# US 9,437,011 B2 · Method and apparatus for estimating a pose of a head for a person

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

- Number: US 9,437,011 B2
- Document ID: US9437011B2
- Status: Granted
- Kind: B2
- Application number: 13/908,282
- Filed: 2013-06-03
- Published: 2013-12-12
- Granted: 2016-09-06
- Earliest claimed: 2012-06-11
- Assignees: Samsung Electronics Co., Ltd.
- Applicants: Samsung Electronics Co., Ltd.
- Inventors: Hariprasad Kannan; Anant Vidur Puri
- Web page: https://patentgrep.com/patent/US9437011B2

## CPC

- G06V 40/171: {Local features and components; Facial parts (eye characteristics G06V40/18); Occluding parts, e.g. glasses; Geometrical relationships}
- G06T 7/75: {involving models}
- G06V 40/167: {using comparisons between temporally consecutive images}
- G06T 2207/10016: Video; Image sequence
- G06T 2207/30201: Face
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A method of estimating a pose of a head for a person, includes estimating the pose of the head for the person based on a content, and generating a three-dimensional (3D) model of a face for the person. The method further includes generating pictorial structures of the face based on the estimated pose and the 3D model, and determining a refined pose of the head by locating parts of the face in the pictorial structures.

## Claims

### Claim 1

A method of estimating a refined pose of a head for a person,
the method comprising:
estimating a rough pose of the head for the person based on a two-dimensional (2D) image;
generating a three-dimensional (3D) model of a face for the person;
rotating the 3D model based on the estimated rough pose;
determining parts of the face from the rotated 3D model;
projecting the rotated 3D model and the determined parts of the face to a 2D model; and
determining the refined pose of the head based on the 2D model.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
capturing the 2D image of preference to a user, the 2D image comprising a video and/or an image.

### Claim 3 (depends on 1)

The method of claim 1, further comprising:
tracking the head in the 2D image to generate a tracked head image; and
separating a foreground and a background in the tracked head image to generate a separated image.

### Claim 4 (depends on 3)

The method of claim 3, further comprising:
extracting the outline of the tracked head from the separated image.

### Claim 5 (depends on 3)

The method of claim 3, further comprising:
estimating the rough pose of the tracked head in the separated image.

### Claim 6 (depends on 1)

The method of claim 1, further comprising:
generating the 3D model based on a personalization of a predetermined model of a head.

### Claim 7 (depends on 1)

The method of claim 1, further comprising:
determining locations of the determined parts of the face with respect to a root part of the face.

### Claim 8 (depends on 1)

The method of claim 1, further comprising:
determining vectors from a center of the head to respective centers of the determined parts based on an outline of the head; and
determining the refined pose based on the vectors mapped to the refined pose in a predetermined lookup table.

### Claim 9 (depends on 1)

A non-transitory computer-readable storage medium storing a program comprising instructions to cause a computer to perform the method of claim 1.

### Claim 10

An apparatus that estimates a refined pose of a head for a person, the apparatus comprising:
a processor configured to
estimate a rough pose of the head for the person based on a two-dimensional (2D) image,
generate a three-dimensional (3D) model of a face for the person,
rotate the 3D model based on the estimated rough pose,
determine parts of the face from the estimated rotated 3D model,
project the rotated 3D model and the determined parts of the face to a 2D model, and
determine the refined pose of the head based on the 2D model.

### Claim 11 (depends on 10)

The apparatus of claim 10, wherein the processor is further configured to:
capture the 2D image of preference to a user, the 2D image comprising a video and/or an image.

### Claim 12 (depends on 10)

The apparatus of claim 10, wherein the processor is further configured to:
track the head in the 2D image to generate a tracked head image; and
separate a foreground and a background in the tracked head image to generate a separated image.

### Claim 13 (depends on 12)

The apparatus of claim 12, wherein the processor is further configured to:
extract an outline of the tracked head from the separated image.

### Claim 14 (depends on 12)

The apparatus of claim 12, wherein the processor is further configured to:
estimate the rough pose of the tracked head in the separated image.

### Claim 15 (depends on 10)

The apparatus of claim 10, wherein the processor is further configured to:
generate the 3D model based on a personalization of a predetermined model of a head.

### Claim 16 (depends on 10)

The apparatus of claim 10, wherein the processor is further configured to:
determine locations of the determined parts of the face with respect to a root part of the face.

### Claim 17 (depends on 10)

The apparatus of claim 10, wherein the processor is further configured to:
determine vectors from a center of the head to respective centers of the determined parts based on an outline of the head; and
determine the refined pose based on the vectors mapped to the refined pose in a predetermined lookup table.

### Claim 18 (depends on 1)

The method of claim 1, further comprising determining the refined pose based on an outline of the head.

### Claim 19 (depends on 10)

The apparatus of claim 10, wherein the processor is further configured to: determine the refined pose based on an outline of the head.

### Claim 20 (depends on 1)

The method of claim 1, wherein the rough pose comprises an approximate range of an angle of the head.