System and method of image reconstruction
First Claim
1. A method of image reconstruction, comprising the steps of:
- capturing a stereoscopic pair of images of a scene;
generating, by one or more processors, a disparity map for the pair of images;
defining a series of value ranges corresponding to disparity, each value range in the series having an end point corresponding to a greater disparity than an end point of preceding value ranges in the series;
for each value range in the series, and progressing in order through the series from the first value range in the series,selecting points in the disparity map falling within the respective value range; and
generating, by the one or more processors, a mesh responsive to those selected points,and subsequently merging the generated meshes to form a 3D model of the scene.
3 Assignments
0 Petitions
Accused Products
Abstract
A method of image reconstruction is provided. The method includes capturing a stereoscopic pair of images of a scene, generating a disparity map for the pair of images, and defining a series of value ranges corresponding to disparity with each value range in the series having an end point corresponding to a greater disparity than an end point of preceding value ranges in the series. For each value range in the series, and progressing in order through the series from the first value range in the series, the method includes selecting points in the disparity map falling within the respective value range and generating a mesh responsive to those selected points. The method also includes subsequently merging the series of generated meshes to form a 3D model of the scene.
24 Citations
15 Claims
-
1. A method of image reconstruction, comprising the steps of:
-
capturing a stereoscopic pair of images of a scene; generating, by one or more processors, a disparity map for the pair of images; defining a series of value ranges corresponding to disparity, each value range in the series having an end point corresponding to a greater disparity than an end point of preceding value ranges in the series; for each value range in the series, and progressing in order through the series from the first value range in the series, selecting points in the disparity map falling within the respective value range; and generating, by the one or more processors, a mesh responsive to those selected points, and subsequently merging the generated meshes to form a 3D model of the scene. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11)
-
-
12. A non-transitory computer program product comprising computer readable instructions that when implemented by a computer cause it to perform a method comprising the steps of:
-
capturing a stereoscopic pair of images of a scene; generating a disparity map for the pair of images; defining a series of value ranges corresponding to disparity, each value range in the series having an end point corresponding to a greater disparity than an end point of preceding value ranges in the series; for each value range in the series, and progressing in order through the series from the first value range in the series, selecting points in the disparity map falling within the respective value range; and generating a mesh responsive to those selected points, and subsequently merging the generated meshes to form a 3D model of the scene.
-
-
13. An entertainment device for image reconstruction, comprising:
-
an input operable to receive a stereoscopic pair of captured images of a scene; a processor operable to generate a disparity map for the pair of images and to define a series of value ranges corresponding to disparity, each value range in the series having an end point corresponding to a greater disparity than an end point of preceding value ranges in the series; the processor being operable, for each value range in the series, and progressing in order through the series from the first value range in the series, to select points in the disparity map falling within the respective value range; and generate a mesh responsive to those selected points, the processor being operable to subsequently merge the generated meshes to form a 3D model of the scene. - View Dependent Claims (14, 15)
-
Specification