# US 2008/0187173 A1 · Method and apparatus for tracking video image

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

- Number: US 2008/0187173 A1
- Document ID: US20080187173A1
- Status: Published application
- Kind: A1
- Application number: 11/889,428
- Filed: 2007-08-13
- Published: 2008-08-07
- Earliest claimed: 2007-02-02
- Assignees: Samsung Electronics Co., Ltd.
- Inventors: Jung-bae Kim; Young-su Moon; Gyu-tae Park
- Web page: https://patentgrep.com/patent/US20080187173A1

## CPC

- G06V 40/162: {using pixel segmentation or colour matching}
- G06V 10/24: Aligning, centring, orientation detection or correction of the image
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A video image tracking apparatus and a video image tracking method are provided. The method makes it possible to detect and track a target image having a variety of angles without a multi-view detector, easily adapt to addition of a new target image and removal of an existing target image, reduce calculation time and memory consumption for detecting and tracking the target image having a variety of angles, so that embedded software or chip can be realized, and tracking the target image at high speed.

## Claims

### Claim 1

A video image tracking method comprising:
tracking a target model and determining a target candidate of a tracked frame;
detecting a target image from the tracked frame or a frame subsequent to the tracked frame; and
renewing the target model using the target candidate or the target image and initializing tracking.

### Claim 2 (depends on 1)

The method of claim 1, wherein the target model is tracked using a statistical distribution characteristic of the target candidate and a statistical distribution characteristic of the target model.

### Claim 3 (depends on 1)

The method of claim 1, wherein the renewing of the target model comprises:
if an overlapping region between the target candidate and the target image is greater than a predetermined reference value, removing the target candidate and renewing the target model using the target image.

### Claim 4 (depends on 1)

The method of claim 1, wherein the tracking of the target model comprises:
calculating a similarity or distance between the statistical distribution characteristic of the target model and a statistical distribution characteristic of a target candidate identified as a result of tracking a frame previous to the tracked frame, modifying the location of the target candidate based on the target model and the statistical distribution characteristic of the target candidate, calculating a similarity or distance between the statistical distribution characteristic of the target model and a statistical distribution characteristic of the target candidate according to the modified location of the target candidate, and performing tracking using the similarity or the distance.

### Claim 5 (depends on 1)

The method of claim 1, wherein the target model is determined as a result of detecting a target image from the frame previous to the tracked frame.

### Claim 6 (depends on 2)

The method of claim 2, wherein the statistical distribution characteristic is a color histogram or an edge histogram.

### Claim 7 (depends on 1)

The method of claim 1, wherein the target candidate is determined according to a comparison result obtained by comparing similarity between the target model and the target candidate of the tracked frame and a predetermined reference value.

### Claim 8 (depends on 1)

The method of claim 1, wherein each of an nth frame (n is a positive number greater than 1) through an n+mth frame (m is a positive number) is tracked, and the n+mth frame or a frame subsequent to the n+mth frame further is detected,
wherein the tracking of the target model comprises:
calculating a first similarity between the statistical distribution characteristic of the target model and a statistical distribution characteristic of a first target candidate of the nth frame having the same location as the target model, and determining the location of a second target candidate of the nth frame according to the first similarity;
calculating a second similarity between the statistical distribution characteristic of the target model and a statistical distribution characteristic of the second target candidate having the determined location; and
comparing first and second similarities, selectively determining the location of a third target candidate according to the comparison result, and calculating a third similarity between a statistical distribution characteristic of the third target candidate and the statistical distribution characteristic of the target model,
wherein one of the first, second, and third target candidates that has the maximum similarity value is selected as the target candidate of the tracked frame.

### Claim 9 (depends on 4)

The method of claim 4, wherein the tracking of the target model is based on similarity or distance between the statistical distribution characteristic of the target model and statistical distribution characteristics obtained by regulating the scale of the target candidate.

### Claim 10 (depends on 1)

The method of claim 1, wherein the target image is detected using full face features thereof.

### Claim 11 (depends on 6)

The method of claim 6, wherein the color histogram of the target model is calculated according to the following equation,
q
u
=
∑
i
=
1
n

δ

[
b

(
x
i
)
-
u
]
wherein, xi denotes the pixel location of the target model, b(xi) denotes a bin value of a pixel, u denotes a color of the pixel, and qu denotes a histogram according to the pixel color u.

### Claim 12 (depends on 4)

The method of claim 4, wherein the distance is calculated according to the following equation,
d

(
y
)
=
1
N
p

(
y
)

N
q

∑
u
=
1
m


N
q

p
u

(
y
)
-
N
p

(
y
)

q
u

wherein, d(y) denotes a distance between the target model and the target candidate, Nq denotes the number of pixels of the target model, Np(y) denotes the number of pixels of the target candidate, Pu(y) denotes a color histogram of the target candidate, and qu denotes a color histogram of the target model.

### Claim 13 (depends on 8)

The method of claim 8, wherein the first and second similarities are compared when the second similarity is greater than or the same as the first similarity.

### Claim 14 (depends on 1)

A computer readable recording medium having embodied thereon a computer program for executing the method of claim 1.

### Claim 15

A video image tracking apparatus comprising:
a tracking unit tracking a target model and determining a target candidate of each frame;
a detector detecting a target image at predetermined frame intervals; and
a controller renewing the target model using the target candidate determined by the tracking unit and the target image detected by the detector, and initializing tracking.

### Claim 16 (depends on 13)

The apparatus of claim 13, wherein the tracking unit comprises:
a tracking location determiner determining the target candidate in a frame to be tracked based on a statistical distribution characteristic of the target model; and
a histogram extractor extracting a histogram reflecting a statistical distribution characteristic of the target candidate determined by the tracking location determiner.

### Claim 17 (depends on 15)

The apparatus of claim 15, wherein the controller comprises:
a scheduler managing a tracking process performed by the tracking unit and a detecting process performed by the detector; and
a combiner combining the target candidate and the target image and renewing the target model.

### Claim 18 (depends on 15)

The apparatus of claim 15, wherein, if an overlapping region between the target candidate of the tracked frame and the target image is greater than a predetermined reference value, the combiner removes the target candidate, and the controller initializes tracking by the target image.

### Claim 19 (depends on 15)

The apparatus of claim 15, wherein, if the overlapping region between the target candidate of the tracked frame and the target image is smaller than the predetermined reference value, the controller determines the target image to be a tracking model.

### Claim 20 (depends on 16)

The apparatus of claim 16, wherein the tracking location determiner determines the target candidate of the tracked frame based on statistical distribution characteristics obtained by regulating the scale of the target candidate.