公开(公告)号：US09240048B2

公开(公告)日：2016-01-19

申请号：US13690514

申请日：2012-11-30

Applicant: Adobe Systems Incorporated

Inventor： Michael Wish Tao ,  Sunil Hadap

IPC: G06T7/00

CPC classification number: G06T7/0065 ,  G06T7/557

Abstract: Systems and methods are provided for depth map estimation using three-dimensional epipolar data structures. The image manipulation application receives image data depicting an image space from a multiple perspectives. The image manipulation application generates at least one three-dimensional epipolar data structure from the image data. The at least one three-dimensional epipolar data structure includes data describing the difference in position of at least one object between the perspectives. The at least one three-dimensional epipolar data structure corresponds to at least one region of the image space. The image manipulation application generates a depth map based on the at least one three-dimensional epipolar data structure.

Abstract translation: 提供了使用三维对极数据结构进行深度图估计的系统和方法。 图像处理应用从多个视角接收描绘图像空间的图像数据。 图像处理应用从图像数据生成至少一个三维对极数据结构。 所述至少一个三维对极数据结构包括描述在透视图之间的至少一个对象的位置差异的数据。 所述至少一个三维对极数据结构对应于图像空间的至少一个区域。 所述图像处理应用基于所述至少一个三维对极数据结构生成深度图。

公开(公告)号：US20140152647A1

公开(公告)日：2014-06-05

申请号：US13690514

申请日：2012-11-30

Applicant: ADOBE SYSTEMS INCORPORATED

Inventor： Michael Wish Tao ,  Sunil Hadap

IPC: G06T15/00

CPC classification number: G06T7/0065 ,  G06T7/557

Abstract: Systems and methods are provided for depth map estimation using three-dimensional epipolar data structures. The image manipulation application receives image data depicting an image space from a multiple perspectives. The image manipulation application generates at least one three-dimensional epipolar data structure from the image data. The at least one three-dimensional epipolar data structure includes data describing the difference in position of at least one object between the perspectives. The at least one three-dimensional epipolar data structure corresponds to at least one region of the image space. The image manipulation application generates a depth map based on the at least one three-dimensional epipolar data structure.

Abstract translation: 提供了使用三维对极数据结构进行深度图估计的系统和方法。 图像处理应用从多个视角接收描绘图像空间的图像数据。 图像处理应用从图像数据生成至少一个三维对极数据结构。 所述至少一个三维对极数据结构包括描述在透视图之间的至少一个对象的位置差异的数据。 所述至少一个三维对极数据结构对应于图像空间的至少一个区域。 所述图像处理应用基于所述至少一个三维对极数据结构生成深度图。

公开(公告)号：US09214025B2

公开(公告)日：2015-12-15

申请号：US13690460

申请日：2012-11-30

Applicant: Adobe Systems Incorporated

Inventor： Michael Wish Tao ,  Sunil Hadap

IPC: G06K9/00 ,  G06T7/00

CPC classification number: G06T7/0075 ,  G06T7/557 ,  G06T2207/10021 ,  G06T2207/10052

Abstract: Systems and methods are provided for depth map estimation using normalized displacement of image pairs. In one embodiment, an image manipulation application identifies image pairs from an input image. Each of the image pairs includes an image portion from a first perspective and the image portion from a second perspective. The image manipulation generates displacement vectors for the image pairs. Each of the displacement vectors represents at least one of a horizontal displacement and a vertical displacement of the image portion from the first perspective to the second perspective. The image manipulation application generates normalized displacement vectors corresponding to the displacement vectors. Each of the normalized displacement vectors is generated by transforming a respective orientation of a corresponding one of the displacement vectors to a common reference direction. The image manipulation application generates a depth map based on the normalized displacement vectors.

Abstract translation: 提供了使用图像对的归一化位移的深度图估计的系统和方法。 在一个实施例中，图像操作应用从输入图像识别图像对。 每个图像对包括来自第一透视图的图像部分和来自第二透视图的图像部分。 图像处理产生图像对的位移矢量。 每个位移矢量表示图像部分从第一透视图到第二透视图的水平位移和垂直位移中的至少一个。 图像处理应用产生对应于位移矢量的归一化位移矢量。 通过将相应的一个位移矢量的相应取向变换为公共参考方向来生成每个归一化位移矢量。 图像处理应用基于归一化的位移矢量生成深度图。

公开(公告)号：US20140153783A1

公开(公告)日：2014-06-05

申请号：US13690460

申请日：2012-11-30

Applicant: ADOBE SYSTEMS INCORPORATED

Inventor： Michael Wish Tao ,  Sunil Hadap

IPC: G06T7/00

CPC classification number: G06T7/0075 ,  G06T7/557 ,  G06T2207/10021 ,  G06T2207/10052

Abstract: Systems and methods are provided for depth map estimation using normalized displacement of image pairs. In one embodiment, an image manipulation application identifies image pairs from an input image. Each of the image pairs includes an image portion from a first perspective and the image portion from a second perspective. The image manipulation generates displacement vectors for the image pairs. Each of the displacement vectors represents at least one of a horizontal displacement and a vertical displacement of the image portion from the first perspective to the second perspective. The image manipulation application generates normalized displacement vectors corresponding to the displacement vectors. Each of the normalized displacement vectors is generated by transforming a respective orientation of a corresponding one of the displacement vectors to a common reference direction. The image manipulation application generates a depth map based on the normalized displacement vectors.

Abstract translation: 提供了使用图像对的归一化位移的深度图估计的系统和方法。 在一个实施例中，图像操作应用从输入图像识别图像对。 每个图像对包括来自第一透视图的图像部分和来自第二透视图的图像部分。 图像处理产生图像对的位移矢量。 每个位移矢量表示图像部分从第一透视图到第二透视图的水平位移和垂直位移中的至少一个。 图像处理应用产生对应于位移矢量的归一化位移矢量。 通过将相应的一个位移矢量的相应取向变换为公共参考方向来生成每个归一化位移矢量。 图像处理应用基于归一化的位移矢量生成深度图。