公开(公告)号：US20150172547A1

公开(公告)日：2015-06-18

申请号：US14105554

申请日：2013-12-13

Applicant: Adobe Systems Incorporated

Inventor： Jue Wang ,  Sunghyun Cho ,  Zhe Hu

IPC: H04N5/232

CPC classification number: H04N5/23267 ,  G06T5/003

Abstract: A blurred image having a spatially invariant motion blur resulting from camera motion during image capture is deblurred based on one or more light streaks identified and extracted from the blurred image. A blur kernel for the blurred image is estimated by performing an optimization procedure having a blur kernel constraint based at least in part on the light streak. One or more light streaks can in some embodiments be posed as the blur kernel constraint. A modeled light streak may be defined as a convolution between the blur kernel and a simulated light source, with the optimization procedure being to minimize a distance between the modeled light streak and the corresponding identified light streak from the blurred image.

Abstract translation: 基于从模糊图像识别和提取的一条或多条光条纹，在图像拍摄期间具有由相机运动产生的空间不变运动模糊的模糊图像被去毛刺。 用于模糊图像的模糊内核通过至少部分地基于光条纹执行具有模糊内核约束的优化过程来估计。 在一些实施例中，可以提供一个或多个光条纹作为模糊核心约束。 模拟的光条可以被定义为模糊核与模拟光源之间的卷积，优化过程是将建模的条纹与从模糊图像相应的识别光条纹之间的距离最小化。

公开(公告)号：US20140140626A1

公开(公告)日：2014-05-22

申请号：US13680952

申请日：2012-11-19

Applicant: ADOBE SYSTEMS INCORPORATED

Inventor： Sunghyun Cho ,  Jue Wang ,  Lin Zhong

IPC: G06T5/00

CPC classification number: G06T5/003 ,  G06T7/13

Abstract: An image de-blurring system obtains a blurred input image and generates, based on the blurred input image, a blur kernel. The blur kernel is an indication of how the image capture device was moved and/or how the subject captured in the image moved during image capture. Based on the blur kernel and the blurred input image, a de-blurred image is generated. The blur kernel is generated based on the direction of edges identified in the blurred input image and/or based on curves having a high curvature identified in the image (e.g., corners identified in the image).

Abstract translation: 图像去模糊系统获得模糊的输入图像，并且基于模糊的输入图像生成模糊核。 模糊内核是图像捕获设备如何被移动的指示和/或在图像捕获期间如何在图像中捕获的图像移动。 基于模糊内核和模糊输入图像，生成去模糊图像。 基于在模糊输入图像中识别的边缘的方向和/或基于在图像中识别的具有高曲率的曲线（例如，图像中标识的角），生成模糊核。

公开(公告)号：US20130129156A1

公开(公告)日：2013-05-23

申请号：US13742213

申请日：2013-01-15

Applicant: Adobe Systems Incorporated

Inventor： Jue Wang ,  David P. Simons ,  Daniel M. Wilk ,  Xue Bai

IPC: G06T7/20

CPC classification number: G06T7/20 ,  G06K9/6277 ,  G06T7/12 ,  G06T7/143 ,  G06T7/149 ,  G06T2207/10024 ,  G06T2207/20121 ,  G06T2207/20124 ,  G06T2207/30196

Abstract: Systems, methods, and computer-readable storage media for chatter reduction in video object segmentation using a variable bandwidth search region. A variable bandwidth search region generation method may be applied to a uniform search region to generate a variable bandwidth search region that reduces the search range for segmentation methods such as a graph cut method. The method may identify parts of the contour that are moving slowly, and reduce the search region bandwidth in those places to stabilize the segmentation. This method may determine a bandwidth for each of a plurality of local windows of an image according to an estimate of how much an object in the image has moved from a previous image. The method may blend the bandwidths for the plurality of local windows to generate a blended map. The method may then generate a variable bandwidth search region for an object according to the blended map.

Abstract translation: 用于使用可变带宽搜索区域的视频对象分割中的抖动减少的系统，方法和计算机可读存储介质。 可变带宽搜索区域生成方法可以应用于均匀搜索区域以生成可变带宽搜索区域，其减少诸如图形切割方法的分割方法的搜索范围。 该方法可以识别正在缓慢移动的轮廓的部分，并且减少那些位置的搜索区域带宽以稳定分割。 该方法可以根据对图像中的对象已经从先前图像移动多少的估计来确定图像的多个局部窗口中的每一个的带宽。 该方法可以混合多个本地窗口的带宽以生成混合映射。 然后，该方法可以根据混合映射生成对象的可变带宽搜索区域。

公开(公告)号：US20190019052A1

公开(公告)日：2019-01-17

申请号：US15650669

申请日：2017-07-14

Applicant: Adobe Systems Incorporated

Inventor： I-Ming Pao ,  Jue Wang ,  Ke Ma ,  Zhe Lin

IPC: G06K9/32 ,  G06K9/00 ,  G06K9/36 ,  G06K9/46 ,  G06T11/60

CPC classification number: G06K9/00456 ,  G06K9/3258

Abstract: Text region detection techniques and systems for digital images using image tag filtering are described. These techniques and systems support numerous advantages over conventional techniques through use of image tags to filter text region candidates. A computing device, for instance, may first generate text region candidates through use of a variety of different techniques, such as text line detection. The computing device then assigns image tags to the text region candidates. The assigned image tags are then used by the computing device to filter the text region candidates based on whether image tags assigned to respective candidates are indicative of text.

公开(公告)号：US09978120B2

公开(公告)日：2018-05-22

申请号：US15268299

申请日：2016-09-16

Applicant: Adobe Systems Incorporated

Inventor： Xue Bai ,  Jue Wang

IPC: G06T3/00 ,  G06T11/60

CPC classification number: G06T3/0093 ,  G06T3/4038 ,  G06T11/60 ,  G06T2210/12

Abstract: The present disclosure is directed toward systems and method for warping a panoramic image to fit a predetermined shape using content-unaware warping techniques. For example, systems and methods described herein involve generating a mesh grid for a panoramic image with skewed edges by sampling boundary points around edges of the panoramic image and interpolated interior vertex points from the boundary points. Further, systems and methods described herein involve warping the mesh grid and underlying pixels of the panoramic image to fit a predetermined boundary. Further, systems and methods described herein involve generating the mesh grid and warping the panoramic image without consideration of content included therein and without overly-warping individual cells of the mesh grid and underlying pixels of the panoramic image.

公开(公告)号：US09466096B2

公开(公告)日：2016-10-11

申请号：US14937395

申请日：2015-11-10

Applicant: ADOBE SYSTEMS INCORPORATED

Inventor： Jue Wang ,  Sunghyun Cho ,  Jen-Chan Chien ,  Dong Feng

IPC: G06K9/40 ,  G06T5/00 ,  G06T5/20 ,  G06T7/00

CPC classification number: G06T5/003 ,  G06T5/20 ,  G06T7/11 ,  G06T2207/10004 ,  G06T2207/20012

Abstract: A computer-implemented method and system are described for deblurring an image. The method may include accessing an image having a first blurred region and a second blurred region, and generating a first blur kernel for the first blurred region and a second blur kernel for the second blurred region. Thereafter, the first blur kernel is positioned with respect to the first blurred region, and the second blur kernel is positioned with respect to the second blurred region based on the position of the first blur kernel. The image is then deblurred by deconvolving the first blurred region with the first blur kernel, and deconvolving the second blurred region with the second blur kernel.

公开(公告)号：US20160267690A1

公开(公告)日：2016-09-15

申请号：US14643630

申请日：2015-03-10

Applicant: Adobe Systems Incorporated

Inventor： Xue Bai ,  Jue Wang ,  Yao Lu

IPC: G06T11/20 ,  G06T11/00 ,  G06T7/00 ,  G06T7/60 ,  G06T15/00 ,  G06T15/10

CPC classification number: G06T11/203 ,  G06T7/12 ,  G06T7/149 ,  G06T7/246 ,  G06T7/60 ,  G06T11/001 ,  G06T15/005 ,  G06T15/10 ,  G06T2200/04 ,  G06T2207/10016 ,  G06T2207/10024 ,  G06T2207/20096 ,  G06T2207/20192 ,  G06T2210/12

Abstract: Techniques involving flexible video object boundary tracking are described. One or more curves, such as Bezier curves, are received as drawn by a user on an initial frame of video to define a boundary of an object in the frame. The curves are then mapped to a subsequent or previous frame of the video where the object is included but has a new or changed boundary. A segmentation boundary is determined for the object in the subsequent frame and endpoints of segments of the curves are snapped to the segmentation boundary. Additionally, confidence values are determined for subregions of the frame that include portions of the curves. These confidence values are used to update control points on the curve segments to fit the curve segments to the new or changed boundary of the object in the frame.

Abstract translation: 描述了涉及灵活视频对象边界跟踪的技术。 一个或多个曲线，例如Bezier曲线，由用户在视频的初始帧上绘制，以定义帧中的对象的边界。 然后将曲线映射到包含对象的视频的后续或前一帧，但具有新的或改变的边界。 为后续帧中的对象确定分割边界，并且将曲线的段的端点捕捉到分割边界。 另外，对包括曲线部分的帧的子区域确定置信度值。 这些置信度值用于更新曲线段上的控制点，以将曲线段拟合到框架中对象的新边界或更改的边界。

公开(公告)号：US20160267629A1

公开(公告)日：2016-09-15

申请号：US15161073

申请日：2016-05-20

Applicant: ADOBE SYSTEMS INCORPORATED

Inventor： Sunghyun Cho ,  Jue Wang ,  Jen-Chan Chien ,  Dong Feng

IPC: G06T5/00 ,  G06T11/00 ,  G06T5/20

CPC classification number: G06T5/003 ,  G06T5/20 ,  G06T11/001 ,  G06T2207/10024 ,  G06T2207/20192 ,  G06T2207/20201 ,  G06T2207/20221

Abstract: Embodiments of the present invention provide systems, methods, and computer storage media directed towards automatic selection of regions for blur kernel estimation. In one embodiment, a process divides a blurred image into a regions. From these regions a first region and a second region can be selected based on a number of edge orientations within the selected regions. A first blur kernel can then be estimated based on the first region and a second blur kernel can be estimated for the second region. The first and second blur kernel can then be utilized to respectively deblur a first and second portion of the image to produce a deblurred image. Other embodiments may be described and/or claimed.

Abstract translation: 本发明的实施例提供了针对模糊核估计的区域的自动选择的系统，方法和计算机存储介质。 在一个实施例中，过程将模糊图像划分为区域。 可以从这些区域中，基于所选择的区域内的多个边缘取向来选择第一区域和第二区域。 然后可以基于第一区域估计第一模糊核，并且可以为第二区域估计第二模糊核。 然后可以利用第一和第二模糊核来分别去除图像的第一和第二部分以产生去模糊图像。 可以描述和/或要求保护其他实施例。

公开(公告)号：US09305339B2

公开(公告)日：2016-04-05

申请号：US14320987

申请日：2014-07-01

Applicant: Adobe Systems Incorporated

Inventor： Jianchao Yang ,  Jue Wang ,  Ketan Tang

IPC: G06K9/40 ,  G06T5/00 ,  H04N5/232 ,  G06K9/46

CPC classification number: G06K9/4661 ,  G06K9/40 ,  G06T5/003 ,  G06T2207/10016 ,  G06T2207/10024 ,  G06T2207/20024 ,  H04N5/217 ,  H04N5/23229

Abstract: Multi-feature image haze removal is described. In one or more implementations, feature maps are extracted from a hazy image of a scene. The feature maps convey information about visual characteristics of the scene captured in the hazy image. Based on the feature maps, portions of light that are not scattered by the atmosphere and are captured to produce the hazy image are computed. Additionally, airlight of the hazy image is ascertained based on at least one of the feature maps. The calculated airlight represents constant light of the scene. Using the computed portions of light and the ascertained airlight, a dehazed image is generated from the hazy image.

Abstract translation: 描述了多特征图像雾度去除。 在一个或多个实现中，从场景的模糊图像中提取特征图。 特征图传达关于在朦胧图像中捕获的场景的视觉特征的信息。 基于特征图，计算不被大气散射并被捕获以产生模糊图像的部分光。 此外，基于特征图中的至少一个来确定模糊图像的飞行器。 计算出的空气动力表示场景的恒定光。 使用所计算的光部分和所确定的空气光，从模糊图像产生脱色图像。

公开(公告)号：US20150147047A1

公开(公告)日：2015-05-28

申请号：US14091658

申请日：2013-11-27

Applicant: Adobe Systems Incorporated

Inventor： Jue Wang ,  Sunghyun Cho ,  Shuaicheng Liu

IPC: H04N5/262 ,  G06T5/00 ,  G06T5/50 ,  G11B27/02

CPC classification number: H04N5/2621 ,  G06T5/003 ,  G06T5/50 ,  G06T2207/10016 ,  G06T2207/20201 ,  G11B27/02

Abstract: A simulated tracking shot is generated from an image sequence in which a foreground feature moves relative to a background during capturing of the image sequence. The background is artificially blurred in the simulated tracking shot in a spatially-invariant manner corresponding to foreground motion relative to the background during a time span of the image sequence. The foreground feature can be substantially unblurred relative to a reference image selected from the image sequence. A system to generate the simulated tracking shot can be configured to derive spatially invariant blur kernels for a background portion by reconstructing or estimating a 3-D space of the captured scene, placing virtual cameras along a foreground trajectory in the 3-D space, and projecting 3-D background points on to the virtual cameras.

Abstract translation: 在拍摄图像序列期间，前景特征相对于背景移动的图像序列生成模拟跟踪拍摄。 在图像序列的时间跨度期间，在相对于背景的前景运动相应于空间不变方式的模拟跟踪拍摄中，背景被人为地模糊。 相对于从图像序列中选择的参考图像，前景特征可以基本上未被翻转。 用于生成模拟跟踪镜头的系统可以被配置为通过重建或估计所捕获的场景的3-D空间，将虚拟相机沿着3D空间中的前景轨迹放置，从而为背景部分导出空间不变的模糊内核，以及 将3-D背景点投影到虚拟相机。