Matching un-synchronized image portions
AssigneeMicrosoft Corporation
InventorsAntonio Criminisi, Gavin Smyth, Geoffrey Cross
Life of this patent
Estimating the term…
Abstract
A matching system may be used to match unsynchronized camera signals. In one example, the images from each camera may be matched by synchronizing images from each camera such that the scene of each image is determined to be static. Alternatively or additionally, the images from one camera may be compared to images from another camera to determine the best synchronous match. These matched, synchronized image portions may be processed to generate a transformation structure that may be used to rectify the images, e.g., transform images from the cameras as if the cameras had co-planar image planes, or used in any other process, such as calibration of the cameras.
Claims41
A method comprising:
- a) receiving a plurality of images of a scene from a first camera position;
- b) at least one image of the scene from a second camera position, the at least one image from a second camera position being un-synchronized in time relative to the plurality of images from the first camera position;
- c) detecting movement of at least a portion of the scene in at least a portion of a first image of the plurality of images from the first camera position, an image portion without detected movement being a static image portion of the first camera position;
- d) synchronizing the static image portion of the first camera position with at least a portion of the at least one image from the second camera position.
A method comprising:
- a) instructing a user to set up a first static scene in a viewing area of a first camera and a second camera, the first camera being un-synchronized in time relative to the second camera;
- b) capturing at least one first image of the first static scene with each of the first camera and the second camera;
- c) associating the at least one first image from the first camera and the second camera to form a first synchronized group of images;
- d) instructing the user to set up a second static scene in the viewing area of the first camera and the second camera;
- e) capturing at least one second image of the second static scene with each of the first camera and the second camera;
- and
- f) associating the at least one second image from the first camera and the second camera to form a second synchronized group of images.
One or more computer readable media containing executable components comprising:
- a) a first digital camera providing a first modulated data signal representing a first plurality of images;
- b) a second digital camera providing a second modulated data signal representing a second plurality of images;
- c) a motion detection engine for receiving the first modulate data signal and the second modulated data signal, and for determining which images from the first camera and the second camera represent a static first scene;
- d) a synchrony engine for receiving the static images from the first camera and the second camera and for synchronizing at least one static image from the first camera with at least one static image from the second camera;
- e) a feature detector engine for detecting at least one feature of the at least one static image from the first camera, and for detecting at least one feature of the at least one static image from the second camera;
- and
- f) a feature point matcher engine for matching the at least one feature of the at least one static image from the first camera with the at least one feature of the at least one static image from the second camera.
The one or more computer readable media of claim 31, wherein the movement engine associates a weight with each image portion indicating movement of at least a portion of the scene.
One or more computer readable media containing executable components comprising:
- a) first means for generating a first digital image and a third digital image;
- b) second means for generating a second digital image and a fourth digital image;
- c) means for detecting movement of the scene based on the first image;
- d) means for synchronizing the first image and the second image;
- e) means for generating a transformation structure;
- and
- f) means for transforming the third and fourth digital images as if an image plane of the first means for generating and an image plane of the second means for generating were co-planar.
One or more computer readable media containing computer executable instructions that, when implemented, perform a method comprising:
- a) receiving a first set of images over time from a first camera;
- b) receiving a second set of images over time from a second camera;
- c) comparing a first image from the first set of images with at least a portion of the second set of images;
- d) based on the comparison, determining a best match image from the at least a portion of the second set of images;
- e) detecting at least one feature in the first image;
- f) detecting at least one feature in the best match image;
- g) matching the at least one feature in the first image with the at least one feature in the best match image;
- h) generating a transformation structure based on the matching features of the first image and the best match image, the transformation structure for transforming images from the first camera and the second camera t images as if the first and second cameras had co-planar image planes.
One or more computer readable media containing computer executable components comprising:
- a) first means for generating a first digital image and a third digital image;
- b) second means for generating a second digital image, a fourth digital image, and a fifth digital image;
- c) means for comparing the first digital image with the second image and the fifth image, wherein the second and fifth images each have a time stamp within a predetermined threshold of a time stamp of the first image.
- d) means for generating a transformation structure;
- and
- e) means for transforming the third and fourth digital images as if an image plane of the first means for generating and an image plane of the second means for generating were co-planar.
Dependencies are inferred from the claim text.