Information processing device and method, and program

Patent metadata

Field	Value
Publication number	US-10600202-B2
Application number	US-201615565845-A
Country	US
Kind code	B2
Filing date	Jun 9, 2016
Priority date	Jun 23, 2015
Publication date	Mar 24, 2020
Grant date	Mar 24, 2020

Abstract

A position/attitude calculation unit calculates a photographing position and a photographing attitude of a captured image to be processed relative to a known photographing position and photographing attitude of one captured image in a reference space on the basis of two captured images having five or more feature points that correspond to each other. In addition, the position/attitude calculation unit calculates a photographing position and a photographing attitude of the captured image to be processed in the reference space on the basis of a position of a subject in the reference space corresponding to a common feature point in the two captured images and another captured image of which a photographing position and a photographing attitude are known and the photographing position and the photographing attitude of the captured image to be processed relative to the photographing position and the photographing attitude of the one captured image. The present technology can be applied to smartphones.

First claim

The invention claimed is:

1. An information processing method comprising:
determining, with an electronic processor, whether a first photographing position and a first photographing attitude of a first captured image in a reference space can be inferred;
determining, with the electronic processor, whether a second photographing position and a second photographing attitude of a second captured image in the reference space can be inferred;
responsive to determining that the first photographing position and the first photographing attitude of the first captured image and the second photographing position and the second photographing attitude of the second captured image in the reference space can be inferred, inferring a relative positional relation between the second photographing position and a third photographing position of a third captured image on a basis of a common corresponding point included in each of the second captured image and the third captured image, the common corresponding point based on a physical feature in the each of the second captured image and the third captured image; and
responsive to determining that the first photographing position and the first photographing attitude of the first captured image and the second photographing position and the second photographing attitude of the second captured image in the reference space can be inferred, inferring the first photographing position, the first photographing attitude, the second photographing position, the second photographing attitude, the third photographing position, and a third photographing attitude of the third captured image in the reference space on a basis of the relative positional relation between the second photographing position and the third photographing position and a common corresponding point included in each of the first captured image and the third captured image, the common corresponding point based on a physical feature in the each of the first captured image and the third captured image.

2. The information processing method according to claim 1 , wherein the second captured image and the third captured image have at least five or more common corresponding points.

3. The information processing method according to claim 1 ,
wherein a relative positional relation between the first photographing position and the third photographing position is further inferred on a basis of the common corresponding point included in the first captured image and the third captured image respectively, and
the first photographing position, the first photographing attitude, the second photographing position, the second photographing attitude, the third photographing position, and the third photographing attitude in the reference space are inferred on a basis of the relative positional relation between the second photographing position and the third photographing position and the relative positional relation between the first photographing position and the third photographing position.

4. The information processing method according to claim 3 , wherein the first captured image and the third captured image have at least five or more common corresponding points.

5. The information processing method according to claim 3 , wherein the third photographing position in the reference space is inferred by calculating an intersection position between a first straight line which connects the second photographing position and the third photographing position and is determined on a basis of the relative positional relation between the second photographing position and the third photographing position, and a second straight line which connects the first photographing position and the third photographing position and is determined on a basis of the relative positional relation between the first photographing position and the third photographing position, or a position that is closest from the first straight line and the second straight line.

6. The information processing method according to claim 3 , wherein the third photographing attitude in the reference space is inferred on a basis of at least one of an inference result of the third photographing attitude based on the common corresponding point included in each of the second captured image and the third captured image and an inference result of the third photographing attitude based on the common corresponding point included in each of the first captured image and the third captured image.

7. The information processing method according to claim 1 , wherein the third photographing position in the reference space is inferred on a basis of the relative positional relation between the second photographing position and the third photographing position and a common corresponding point included in each of the first captured image, the second captured image, and the third captured image.

8. The information processing method according to claim 1 , wherein the first captured image, the second captured image, and the third captured image are images captured by different photographing units.

9. The information processing method according to claim 1 , wherein the first captured image, the second captured image, and the third captured image are images captured by a single photographing unit.

10. The information processing method according to claim 1 , further comprising:
performing predetermined image processing on a basis of the third photographing position and the third photographing attitude in the reference space and the third captured image.

11. The information processing method according to claim 1 , further comprising:
performing a reproduction process using the first captured image, the second captured image, and the third captured image on a basis of a positional relation between the first photographing position, the second photographing position, and the third photographing position.

12. An information processing device comprising:
a memory including a relative positional relation inferring program and a position/attitude inferring program; and
an electronic processor configured to
determine whether a first photographing position and a first photographing attitude of a first captured image in a reference space can be inferred, and
determine whether a second photographing position and a second photographing attitude of a second captured image in the reference space can be inferred,
infer, with the relative positional relation inferring program, a relative positional relation between the second photographing position and a third photographing position of a third captured image on a basis of a common corresponding point included in each of the second captured image and the third captured image in response to the electronic processor determining that the first photographing position and the second photographing position in the reference space can be inferred, the common corresponding point based on a physical feature in the each of the second captured image and the third captured image, and
infer, with the position/attitude inferring program, the first photographing position, the first photographing attitude, the second photographing position, the second photographing attitude, the third photographing position, and a third photographing attitude of the third captured image in the reference space on a basis of the relative positional relation between the second photographing position and the third photographing position and a common corresponding point included in each of the first captured image and the third captured image in response to the electronic processor determining that the first photographing position and the second photographing position in the reference space can be

Assignees

• Sony Corp

Inventors

• Ikeda Hiroshi
• Inaba Seijiro

Classifications

• G01B11/002

  for measuring two or more coordinates · CPC title

• G06T7/73

  using feature-based methods · CPC title

• G06T2207/30244

  Camera pose · CPC title

• G06T7/74 Primary

  involving reference images or patches · CPC title

• G01B11/26 Primary

  for measuring angles or tapers; for testing the alignment of axes · CPC title