# US 2021/0174129 A1 · Information processing apparatus, control method, and program

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

- Number: US 2021/0174129 A1
- Document ID: US20210174129A1
- Status: Published application
- Kind: A1
- Application number: 16/759,164
- Filed: 2017-11-07
- Published: 2021-06-10
- Earliest claimed: 2017-11-07
- Assignees: NEC Corporation
- Applicants: NEC Corporation
- Inventors: Kota IWAMOTO
- Web page: https://patentgrep.com/patent/US20210174129A1

## CPC

- G06V 10/48: by mapping characteristic values of the pattern into a parameter space, e.g. Hough transformation
- G06F 18/22: Matching criteria, e.g. proximity measures
- G06T 7/00: Image analysis
- G06V 10/25: Determination of region of interest [ROI] or a volume of interest [VOI]
- G06V 10/40: Extraction of image or video features
- G06V 10/751: {Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching}
- G06V 10/757: {Matching configurations of points or features}
- G06V 10/759: {Region-based matching}
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

The information processing apparatus (2000) includes a feature point detection unit (2020), a determination unit (2040), an extraction unit (2060), and a comparison unit (2080). A feature point detection unit (2020) detects a plurality of feature points from the query image. The determination unit (2040) determines, for each feature point, one or more object images estimated to include the feature point. The extraction unit (2060) extracts an object region estimated to include the object in the query image in association with the object image of the object estimated to be included in the object region, on the basis of the result of the determination. The comparison unit (2080) cross-checks the object region with the object image associated with the object region and determines an object included in the object region.

## Claims

### Claim 1

An information processing apparatus comprising:
a feature point detection unit that acquires a query image including a plurality of objects and detecting a plurality of feature points from the acquired query image;
a determination unit that determines one or more object images estimated to include the feature point for each feature point;
an extraction unit that extracts an object region estimated to include an object in the query image in association with an object image of the object estimated to be included in the object region, based on a result of the determination; and
a comparison unit that compares the object region with the object image associated with the object region, thereby determining an object included in the object region.

### Claim 2 (depends on 1)

The information processing apparatus according to claim 1,
wherein when the local feature value of the feature point included in an object image is similar to the local feature value of the feature point in the query image, the determination unit determines the object image as the object image estimated to include the feature point in the query image.

### Claim 3 (depends on 2)

The information processing apparatus according to claim 2,
wherein the determination unit computes a degree of similarity between the local feature value of the feature point and the local feature value of the feature point in the query image, for some of the feature points included in the object image.

### Claim 4 (depends on 3)

The information processing apparatus according to claim 3,
wherein the determination unit computes, as the degree of similarity, a distance value between a part of a vector constituting the local feature value of the feature point included in the object image and a part of a vector constituting the local feature value of the feature point in the query image.

### Claim 5 (depends on 1, 2)

The information processing apparatus according to claim 1 or 2,
wherein an image identifier of each of the object images is stored in a storage device in association with an index value computed using the local feature value of each feature point included in the object image, and
wherein the determination unit computes an index value from the local feature value of the feature point in the query image, and determines an object image having the image identifier, which is stored in association with the index value, as the object image estimated to include the feature point.

### Claim 6 (depends on 1)

The information processing apparatus according to claim 1,
wherein the extraction unit performs: setting an image position corresponding to the feature point in the query image as a voting position of the feature point; associating the voting position with the object image estimated to include the feature point; and extracting, as the object region corresponding to the object, an image region in the query image including a predetermined number or more of the voting positions associated with the same object image.

### Claim 7 (depends on 1)

The information processing apparatus according to claim 1,
wherein the extraction unit performs: setting an image position corresponding to the feature point in the query image as a voting position; associating the voting position with the object image estimated to include the feature point by assigning a weight to the voting position; and extracting, as the object region corresponding to the object, an image region in the query image in which the sum of the weights assigned to the voting positions associated with the same object image is equal to or greater than a predetermined value.

### Claim 8 (depends on 6)

The information processing apparatus according to claim 6,
wherein the voting position of the feature point is an image position of the feature point.

### Claim 9 (depends on 6)

The information processing apparatus according to claim 6,
wherein, in the object image,
a reference vector starting from a position of a reference point is defined, and
a feature vector, whose initial point, direction, and magnitude are respectively a position of the feature point, an orientation of the local feature value of the feature point, and a scale of the local feature value of the feature point, is defined, and
wherein the extraction unit performs:
computing, for the feature point in the query image, a feature vector whose initial point, direction, and magnitude are respectively a position of the feature point, an orientation of the local feature value of the feature point, and a scale of the local feature value of the feature point, and
applying a conversion expression, which is for converting a feature vector of a feature point corresponding to the feature point in the object image estimated to include the feature point in the query image into the reference vector, to the feature vector of the feature point in the query image, thereby converting the feature vector and setting an image position at which an initial point of the vector after the conversion is the voting position of the feature point in the query image.

### Claim 10

A control method executed by a computer, the control method comprising:
acquiring a query image including a plurality of objects and detecting a plurality of feature points from the acquired query image;
determining one or more object images estimated to include the feature point for each feature point;
extracting an object region estimated to include an object in the query image in association with an object image of the object estimated to be included in the object region, based on a result of the determination; and
comparing the object region with the object image associated with the object region, thereby determining an object included in the object region.

### Claim at position 11

-18. (canceled)

### Claim 19 (depends on 10)

A non-transitory computer-readable storage medium storing a program causing a computer to execute each step of the control method according to claim 10.
