Still-image extracting method and image processing device for implementing the same

Patent metadata

Field	Value
Publication number	US-9843789-B2
Application number	US-201514803960-A
Country	US
Kind code	B2
Filing date	Jul 20, 2015
Priority date	Sep 8, 2014
Publication date	Dec 12, 2017
Grant date	Dec 12, 2017

Abstract

A still-image extracting method is disclosed. Frames of an object are extracted as still images from a moving image stream chronologically continuously captured by a camera. The camera moves relative to the object. First frames are extracted from the moving image stream. Image capture times of the extracted first frames are obtained. Image capture positions of the camera at the image capture times of the first frames are identified based on the first frames. Image capture times of the frames captured at image capture positions spaced at equal intervals are estimated based on both the image capture positions, identified by the first frames, of the camera and the obtained image capture times. Second frames at the estimated image capture times are extracted as frames captured and obtained at image capture positions spaced apart at equal intervals from the moving image stream.

First claim

What is claimed is:

1. A still-image extracting method for extracting a plurality of frames of an object as still images from a moving image stream chronologically continuously captured by a camera, the camera moving relative to the object, the method comprising:
extracting a plurality of first frames from the moving image stream and obtaining image capture times of each of the first frames extracted;
identifying image capture positions of the camera at the image capture times of each of the first frames based on the plurality of first frames;
estimating image capture times of the plurality of frames captured at a plurality of image capture positions spaced at equal intervals based on both the image capture positions, identified by each of the first frames, of the camera and the obtained image capture times;
extracting, from the moving image stream, second frames at the estimated image capture times as a plurality of frames captured at a plurality of image capture positions spaced apart at equal intervals,
identifying the image capture positions of the second frames extracted from the moving image stream based on the second frames;
determining whether or not a positional deviation between each of the image capture positions of the second frames and a corresponding one of the plurality of image capture positions spaced apart at equal intervals falls within a predetermined range, and when the positional deviation falls out of the predetermined range, re-estimating the image capture times of the frames captured by the camera at the positions spaced at equal intervals based on both the image capture positions and image capture times of the extracted second frames; and
extracting the second frames at the re-estimated image capture times as the frames captured at the plurality of image capture positions spaced apart at equal intervals from the moving image stream.

2. The still-image extracting method according to claim 1 , wherein estimating the image capture times includes estimating the image capture times of the plurality of frames captured at a plurality of image capture positions spaced apart at equal intervals by using linear interpolations based on the identified image capture positions of the plurality of first frames and the obtained image capture times.

3. The still-image extracting method according to claim 1 , wherein identifying the image capture positions of the plurality of first frame includes matching a plurality of feature points extracted from the plurality of first frames, calculating corresponding points of the frames, applying an Structure From Motion algorithm to the calculated corresponding points of the frames, and identifying the image capture positions.

4. The still-image extracting method according to claim 1 , wherein extracting the second frames at the estimated image capture times includes extracting the second frames as a plurality of frames captured by the camera at a plurality of image capture positions aligned at equal angular intervals from a moving image stream captured by the camera that moves on a circumference centering the object.

5. The still-image extracting method according to claim 1 , wherein extracting the second frames at the estimated image capture times includes extracting the plurality of second frames from a moving image stream captured by the camera that moves in a tangent direction of a circumference centering the object so as to generate a three-dimensional image with parallax.

6. The still-image extracting method according to claim 1 , wherein extracting the second frames at the estimated image capture times includes extracting the second frames as a plurality of frames captured by the camera at a plurality of image capture positions spaced apart at equal intervals from a moving image stream captured by the camera being moved along a straight line.

7. An image processing device for implementing a still-image extracting method for extracting a plurality of frames of an object as still images from a moving image stream chronologically continuously captured by a camera, the camera moving relative to the object, the device including:
an image extracting device; and
an image capture position calculation device,
wherein the image extracting device extracts a plurality of first frames from the moving image stream and obtains image capture times of the extracted plurality of first frames,
wherein the image capture position calculation device identifies information on image capture positions of the camera at the image capture times obtained by the plurality of first frames, and estimates image capture times of the plurality of frames captured at a plurality of image capture positions spaced at equal intervals based on both the information on the image capture positions, identified by the plurality of first frames, of the camera and the obtained image capture times,
wherein the image extracting device extracts, from the moving image stream, second frames corresponding to the image capture times estimated by the image capture position calculation device as a plurality of frames captured at a plurality of image capture positions spaced apart at equal intervals,
wherein the image capture position calculation device further identifies the image capture positions of the second frames extracted from the moving image stream based on the second frames, and determines whether or not a positional deviation between each of the image capture positions of the second frames and a corresponding one of the plurality of image capture positions spaced apart at equal intervals falls within a predetermined range, and when the positional deviation falls out of the predetermined range, re-estimates the image capture times of the frames captured by the camera at the positions spaced at equal intervals based on both the image capture positions and image capture times of the extracted second frames; and
wherein the image extracting device further extracts the second frames at the re-estimated image capture times as the frames captured at the plurality of image capture positions spaced apart at equal intervals from the moving image stream.

8. The image processing device of claim 7 , wherein estimating the image capture times includes estimating the image capture times of the plurality of frames captured at a plurality of image capture positions spaced apart at equal intervals by using linear interpolations based on the identified image capture positions of the plurality of first frames and the obtained image capture times.

9. The image processing device of claim 7 , wherein identifying the image capture positions of the plurality of first frame includes matching a plurality of feature points extracted from the plurality of first frames, calculating corresponding points of the frames, applying an Structure From Motion algorithm to the calculated corresponding points of the frames, and identifying the image capture positions.

10. The image processing device of claim 7 , wherein extracting the second frames at the estimated image capture times includes extracting the second frames as a plurality of frames captured by the camera at a plurality of image capture positions aligned at equal angular intervals from a moving image stream captured by the camera that moves on a circumference centering the object.

11. The image processing device of claim 7 , wherein extracting the second frames at the estimated image capture times includes extracting the plurality of second frames from a moving image stream captured by the camera that moves in a tangent direction of a circumference centering the object so as to generate a three-dimensional image with parallax.

12. The image proces

Assignees

• Panasonic Ip Man Co Ltd

Inventors

• Ogura Yasunobu

Classifications

• G06T7/596

  from three or more stereo images · CPC title

• G06T2207/10016

  Video; Image sequence · CPC title

• H04N13/0282 Primary

  Electricity · mapped topic

• H04N13/0221

  Electricity · mapped topic

• H04N13/221

  using the relative movement between cameras and objects · CPC title