# US 2021/0176444 A1 · Video conference panoramic image spreading method

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

- Number: US 2021/0176444 A1
- Document ID: US20210176444A1
- Status: Published application
- Kind: A1
- Application number: 16/782,236
- Filed: 2020-02-05
- Published: 2021-06-10
- Earliest claimed: 2019-12-09
- Assignees: Magic Control Technology Corporation
- Applicants: Magic Control Technology Corp.
- Inventors: PEI-CHUNG LIU
- Web page: https://patentgrep.com/patent/US20210176444A1

## CPC

- H04N 7/147: {Communication arrangements, e.g. identifying the communication as a video-communication, intermediate storage of the signals (selecting H04Q)}
- H04N 13/111: Transformation of image signals corresponding to virtual viewpoints, e.g. spatial image interpolation
- H04N 7/157: {defining a virtual conference space and using avatars or agents (computer conference optimisation or adaptation H04L12/1827)}
- H04N 13/282: for generating image signals corresponding to three or more geometrical viewpoints, e.g. multi-view systems
- H04N 23/611: where the recognised objects include parts of the human body
- H04N 23/698: for achieving an enlarged field of view, e.g. panoramic image capture
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

A video conference panoramic image spreading method is provided, in which an image capturing (acquiring) step is applied by using at least one panoramic camera to capture at least one video conference panoramic picture or image and a bodily human features recognition step is applied to achieve bodily human features recognition for recognizing each of conferee photographed subjects in the video conference panoramic picture or image, and then, based on the result of bodily human features recognition, steps of searching out a dividing point for a two-dimensional (2D)spread picture, image cutting and two-dimensional picture spreading, and stitching are carried out so as to avoid, in a real-time manner, conflicts of cutting of a human body in the conference panoramic picture or image based on the dividing point for two-dimensional spreading and correct the dividing point, for acquiring a two-dimensional panoramic picture that instantaneously reflects and completely and correctly corresponds to a primitive video conference panoramic picture.

## Claims

### Claim 1

A video conference panoramic image spreading method, comprising the following steps:
(a) acquiring a primitive panoramic picture or image, in which at least one panoramic camera is used to photograph and capture and thus acquire at least one primitive video conference panoramic picture or image;
(b) performing bodily human features recognition, in which conferee photographed subjects contained in the at least one primitive video conference panoramic picture or image that is photographed and captured and thus acquired by the panoramic camera in Step (a) is individually subjected to bodily human features recognition;
(c) searching out a dividing point for two-dimensional picture spreading and defining the dividing point as a starting point for the two-dimensional panoramic image spreading, in which searching is performed in the primitive video conference panoramic picture or image in which the conferee photographed subjects have been subjected to bodily human features recognition in Step (b) to find a dividing point that does not lead to cutting or dividing of bodily human features of the conferee photographed subjects, and the dividing point is taken as the starting point for spreading a two-dimensional panoramic image;
(d) intercepting desired image regions according to the dividing point, in which cutting and intercepting regions of the primitive video conference panoramic picture or image are performed according to the dividing point found in Step (c), wherein cutting and intercepting are carried out primarily based on a region of each of the conferee photographed subjects in the primitive video conference panoramic picture or image in order to acquire at least one quasi-panoramic image;
(e) carrying out combining and stitching on the quasi-panoramic image by using the starting point for two-dimensional panoramic image spreading, in which the quasi-panoramic image intercepted in Step (d) is subjected to combining and stitching according to the starting point for two-dimensional panoramic image spreading, which is the dividing point of Step (c); and
(f) forming a two-dimensional panoramic picture, in which the quasi-panoramic image that has been subjected to combining and stitching in Step (e) forms a spread two-dimensional panoramic picture.

### Claim 2 (depends on 1)

The video conference panoramic image spreading method according to claim 1, wherein the panoramic camera of Step (a) comprises a panoramic camera having a fisheye lens.

### Claim 3 (depends on 1)

The video conference panoramic image spreading method according to claim 1, wherein the panoramic camera of Step (a) comprises a panoramic camera having multiple lenses.

### Claim 4 (depends on 1)

The video conference panoramic image spreading method according to claim 1, wherein bodily human features recognition of Step (b) is applied as features recognition for at least one or a combination of at least two of human face, facial features, upper body contour, and dynamic face image of the conferee photographed subjects.

### Claim 5 (depends on 1)

The video conference panoramic image spreading method according to claim 1, wherein an operation of searching out the dividing point for two-dimensional picture spreading and defining the dividing point as the starting point of the panoramic image carried out in Step (c) further comprises:
(c1) randomly selecting a test dividing point, in which a test dividing point is randomly picked in the primitive video conference panoramic picture or image of which the conferee photographed subjects contained therein have been individually subjected to the bodily human features recognition in Step (b);
(c2) identifying if the test dividing point cuts and divides bodily human features and carrying out Step (c21), if positive, and carrying out Step (c22), if negative, in which determination is made on whether bodily human features of any one of the conferee photographed subjects of Step (c1) is cut and divided by the test dividing point picked in Step (c1);
(c21) moving the test dividing point, in which the test dividing point of Step (c1) is moved in a clockwise direction or a counterclockwise direction and repeating Step (c2);
(c22) setting the test dividing point as the dividing point for two-dimensional picture spreading, in which a test dividing point that does not cut and divide, in Step (c2), the bodily human features of any one of the conferee photographed subjects of Step (c1) is taken as the dividing point for spreading of the two-dimensional picture; and
(c3) ending.

### Claim 6 (depends on 1)

The video conference panoramic image spreading method according to claim 1, wherein Step (a)-Step (f) are provided as image processing software pre-loaded in the panoramic camera of Step (a).

### Claim 7 (depends on 1)

The video conference panoramic image spreading method according to claim 1, wherein the panoramic camera of Step (a) is connected to at least one image processor, and Step (a)-Step (f) are provided as image processing software pre-loaded in the image processor.

### Claim 8 (depends on 1)

The video conference panoramic image spreading method according to claim 1, wherein the panoramic camera of Step (a) is connected to at least one smart communication device, and Step (a)-Step (f) are provided as image processing software pre-loaded in the smart communication device.

### Claim 9 (depends on 8)

The video conference panoramic image spreading method according to claim 8, wherein the smart communication device to which the panoramic camera of Step (a) is connected comprises a smart mobile phone.

### Claim 10 (depends on 8)

The video conference panoramic image spreading method according to claim 8, wherein the smart communication device to which the panoramic camera of Step (a) is connected comprises a tablet computer.

### Claim 11 (depends on 1)

The video conference panoramic image spreading method according to claim 1, wherein the panoramic camera of Step (a) is connected to at least one desk-top computer, and Step (a)-Step (f) are provided as image processing software pre-loaded in the desk-top computer.

### Claim 12 (depends on 1)

The video conference panoramic image spreading method according to claim 1, wherein the panoramic camera of Step (a) is connected to at least one notebook computer, and Step (a)-Step (f) are provided as image processing software pre-loaded in the notebook computer.
