SYSTEMS AND METHODS FOR HIGHER ORDER DIMENSIONAL SPACE VIDEO SEGMENTATION

Number: 20170345151
Published: 2017-11-30
Filed: 2016-05-26
Assignee: DISNEY ENTERPRISES, INC.
Inventors: Hornung; Alexander Sorkine, Perazzi; Federico, Wang; Oliver, Marki; Nicolas
CPC: G11B27/28; G11B27/3081; G06T7/10; G11B27/34; G06V20/49
Verdict: Low Notable software
Source: Google Patents · FreePatentsOnline

The keeper's note

Segments video by representing pixels in a higher-order dimensional space combining color and position for more accurate object boundaries.

Abstract

This disclosure relates to system and methods for segmenting a video in a higher order dimensional space. A video may be segmented by obtaining visual informationdefining an image of the video. The visual information may include pixels of the image and may be represented in a display space having a first dimensionality. A designation of a subset of the visual information represented in the display space as a part of an object portrayed in the image may be obtained. The visual information and the designation may be represented in the higher order dimensional space having a second dimensionality greater than the first dimensionality. An association of the visual information represented in the higher order dimensional space with the object may be obtained. The association may be correlated with the visual information represented in the display space. The correlation may define a location of the object in the image.

Background

BRIEF DESCRIPTION OF THEDRAWINGS

FIG. 1 illustrates a system for segmenting a video in a higher order dimensional space.

FIG. 2 illustrates a method for segmenting a video in a higher order dimensional space.

FIG. 3 illustrates an example of a one-dimensional image.

FIG. 4 illustrates an example of a one-dimensional image lifted into a two-dimensional feature space.

FIG. 5 illustrates an example of accumulating values on individual vertices of a two-dimensional feature space.

FIG. 6 illustrates an example of assigning labels to individual vertices of a two-dimensional feature space.

FIG. 7 illustrates anexample of interpolating labels of individual vertices at positions of pixels corresponding to the individual vertices.

FIG. 8 illustrates an example of a segmented one-dimensional image.

FIG. 9A illustrates an example of a nearest neighbor interpolation scheme.

FIG. 9B illustrates an example of a multi-linear interpolation scheme.

FIG. 9C illustrates an example of an adjacent interpolation scheme.

FIGS. 10A-10C illustrate examples of a user provided mask propagated through images of a video.

FIGS.11A-11C illustrate examples of a video segmentation based on user strokes.DETAILED DESCRIPTION

FIG. 1 illustrates a system 10 for segmenting a video in a higher order dimensional space. System 10 may include one or more of processor 11, electronic storage 12, bus13, and/or other components. To seg

Claims

1. A system for segmenting a videoin a higher order dimensional space, the system comprising: one or more processors configured by machine readable instructions to: obtain visual information defining an image of the video, wherein the visual information includes pixels of the image and is represented in a display space having a first dimensionality such that the visual information represented in the display space facilitates visual reproduction of the image on a display device; obtain a designation of a subset of the visual information represented in the display space as a part an object portrayed in the image; represent the visual information and the designation in the higher order dimensional space, the higher order dimensional space having a second dimensionality greater than the first dimensionality; determine an association ofthe visual information represented in the higher order dimensional space with the object based on the higher order dimensional space representation of the visual information and the designation; and correlate the association with the visual information represented in the display space, the correlation defining a location of the object in the image. 11. A method for segmenting a video in a higher order dimensional space, the method comprising: obtaining visual information defining an image of the video, wherein the visual information includes pixels of the image and is represented in a display space having a first dimensionality such that the visual information represented in the display space facilitates visual reproduction of the image on a display device; obtaining a designation of a subset of the visual information represented in the display space as a part an object portrayed in theimage; representing the visual information and the designation in the higher order dimensional space, the higher order dimensional space having a second dimensionality greater than the first dimensionality; determining an association of the visual information represented in the higher order dimensional space with the object based on the higher order dimensional space representation of the visual information and the designation; and correlating the association with the visual information represented in the display space, the correlation defining a location of the object in the image.