# US 2021/0174073 A1 · A Method and Apparatus for Swimmer Tracking

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

- Number: US 2021/0174073 A1
- Document ID: US20210174073A1
- Status: Published application
- Kind: A1
- Application number: 17/048,097
- Filed: 2019-04-14
- Published: 2021-06-10
- Earliest claimed: 2018-04-16
- Assignees: Lynxight Ltd.
- Applicants: Lynxight Ltd.
- Inventors: Omer Bar-Ilan; Roee Mor
- Web page: https://patentgrep.com/patent/US20210174073A1

## CPC

- G06V 40/23: {Recognition of whole body movements, e.g. for sport training}
- G06T 7/246: using feature-based methods, e.g. the tracking of corners or segments
- G06V 20/52: Surveillance or monitoring of activities, e.g. for recognising suspicious objects (recognising microscopic objects G06V20/69)
- G06V 40/10: Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06T 2207/30196: Human being; Person
- G06T 2207/30221: Sports video; Sports image
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A method, system and computer program product, the method comprising: obtaining a first multiplicity of images from an overhead capture device overlooking a body of water; detecting a swimmer in some of the images, wherein the swimmer is outside the body of water or separated from other swimmers; determining two descriptors of the person, based on the images, the first descriptor related to a portion of the swimmer outside the water, the second related to a portion inside the water, the descriptors determined upon visual characteristics extracted from the images; obtaining a second multiplicity of images from the device; and tracking the swimmer in the second multiplicity of images, said tracking comprises detecting the swimmer using either descriptor, said detecting the swimmer is performed with respect to an image from the second multiplicity of images in which the swimmer is partially occluded by other swimmers or adjacent to other swimmers.

## Claims

### Claim 1

A method comprising:
obtaining a first multiplicity of images taken by at least one overhead capture device overlooking a body of water having a surface;
detecting a swimmer in at least some images from the first multiplicity of images, wherein the swimmer is outside the body of water or separated from other swimmers in at least some of the first multiplicity of images;
determining a first descriptor and a second descriptor of the person, based on the at least some images from the first multiplicity of images,
wherein the first descriptor is related to a portion the swimmer that is outside the body of water,
wherein the second descriptor is related to a portion of the swimmer that is inside the body of water,
the first descriptor and the second descriptor are determined based on visual characteristics extracted from the at least some images;
obtaining a second multiplicity of images taken by the at least one overhead capture device; and
tracking the swimmer in the second multiplicity of images, wherein said tracking comprises detecting the swimmer using at least one of the first descriptor and the second descriptor, wherein said detecting the swimmer is performed with respect to at least one image in which the swimmer is at least one image from the second multiplicity of images in which the swimmer is at least partially occluded by other swimmers or adjacent to other swimmers.

### Claim 2 (depends on 1)

The method of claim 1, wherein the second descriptor is determined based on the first descriptor.

### Claim 3 (depends on 1)

The method of claim 1, wherein the second descriptor is determined based on the first descriptor and predetermined modifications.

### Claim 4 (depends on 1)

The method of claim 1, further comprising:
determining whether the swimmer is above or below the water in at least some of the second multiplicity of images; and
subject to determining that the swimmer is below the water in at least some of the second multiplicity of images, taking an action.

### Claim 5 (depends on 4)

The method of claim 4, wherein the action is selected from the group consisting of: setting off an alarm; sounding an alarm; generating a visual or vibrating alert; and notifying a life guard or the like.

### Claim 6 (depends on 1)

The method of claim 1, further comprising:
analyzing presence of the swimmer in the pool and its environment; and
reporting the statistics to an operator.

### Claim 7 (depends on 1)

The method of claim 1, further comprising:
analyzing presence of the swimmer in the pool and its environment; and
taking an action related to the presence.

### Claim 8 (depends on 1)

The method of claim 1, wherein tracking the person is performed also in an image in which a first part of the person is separated from a second part of the person by a part of another person.

### Claim 9 (depends on 1)

The method of claim 1, wherein one device of the at least one overhead capture device has a viewing angle of at most 70° between a first line connecting the device and a projection of the device on a plane containing the surface, and a second line connecting the device and point on the surface farthest from the device.

### Claim 10 (depends on 1)

The method of claim 1, wherein the at least one overhead capture device captures an entire surface the body of water.

### Claim 11 (depends on 1)

The method of claim 1, wherein said detecting comprises differentiating a background and foreground of the image.

### Claim 12 (depends on 1)

The method of claim 1, wherein the first descriptor or the second descriptor is updated based on the at least one visual attribute as extracted from images in the second multiplicity of images.

### Claim 13 (depends on 1)

The method of claim 1, wherein the first descriptor or the second descriptor is selected from the group consisting of: a color histogram; an edge; an edge histogram; a gradient histogram; a contour edge between colors; a contour edge between shapes; a distinct feature; a unique pattern; a corner; an edge; a color co-occurrence matrix, and a texture.

### Claim 14

A system having a processor, the processor being adapted to perform the steps of:
obtaining a first multiplicity of images taken by at least one overhead capture device overlooking a body of water having a surface;
detecting a swimmer in at least some images from the first multiplicity of images, wherein the swimmer is outside the body of water or separated from other swimmers in at least some of the first multiplicity of images;
determining a first descriptor and a second descriptor of the person, based on the at least some images from the first multiplicity of images,
wherein the first descriptor is related to a portion the swimmer that is outside the body of water,
wherein the second descriptor is related to a portion of the swimmer that is inside the body of water,
the first descriptor and the second descriptor are determined based on visual characteristics extracted from the at least some images;
obtaining a second multiplicity of images taken by the at least one overhead capture device; and
tracking the swimmer in the second multiplicity of images, wherein said tracking comprises detecting the swimmer using at least one of the first descriptor and the second descriptor, wherein said detecting the swimmer is performed with respect to at least one image in which the swimmer is at least one image from the second multiplicity of images in which the swimmer is at least partially occluded by other swimmers or adjacent to other swimmers.

### Claim 15

A computer program product comprising a non-transitory computer readable medium retaining program instructions, which instructions when read by a processor, cause the processor to perform:
obtaining a first multiplicity of images taken by at least one overhead capture device overlooking a body of water having a surface;
detecting a swimmer in at least some images from the first multiplicity of images, wherein the swimmer is outside the body of water or separated from other swimmers in at least some of the first multiplicity of images;
determining a first descriptor and a second descriptor of the person, based on the at least some images from the first multiplicity of images,
wherein the first descriptor is related to a portion the swimmer that is outside the body of water,
wherein the second descriptor is related to a portion of the swimmer that is inside the body of water,
the first descriptor and the second descriptor are determined based on visual characteristics extracted from the at least some images;
obtaining a second multiplicity of images taken by the at least one overhead capture device; and
tracking the swimmer in the second multiplicity of images, wherein said tracking comprises detecting the swimmer using at least one of the first descriptor and the second descriptor, wherein said detecting the swimmer is performed with respect to at least one image from the second multiplicity of images in which the swimmer is at least partially occluded by other swimmers or adjacent to other swimmers.
