# US 2021/0279472 A1 · Systems and methods for anti-spoofing protection using motion detection and video background analysis

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

- Number: US 2021/0279472 A1
- Document ID: US20210279472A1
- Status: Published application
- Kind: A1
- Application number: 17/189,454
- Filed: 2021-03-02
- Published: 2021-09-09
- Earliest claimed: 2020-03-03
- Assignees: Cyberlink Corp.
- Applicants: CyberLink Corp.
- Inventors: Fu-Kai CHUANG; Sheng-Hung LIU
- Web page: https://patentgrep.com/patent/US20210279472A1

## CPC

- G06V 20/41: {Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items (segmenting video sequences G06V20/49)}
- G06V 40/16: Human faces, e.g. facial parts, sketches or expressions
- G06T 7/215: Motion-based segmentation
- G06V 40/166: {using acquisition arrangements}
- G06V 40/40: Spoof detection, e.g. liveness detection
- G06T 2207/10016: Video; Image sequence
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A computing device captures a live video of a user. For a first frame of the live video, the computing device obtains first target positional coordinates of a first target point located a first predetermined distance from the computing device and obtains first background data. For a second frame, the computing device obtains second target positional coordinates of a second target point located a second predetermined distance from the computing device and obtains second background data. The computing device calculates a target motion vector based on the first target point and the second target point and calculates a background motion vector based on feature points in the first background data and the second background data. The computing device determines a difference value between the target motion vector and the background motion vector and determines whether the user is spoofing the computing device based on the difference value.

## Claims

### Claim 1

A method implemented in a computing device, comprising:
capturing a live video of a user;
for a first frame of the live video:
obtaining first target positional coordinates of a first target point located a first predetermined distance from the computing device; and
obtaining first background data;
for a second frame of the live video:
obtaining second target positional coordinates of a second target point located a second predetermined distance from the computing device; and
obtaining second background data;
calculating a target motion vector based on the first target point and the second target point;
calculating a background motion vector based on feature points in the first background data of the first frame and the second background data in the second frame;
determining a difference value between the target motion vector and the background motion vector; and
determining whether the user is spoofing the computing device based on the difference value.

### Claim 2 (depends on 1)

The method of claim 1, wherein obtaining the first background data and the second background data comprises performing facial detection and filtering all individuals depicted in the first frame and the second frame.

### Claim 3 (depends on 1)

The method of claim 1, wherein calculating the target motion vector comprises:
obtaining first yaw data, first pitch data, and first roll data of the computing device, second yaw data, second pitch data, and second roll data of the computing device, first positional coordinates of the computing device, and second positional coordinates of the computing device;
performing three dimensional (3D) coordinate transformation on the first target point based on the first yaw data, the first pitch data, and the first roll data of the computing device, the second yaw data, the second pitch data, the second roll data, the first positional coordinates of the computing device, and the second positional coordinates of the computing device to generate 3D coordinates of the first target point;
determining a difference value between the 3D coordinates of the first target point and 3D coordinates of the second target point to generate a 3D target motion vector; and
transforming the 3D target motion vector into a two dimensional (2D) target motion vector.

### Claim 4 (depends on 3)

The method of claim 3, wherein the first and second positional coordinates of the computing device, first and second yaw data, first and second pitch data, and first and second roll data of the computing device are obtained based on acceleration measured by a gyroscope in the computing device and displacement measured by an accelerometer in the computing device.

### Claim 5 (depends on 3)

The method of claim 3, wherein the first target positional coordinates of the first target point located and the second target positional coordinates of the second target point located from the computing device are calculated based on the first positional coordinates, the first yaw data, first pitch data, first roll data of the computing device, the second positional coordinates, the second yaw data, second pitch data, and second roll data of the computing device.

### Claim 6 (depends on 3)

The method of claim 3, wherein the predetermined distances in which the first target point and the second target point are located from the computing device are real number multiples of a distance (x) such that the predetermined distance is equal to n*(x), wherein n>1, x is a focal distance, and wherein a value of (n) is set based on an average distance between an object in a background and the computing device.

### Claim 7 (depends on 6)

The method of claim 6, wherein a determination is made that the user is spoofing the computing device when the difference value is greater than a threshold value.

### Claim 8 (depends on 1)

The method of claim 1, wherein calculating the target motion vector comprises:
for a first frame of the live video:
obtaining the first target point in a first vertical plane located the first predetermined distance from the computing device, wherein the first target point is a real number multiple of a distance (x) such that the first predetermined distance is equal to n*(x), wherein n=1 and x is a focal distance; and
modifying the first target point in the first vertical plane based on a first camera focus to determine a first point in the first vertical plane outside a facial region of the user;
for a second frame of the live video:
obtaining the second target point in a second vertical plane equal to the second predetermined distance n*(x) to obtain a focal point of the camera in the facial region of the user, wherein n=1; and
modifying the second target point in the second vertical plane based on a second camera focus to determine a second point in the second vertical plane outside the facial region of the user; and
determining a difference value between the second target point and the first target point to generate the target motion vector.

### Claim 9 (depends on 8)

The method of claim 8, wherein a determination is made that the user is spoofing the computing device when the difference value is less than a threshold value.

### Claim 10 (depends on 1)

The method of claim 1, wherein determining the difference value between the target motion vector and the background motion vector comprises determining cosine similarity between the target motion vector and the background motion vector.

### Claim 11 (depends on 1)

The method of claim 1, wherein calculating the background motion vector based on feature points in the first frame of first background data and the second frame of the second background data comprises:
selecting a plurality of feature points in the first background data and in the second background data meeting a threshold degree of similarity;
subtracting feature points in the second background data from corresponding feature points in the first background data to generate difference reference values; and
averaging the reference values to generate the background motion vector.

### Claim 12

A system, comprising:
a memory storing instructions;
a processor coupled to the memory and configured by the instructions to at least:
capture a live video of a user;
for a first frame of the live video:
obtain first target positional coordinates of a first target point located a first predetermined distance from the system; and
obtain first background data;
for a second frame of the live video:
obtain second target positional coordinates of a second target point located a second predetermined distance from the system; and
obtain second background data;
calculate a target motion vector based on the first target point and the second target point;
calculate a background motion vector based on feature points in the first background data of the first frame and the second background data in the second frame;
determine a difference value between the target motion vector and the background motion vector; and
determine whether the user is spoofing the system based on the difference value.

### Claim 13 (depends on 12)

The system of claim 12, wherein the processor calculates the target motion vector by:
obtaining first yaw data, first pitch data, and first roll data of the system, second yaw data, second pitch data, and second roll data of the system, first positional coordinates of the system, and second positional coordinates of the system;
performing three dimensional (3D) coordinate transformation on the first target point based on the first yaw data, the first pitch data, and the first roll data of the system, the second yaw data, the second pitch data, the second roll data, the first positional coordinates of the system, and the second positional coordinates of the system to generate 3D coordinates of the first target point;
determining a difference value between the 3D coordinates of the first target point and 3D coordinates of the second target point to generate a 3D target motion vector; and
transforming the 3D target motion vector into a two dimensional (2D) target motion vector.

### Claim 14 (depends on 13)

The system of claim 13, wherein the predetermined distances in which the first target point and the second target point are located from the system are real number multiples of a distance (x) such that the predetermined distance is equal to n*(x), wherein n>1, and wherein a value of (n) is set based on an average distance between an object in a background and the system.

### Claim 15 (depends on 14)

The system of claim 14, wherein the processor determines that the user is spoofing the system when the difference value is greater than a threshold value.

### Claim 16 (depends on 12)

The system of claim 12, wherein the processor calculates the target motion vector by:
for a first frame of the live video:
obtaining the first target point in a first vertical plane located the first predetermined distance from the computing device, wherein the first target point is a real number multiple of a distance (x) such that the first predetermined distance is equal to n*(x), wherein n=1 and x is a focal distance; and
modifying the first target point in the first vertical plane based on a first camera focus to determine a first point in the first vertical plane outside a facial region of the user;
for a second frame of the live video:
obtaining the second target point in a second vertical plane equal to the second predetermined distance n*(x) to obtain a focal point of the camera in the facial region of the user, wherein n=1; and
modifying the second target point in the second vertical plane based on a second camera focus to determine a second point in the second vertical plane outside the facial region of the user; and
determining a difference value between the second target point and the first target point to generate the target motion vector.

### Claim 17 (depends on 16)

The system of claim 16, wherein the processor determines that the user is spoofing the system when the difference value is less than a threshold value.

### Claim 18

A non-transitory computer-readable storage medium storing instructions to be implemented by a computing device having a processor, wherein the instructions, when executed by the processor, cause the computing device to at least:
capture a live video of a user;
for a first frame of the live video:
obtain first target positional coordinates of a first target point located a first predetermined distance from the computing device; and
obtain first background data;
for a second frame of the live video:
obtain second target positional coordinates of a second target point located a second predetermined distance from the computing device; and
obtain second background data;
calculate a target motion vector based on the first target point and the second target point;
calculate a background motion vector based on feature points in the first background data of the first frame and the second background data in the second frame;
determine a difference value between the target motion vector and the background motion vector; and
determine whether the user is spoofing the computing device based on the difference value.

### Claim 19 (depends on 18)

The non-transitory computer-readable storage medium of claim 18, wherein the processor calculates the target motion vector by:
obtaining first yaw data, first pitch data, and first roll data of the computing device, second yaw data, second pitch data, and second roll data of the computing device, first positional coordinates of the computing device, and second positional coordinates of the computing device;
performing three dimensional (3D) coordinate transformation on the first yaw data, the first pitch data, and the first roll data of the computing device, the second yaw data, the second pitch data, the second roll data, the first positional coordinates of the computing device, and the second positional coordinates of the computing device to generate 3D coordinates of the first target point;
determining a difference value between the 3D coordinates of the first target point and 3D coordinates of the second target point to generate a 3D target motion vector; and
transforming the 3D target motion vector into a two dimensional (2D) target motion vector.

### Claim 20 (depends on 18)

The non-transitory computer-readable storage medium of claim 18, wherein the processor calculates the target motion vector by:
for a first frame of the live video:
obtaining the first target point in a first vertical plane located the first predetermined distance from the computing device, wherein the first target point is a real number multiple of a distance (x) such that the first predetermined distance is equal to n*(x), wherein n=1 and x is a focal distance; and
modifying the first target point in the first vertical plane based on a first camera focus to determine a first point in the first vertical plane outside a facial region of the user;
for a second frame of the live video:
obtaining the second target point in a second vertical plane equal to the second predetermined distance n*(x) to obtain a focal point of the camera in the facial region of the user, wherein n=1; and
modifying the second target point in the second vertical plane based on a second camera focus to determine a second point in the second vertical plane outside the facial region of the user; and
determining a difference value between the second target point and the first target point to generate the target motion vector.
