公开(公告)号：US20190034056A1

公开(公告)日：2019-01-31

申请号：US15660284

申请日：2017-07-26

Applicant: Adobe Systems Incorporated

Inventor： Jonathan Eisenmann ,  Bushra Mahmood

IPC: G06F3/0481 ,  G06T19/00 ,  G06T15/20

CPC classification number: G06F3/04815 ,  G06T15/20 ,  G06T19/003 ,  G06T19/006 ,  G06T19/20 ,  G06T2200/24 ,  G06T2219/2004

Abstract: The present disclosure is directed toward systems and methods for manipulating a camera perspective within a digital environment for rendering three-dimensional objects against a background digital image. In particular, the systems and methods described herein display a view of a three-dimensional space including a horizon, a ground plane, and a three-dimensional object in accordance with a camera perspective of the three-dimensional space. The systems and methods further manipulate the camera perspective in response to, and in accordance with, user interaction with one or more options. The systems and methods manipulate the camera perspective relative to the three-dimensional space and thereby change the view of the three-dimensional space within a user interface.

公开(公告)号：US20190139319A1

公开(公告)日：2019-05-09

申请号：US15804908

申请日：2017-11-06

Applicant: ADOBE SYSTEMS INCORPORATED

Inventor： Jonathan Eisenmann ,  Geoffrey Alan Oxholm ,  Elya Shechtman ,  Bryan Russell

IPC: G06T19/20 ,  G06T15/20

CPC classification number: G06T19/20 ,  G06T15/20 ,  G06T2219/2004 ,  G06T2219/2016

Abstract: Embodiments disclosed herein provide systems, methods, and computer storage media for automatically aligning a 3D camera with a 2D background image. An automated image analysis can be performed on the 2D background image, and a classifier can predict whether the automated image analysis is accurate within a selected confidence level. As such, a feature can be enabled that allows a user to automatically align the 3D camera with the 2D background image. For example, where the automated analysis detects a horizon and one or more vanishing points from the background image, the 3D camera can be automatically transformed to align with the detected horizon and to point at a detected horizon-located vanishing point. In some embodiments, 3D objects in a 3D scene can be pivoted and the 3D camera dollied forward or backwards to reduce changes to the framing of the 3D composition resulting from the 3D camera transformation.

公开(公告)号：US20180286061A1

公开(公告)日：2018-10-04

申请号：US15996833

申请日：2018-06-04

Applicant: Adobe Systems Incorporated

Inventor： Xiaohui Shen ,  Scott Cohen ,  Peng Wang ,  Bryan Russell ,  Brian Price ,  Jonathan Eisenmann

IPC: G06T7/50

CPC classification number: G06T7/50 ,  G06T7/13 ,  G06T7/62 ,  G06T2207/20084

Abstract: Techniques for planar region-guided estimates of 3D geometry of objects depicted in a single 2D image. The techniques estimate regions of an image that are part of planar regions (i.e., flat surfaces) and use those planar region estimates to estimate the 3D geometry of the objects in the image. The planar regions and resulting 3D geometry are estimated using only a single 2D image of the objects. Training data from images of other objects is used to train a CNN with a model that is then used to make planar region estimates using a single 2D image. The planar region estimates, in one example, are based on estimates of planarity (surface plane information) and estimates of edges (depth discontinuities and edges between surface planes) that are estimated using models trained using images of other scenes.

公开(公告)号：US09990728B2

公开(公告)日：2018-06-05

申请号：US15261749

申请日：2016-09-09

Applicant: ADOBE SYSTEMS INCORPORATED

Inventor： Xiaohui Shen ,  Scott Cohen ,  Peng Wang ,  Bryan Russell ,  Brian Price ,  Jonathan Eisenmann

IPC: G06T7/00

CPC classification number: G06T7/50 ,  G06T7/0051 ,  G06T2207/20084

Abstract: Techniques for planar region-guided estimates of 3D geometry of objects depicted in a single 2D image. The techniques estimate regions of an image that are part of planar regions (i.e., flat surfaces) and use those planar region estimates to estimate the 3D geometry of the objects in the image. The planar regions and resulting 3D geometry are estimated using only a single 2D image of the objects. Training data from images of other objects is used to train a CNN with a model that is then used to make planar region estimates using a single 2D image. The planar region estimates, in one example, are based on estimates of planarity (surface plane information) and estimates of edges (depth discontinuities and edges between surface planes) that are estimated using models trained using images of other scenes.