# US 2022/0067347 A1 · Systems and methods for matching facial images to reference images

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

- Number: US 2022/0067347 A1
- Document ID: US20220067347A1
- Status: Published application
- Kind: A1
- Application number: 17/410,731
- Filed: 2021-08-24
- Published: 2022-03-03
- Earliest claimed: 2020-08-27
- Assignees: The Code Dating LLC
- Applicants: The Code Dating LLC
- Inventors: William Reinisch; Stacy Meredith Schneider; Charles Anthony Sankowich
- Web page: https://patentgrep.com/patent/US20220067347A1

## CPC

- G06V 10/764: using classification, e.g. of video objects
- G06V 40/168: {Feature extraction; Face representation}
- G06V 40/16: Human faces, e.g. facial parts, sketches or expressions
- G06V 40/172: {Classification, e.g. identification}
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A facial feature matching system comprises a facial feature matching engine. A first user selection of reference facial images is received, and facial features of reference faces are characterized using the facial feature recognition engine comprising a neural network with input, hidden, and output layers. The facial features are weighted. The weighted facial features are used to identify users that have facial features similar to the weighted facial features, wherein the respective reference faces include faces different than the faces of the users. Similarity indicators are generated for the identified users. The generated respective similarity indicators are used to generate a ordering of the identified users which is rendered via the user device. A first user selection of a second user in the ordered identified users is received and the first user and the second user are enabled to communicate over an electronic communication channel.

## Claims

### Claim 1

A facial feature matching system, comprising:
a network interface;
at least one processing device operable to:
populate a first user interface, for display on a user device associated with a first user, with a set of images comprising respective reference facial images of different people;
receive, via the network interface, from the user device a selection by the first user of a plurality of reference facial images in the set of images;
characterize a plurality of facial features of respective reference faces using a facial feature recognition engine, the respective faces comprising faces included in the reference facial images selected by the first user;
apply respective weights to the plurality of facial features of respective reference faces;
use the weighted plurality of facial features to identify a plurality of users that have one or more facial features similar to one or more of the weighted plurality of facial features, wherein the respective reference faces include faces different than the faces of the plurality of users;
generate respective similarity scores for the identified plurality of users that have one or more facial features similar to one or more of the weighted plurality of facial features;
use the generated respective similarity scores for the identified plurality of users that have one or more facial features similar to one or more of the weighted plurality of facial features to generate a ranking of the identified plurality of users;
cause a second user interface to be displayed on the user device, the second user interface populated, in ranked order in association with respective similarity scores, with the identified plurality of users that have one or more facial features similar to one or more of the weighted plurality of facial features;
receive, via the second user interface, a selection, by the first user, from the user device, of a second user in the plurality of users that have one or more facial features similar to one or more of the weighted plurality of facial features; and
enable the first user and the second user to communicate over at least a first electronic communication channel.

### Claim 2 (depends on 1)

The facial feature matching system as defined in claim 1, the system further configured to perform operation comprising:
filtering reference images in a database of reference images to generate the set of images comprising respective reference facial images based at least in part on non-appearance-related preferences.

### Claim 3 (depends on 1)

The facial feature matching system as defined in claim 1, the system further configured to perform operations comprising:
using a neural network comprising:
an input layer,
one or more hidden layers, and
an output layer,
to identify the plurality of users that have one or more facial features similar to the one or more of the weighted plurality of facial features.

### Claim 4 (depends on 1)

The facial feature matching system as defined in claim 1, the system further configured to perform operations comprising:
receiving an image from a given user;
verifying timing information associated with the image of the given user; and
at least partly in response to a successful verification of timing information associated with the image of the given user, including the image of the given user including the image of the third user in a first database comprising user images.

### Claim 5 (depends on 1)

The facial feature matching system as defined in claim 1, the system further configured to perform operations comprising:
drawing respective light masks of vectors for respective facial features of a given face in at least one image; and
weighting respective light masks to enhance accuracy in identifying facial feature similarity.

### Claim 6 (depends on 1)

The facial feature matching system as defined in claim 1, the system further configured to perform operations comprising: using a chain code in performing a process of matching a feature of a given reference face with a feature of a given user.

### Claim 7 (depends on 1)

The facial feature matching system as defined in claim 1, wherein a given similarity score comprises a percentage similarity score.

### Claim 8

A computerized method, the method comprising:
populating, using a computer system comprising hardware, a first user interface, for display on a user device associated with a first user, with images comprising respective reference facial images of different people;
receiving, at the computer system from the user device, a selection by the first user of a plurality of reference facial images;
characterizing a plurality of facial features of respective reference faces using a facial feature recognition engine, the respective faces comprising faces included in the reference facial images selected by the first user;
applying respective weights to the plurality of facial features of respective reference faces;
using the weighted plurality of facial features to identify a plurality of users that have one or more facial features similar to one or more of the weighted plurality of facial features, wherein the respective reference faces include faces different than the faces of the plurality of users;
generating respective similarity scores for the identified plurality of users that have one or more facial features similar to one or more of the weighted plurality of facial features;
using the generated respective similarity scores for the identified plurality of users that have one or more facial features similar to one or more of the weighted plurality of facial features to generate a ranking of the identified plurality of users;
causing a second user interface to be displayed on the user device, the second user interface populated, in ranked order, with the identified plurality of users that have one or more facial features similar to one or more of the weighted plurality of facial features;
receiving, via the second user interface, a selection, by the first user, from the user device, of a second user in the plurality of users that have one or more facial features similar to one or more of the weighted plurality of facial features; and
enabling the first user and the second user to communicate over at least a first electronic communication channel.

### Claim 9 (depends on 8)

The computerized method as defined in claim 8, the method further comprising:
filtering reference images in a database of reference images to generate the images comprising respective reference facial images based at least in part on non-appearance-related preferences.

### Claim 10 (depends on 8)

The computerized method as defined in claim 8, the method further comprising:
using a neural network comprising:
an input layer,
one or more hidden layers, and
an output layer,
to identify the plurality of users that have one or more facial features similar to the one or more of the weighted plurality of facial features.

### Claim 11 (depends on 8)

The computerized method as defined in claim 8, the method further comprising:
receiving an image from a given user;
verifying timing information associated with the image of the given user; and
at least partly in response to a successful verification of timing information associated with the image of the given user, including the image of the given user including the image of the third user in a first database comprising user images.

### Claim 12 (depends on 8)

The computerized method as defined in claim 8, the method further comprising:
drawing respective light masks of vectors for respective facial features of a given face in at least one image; and
weighting respective light masks to enhance accuracy in identifying facial feature similarity.

### Claim 13 (depends on 8)

The computerized method as defined in claim 8, the method further comprising: using a chain code in performing a process of matching a feature of a given reference face with a feature of a given user.

### Claim 14 (depends on 8)

The computerized method as defined in claim 8, wherein a given similarity score comprises a percentage similarity score.

### Claim 15

A non-transitory computer-readable medium comprising instructions that when executed by a processor, cause the processor to perform operations comprising:
populating a first user interface, for display on a user device associated with a first user, with images comprising respective reference facial images of different people;
receiving a selection by the first user of reference facial images;
characterizing a plurality of facial features of respective reference faces using a facial feature recognition engine, the respective faces comprising faces included in the reference facial images selected by the first user;
applying respective weights to the plurality of facial features of respective reference faces;
using the weighted plurality of facial features to identify users that have one or more facial features similar to one or more of the weighted plurality of facial features, wherein the respective reference faces include faces different than the faces of the plurality of users;
generating respective similarity indicators for the identified users that have one or more facial features similar to one or more of the weighted plurality of facial features;
using the generated respective similarity indicators for the identified users that have one or more facial features similar to one or more of the weighted plurality of facial features to generate a ordering of the identified users;
causing a second user interface to be displayed on the user device, the second user interface populated, in accordance with the ordering, with the identified plurality of users that have one or more facial features similar to one or more of the weighted plurality of facial features;
receiving, via the second user interface, a selection, by the first user, from the user device, of a second user users that have one or more facial features similar to one or more of the weighted plurality of facial features; and
enabling the first user and the second user to communicate over at least a first electronic communication channel.

### Claim 16 (depends on 15)

The non-transitory computer-readable medium as defined in claim 15, the operations further comprising:
filtering reference images in a database of reference images to generate the images comprising respective reference facial images based at least in part on non-appearance-related preferences.

### Claim 17 (depends on 15)

The non-transitory computer-readable medium as defined in claim 15, the operations further comprising:
using a neural network comprising:
an input layer,
one or more hidden layers, and
an output layer,
to identify the plurality of users that have one or more facial features similar to the one or more of the weighted plurality of facial features.

### Claim 18 (depends on 15)

The non-transitory computer-readable medium as defined in claim 15, the operations further comprising:
receiving an image from a given user;
verifying timing information associated with the image of the given user; and
at least partly in response to a successful verification of timing information associated with the image of the given user, including the image of the given user including the image of the third user in a first database comprising user images.

### Claim 19 (depends on 15)

The non-transitory computer-readable medium as defined in claim 15, the operations further comprising:
drawing respective light masks of vectors for respective facial features of a given face in at least one image; and
weighting respective light masks to enhance accuracy in identifying facial feature similarity.

### Claim 20 (depends on 15)

The non-transitory computer-readable medium as defined in claim 15, the operations further comprising: using a chain code in performing a process of matching a feature of a given reference face with a feature of a given user.

### Claim 21 (depends on 15)

The non-transitory computer-readable medium as defined in claim 15, wherein a given similarity indicator comprises a percentage similarity indicator.
