depends on 6)

The audio acquisition device positioning method according to claim 6, further comprising:
storing cache coordinate data corresponding to the first coordinate data of the first image in a cache coordinate database;
comparing first coordinate data in a second image that is obtained after the first image with one of the cache coordinate data and the first historical coordinate data; and
determining second coordinates of the audio acquisition device in the second image based on the comparison of the first coordinate data in the second image and the one of the cache coordinate data and the first historical coordinate data.

### Claim 8 (depends on 7)

The audio acquisition device positioning method according to claim 7, wherein
the comparing the first coordinate data in the second image with the cache coordinate data includes determining position deviation data between the first coordinate data in the second image and the cache coordinate data; and
the determining the second coordinates of the audio acquisition device in the second image includes:
clearing the historical coordinate database when the position deviation data is less than a second threshold;
storing historical coordinate data corresponding to the cache coordinate data and the first coordinate data in the second image in the historical coordinate database;
obtaining third coordinate data based on a weighted average processing performed on the historical coordinate data stored in the historical coordinate database; and
determining the second coordinates of the audio acquisition device in the second image based on the third coordinate data.

### Claim 9 (depends on 7)

The audio acquisition device positioning method according to claim 7, wherein
the comparing the first coordinate data in the second image with the cache coordinate data includes determining position deviation data between the first coordinate data in the second image and the cache coordinate data; and
the determining the second coordinates of the audio acquisition device in the second image includes:
comparing the first coordinate data in the second image with the first historical coordinate data to obtain second displacement data when the position deviation data is greater than or equal to a second threshold;
storing historical coordinate data corresponding to the first coordinate data in the second image in the historical coordinate database when the second displacement data corresponding to the second image is less than the first threshold;
obtaining fourth coordinate data based on a weighted average processing performed on the historical coordinate data stored in the historical coordinate database; and
determining the second coordinates of the audio acquisition device in the second image based on the fourth coordinate data.

### Claim 10 (depends on 9)

The audio acquisition device positioning method according to claim 9, further comprising:
clearing the cache coordinate database.

### Claim 11 (depends on 7)

The audio acquisition device positioning method according to claim 7, wherein
the comparing the first coordinate data in the second image with the first historical coordinate data includes comparing the first coordinate data in the second image with the first historical coordinate data, to obtain second displacement data; and
the determining the second coordinates of the audio acquisition device in the second image includes:
determining position deviation data between the first coordinate data in the second image and the cache coordinate data when the second displacement data corresponding to the second image is greater than or equal to the first threshold;
clearing the historical coordinate database when the position deviation data is less than a second threshold;
storing historical coordinate data corresponding to the cache coordinate data and the first coordinate data in the second image in the historical coordinate database;
obtaining third coordinate data based on a weighted average processing performed on the historical coordinate data stored in the historical coordinate database; and
determining the second coordinates of the audio acquisition device in the second image based on the third coordinate data.

### Claim 12 (depends on 7)

The audio acquisition device positioning method according to claim 7, wherein
the comparing the first coordinate data in the second image includes comparing the first coordinate data in the second image with the first historical coordinate data, to obtain second displacement data; and
the determining the second coordinates of the audio acquisition device in the second image includes:
storing historical coordinate data corresponding to the first coordinate data in the second image in the historical coordinate database when the second displacement data corresponding to the second image is less than the first threshold;
obtaining fourth coordinate data based a weighted average processing performed on the historical coordinate data stored in the historical coordinate database,; and
determining the second coordinates of the audio acquisition device in the second image based on the fourth coordinate data.

### Claim 13 (depends on 12)

The audio acquisition device positioning method according to claim 12, further comprising:
clearing the cache coordinate database.

### Claim 14 (depends on 1)

The audio acquisition device positioning method according to claim 1, further comprising:
performing a weighted average processing on historical coordinate data of the audio acquisition device in a historical coordinate database to obtain fifth coordinate data when the audio acquisition device is not recognized in k consecutive images and k is less than a third threshold; and
determining the first coordinates of the audio acquisition device in the first image based on the fifth coordinate data, k being a positive integer.

### Claim 15 (depends on 1)

The audio acquisition device positioning method according to claim 1, further comprising:
generating notification information indicating that the audio acquisition device is not present when the audio acquisition device is not recognized in j consecutive images and j is greater than or equal to a third threshold, j being a positive integer.

### Claim 16

An audio acquisition device positioning apparatus, comprising:
processing circuitry configured to:
obtain a first image that includes an audio acquisition device;
identify the audio acquisition device in the first image;
obtain first coordinate data of the identified audio acquisition device in the first image;
determine first displacement data according to the first coordinate data and historical coordinate data of the audio acquisition device; and
determine first coordinates of the audio acquisition device according to the first displacement data.

### Claim 17 (depends on 16)

The audio acquisition device positioning method according to claim 16, wherein the processing circuitry is configured to:
identify a recognition target conforming to the audio acquisition device in the first image; and
determine, when the recognition target is identified, the first coordinates of the audio acquisition device based on the target coordinate data corresponding to the recognition target.

### Claim 18

A speaker recognition method, comprising:
obtaining an image from an image capture device;
performing face recognition processing on the image to obtain coordinates of at least one face in the image;
identifying an audio acquisition device in the image, to obtain coordinates of the audio acquisition device;
determining a distance between the coordinates of the audio acquisition device and coordinates of each of the at least one face; and
determining, by processing circuitry, an object corresponding to the coordinates of a face of the at least one face with the smallest distance as a speaker.

### Claim 19 (depends on 16)

A speaker recognition system, comprising:
the audio acquisition device according to claim 16; and
an image capture device include an imaging sensor configured to obtain the first image.

### Claim 20

A non-transitory computer-readable storage medium storing instructions which when executed by a processor, cause the processor to perform:
obtaining a first image that includes an audio acquisition device;
identifying the audio acquisition device in the first image;
obtaining first coordinate data of the identified audio acquisition device in the first image;
determining first displacement data according to the first coordinate data and historical coordinate data of the audio acquisition device; and
determining first coordinates of the audio acquisition device according to the first displacement data.
