专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

US20170213112A1 UTILIZING DEEP LEARNING FOR AUTOMATIC DIGITAL IMAGE SEGMENTATION AND STYLIZATION 有权
公开(公告)号：US20170213112A1
公开(公告)日：2017-07-27
申请号：US15005855
申请日：2016-01-25
申请人： Adobe Systems Incorporated
发明人： Ian Sachs , Xiaoyong Shen , Sylvain Paris , Aaron Hertzmann , Elya Shechtman , Brian Price
IPC分类号： G06K9/66 , G06K9/46 , G06T7/00
CPC分类号： G06K9/66 , G06K9/00228 , G06K9/4604 , G06T7/11 , G06T7/73 , G06T7/90 , G06T2207/20084 , G06T2207/30201
摘要： Systems and methods are disclosed for segregating target individuals represented in a probe digital image from background pixels in the probe digital image. In particular, in one or more embodiments, the disclosed systems and methods train a neural network based on two or more of training position channels, training shape input channels, training color channels, or training object data. Moreover, in one or more embodiments, the disclosed systems and methods utilize the trained neural network to select a target individual in a probe digital image. Specifically, in one or more embodiments, the disclosed systems and methods generate position channels, training shape input channels, and color channels corresponding the probe digital image, and utilize the generated channels in conjunction with the trained neural network to select the target individual.

2. 发明授权

US09036909B2 Pyramid collapse color interpolation 有权
标题翻译：金字塔倒塌颜色插值
公开(公告)号：US09036909B2
公开(公告)日：2015-05-19
申请号：US13815053
申请日：2013-01-28
申请人： Adobe Systems Incorporated
发明人： Sylvain Paris , James McCann
IPC分类号： G06K9/00 , G06K9/32 , G06T3/40
CPC分类号： G06T3/4092 , G06T3/4038 , G06T3/4084
摘要： One exemplary embodiment involves receiving an image with a set of undefined pixels and a set of defined pixels and recursively modifying the image to generate a seamless composition comprising only defined pixels. Disclosed are embodiments for recursively modifying the image by recursively down sampling the image by a factor to generate a plurality of down sampled images until the down sampled image generated at each recursive down sampling lacks undefined pixels and then recursively up sampling each one of the down sampled images by the factor to generate an up sampled image from the respective down sampled image. Additionally, at each recursive up sampling instance, pasting the next recursively occurring down sampled image on the up sampled image to generate the next recursively occurring image for up sampling.
摘要翻译：一个示例性实施例涉及用一组未定义像素和一组定义的像素接收图像，并递归地修改图像以生成仅包含限定像素的无缝组合。公开了递归地修改图像的实施例，通过递归地对图像进行下采样，以产生多个下采样图像，直到在每个递归下采样期间产生的下采样图像缺少未定义的像素，然后递归上采样每个下采样通过该因子的图像从相应的下采样图像生成上采样图像。此外，在每个递归采样实例中，将下一个递归递减的采样图像粘贴在上采样图像上，以产生用于上采样的下一个递归出现图像。

3. 发明申请

US20170309053A1 SEAMLESS COMPOSITING USING A SOFT SELECTION 有权
公开(公告)号：US20170309053A1
公开(公告)日：2017-10-26
申请号：US15646509
申请日：2017-07-11
申请人： ADOBE SYSTEMS INCORPORATED
发明人： Sylvain Paris , Sohrab Amirghodsi , Elya Shechtman
IPC分类号： G06T11/60 , G06T7/12 , G06K9/46 , G06K9/62
CPC分类号： G06T11/60 , G06K9/4604 , G06K9/6215 , G06T5/005 , G06T7/12 , G06T7/136 , G06T2207/10144 , G06T2207/20112
摘要： A healing component that heals foreground pixels with background pixels is provided. In some embodiments, the healing component is programmed or otherwise configured to respond to a single healing request by identifying a plurality of regions within a selected area and healing each region of the plurality of regions independently of other regions.

4. 发明授权

US09576351B1 Style transfer for headshot portraits 有权
标题翻译：风格转移为头像肖像
公开(公告)号：US09576351B1
公开(公告)日：2017-02-21
申请号：US14946004
申请日：2015-11-19
申请人： Adobe Systems Incorporated
发明人： Ronen Barzel , Sylvain Paris , Robert Bailey , Ian Sachs
IPC分类号： G06K9/54 , G06T5/50 , G06T5/40 , G06T3/00 , G06T11/60
CPC分类号： G06T11/60
摘要： Techniques are disclosed for automatically transferring a style of at least two reference images to an input image. The resulting transformation of the input image matches the visual styles of the reference images without changing the identity of the subject of the input image. Each image is decomposed into levels of detail with corresponding energy levels and a residual. A style transfer operation is performed at each energy level and residual using the reference image that most closely matches the input image at each energy level. The transformations of each level of detail and the residual of the input image are aggregated to generate an output image having the styles of the reference images. In some cases, the transformations are performed on the foreground of the input image, and the background can be transformed by an amount that is proportional to the aggregated transformations of the foreground.
摘要翻译：公开了用于将至少两个参考图像的风格自动传送到输入图像的技术。输入图像的所得到的变换与参考图像的视觉样式匹配，而不改变输入图像的对象的身份。每个图像被分解成具有相应能量级别和残差的细节级别。在每个能量级别进行风格转移操作，并使用在每个能级与输入图像最匹配的参考图像进行残差。将每个细节级别的变换和输入图像的残差进行聚合以产生具有参考图像样式的输出图像。在某些情况下，在输入图像的前景上执行变换，并且可以将背景变换成与前景的聚合变换成比例的量。

5. 发明申请

US20150098646A1 LEARNING USER PREFERENCES FOR PHOTO ADJUSTMENTS 有权
标题翻译：学习照片调整的用户首选项
公开(公告)号：US20150098646A1
公开(公告)日：2015-04-09
申请号：US14047735
申请日：2013-10-07
申请人： Adobe Systems Incorporated
发明人： Sylvain Paris , Durga Ganesh Grandhi , Srinivasa Madhava Phaneendra Angara , Robert Land Gager
IPC分类号： G06K9/66 , G06T5/00 , G06K9/62
CPC分类号： G06K9/6254 , G06K9/6202 , G06K9/6215 , G06T5/00 , G06T5/50 , G06T2200/24 , G06T2200/28 , G06T2207/20081
摘要： In example embodiments, systems and methods for learning and using user preferences for image adjustments are presented. In example embodiments, a new image is received. A correction parameter based on previously stored user adjustments for similar images is determined. A user style that is an adjusted version of the new image is generated by applying the correction parameter. The user style is provided on a user interface. A user adjustment is received. Based on determining that a user sample image is within a predetermined threshold of closeness to the new image, data corresponding to the user sample image is replaced with new adjustment data for the new image in a database of user sample images used to generate the correction parameter. Based on determining that no user sample images are within the predetermined threshold of closeness, new adjustment data is appended to the database used to generate the correction parameter.
摘要翻译：在示例实施例中，呈现用于学习和使用用于图像调整的用户偏好的系统和方法。在示例实施例中，接收新的图像。确定基于用于类似图像的先前存储的用户调整的校正参数。通过应用校正参数生成作为新图像的调整版本的用户风格。在用户界面上提供用户风格。接收用户调整。基于确定用户采样图像在与新图像的接近度的预定阈值内，与用户样本图像相对应的数据被替换为用于生成修正参数的用户样本图像的数据库中的新图像的新调整数据。基于确定没有用户样本图像在预定的接近阈值内，新的调整数据被附加到用于生成校正参数的数据库。

6. 发明授权

US09892537B2 Seamless compositing using a soft selection 有权
公开(公告)号：US09892537B2
公开(公告)日：2018-02-13
申请号：US15646509
申请日：2017-07-11
申请人： ADOBE SYSTEMS INCORPORATED
发明人： Sylvain Paris , Sohrab Amirghodsi , Elya Shechtman
IPC分类号： G06K9/34 , G06T11/60 , G06T7/12 , G06K9/62 , G06K9/46
CPC分类号： G06T11/60 , G06K9/4604 , G06K9/6215 , G06T5/005 , G06T7/12 , G06T7/136 , G06T2207/10144 , G06T2207/20112
摘要： A healing component that heals foreground pixels with background pixels is provided. In some embodiments, the healing component is programmed or otherwise configured to respond to a single healing request by identifying a plurality of regions within a selected area and healing each region of the plurality of regions independently of other regions.

7. 发明申请

US20170344860A1 UTILIZING DEEP LEARNING FOR AUTOMATIC DIGITAL IMAGE SEGMENTATION AND STYLIZATION 有权
公开(公告)号：US20170344860A1
公开(公告)日：2017-11-30
申请号：US15679989
申请日：2017-08-17
申请人： Adobe Systems Incorporated
发明人： Ian Sachs , Xiaoyong Shen , Sylvain Paris , Aaron Hertzmann , Elya Shechtman , Brian Price
IPC分类号： G06K9/66 , G06K9/46 , G06T7/90
CPC分类号： G06K9/66 , G06K9/00228 , G06K9/4604 , G06T7/11 , G06T7/73 , G06T7/90 , G06T2207/20084 , G06T2207/30201
摘要： Systems and methods are disclosed for segregating target individuals represented in a probe digital image from background pixels in the probe digital image. In particular, in one or more embodiments, the disclosed systems and methods train a neural network based on two or more of training position channels, training shape input channels, training color channels, or training object data. Moreover, in one or more embodiments, the disclosed systems and methods utilize the trained neural network to select a target individual in a probe digital image. Specifically, in one or more embodiments, the disclosed systems and methods generate position channels, training shape input channels, and color channels corresponding the probe digital image, and utilize the generated channels in conjunction with the trained neural network to select the target individual.

8. 发明授权

US09773196B2 Utilizing deep learning for automatic digital image segmentation and stylization 有权
公开(公告)号：US09773196B2
公开(公告)日：2017-09-26
申请号：US15005855
申请日：2016-01-25
申请人： Adobe Systems Incorporated
发明人： Ian Sachs , Xiaoyong Shen , Sylvain Paris , Aaron Hertzmann , Elya Shechtman , Brian Price
IPC分类号： G06K9/62 , G06K9/66 , G06T7/00 , G06K9/46
CPC分类号： G06K9/66 , G06K9/00228 , G06K9/4604 , G06T7/11 , G06T7/73 , G06T7/90 , G06T2207/20084 , G06T2207/30201
摘要： Systems and methods are disclosed for segregating target individuals represented in a probe digital image from background pixels in the probe digital image. In particular, in one or more embodiments, the disclosed systems and methods train a neural network based on two or more of training position channels, training shape input channels, training color channels, or training object data. Moreover, in one or more embodiments, the disclosed systems and methods utilize the trained neural network to select a target individual in a probe digital image. Specifically, in one or more embodiments, the disclosed systems and methods generate position channels, training shape input channels, and color channels corresponding the probe digital image, and utilize the generated channels in conjunction with the trained neural network to select the target individual.

9. 发明申请

US20160253578A1 LEARNING USER PREFERENCES FOR PHOTO ADJUSTMENTS 审中-公开
公开(公告)号：US20160253578A1
公开(公告)日：2016-09-01
申请号：US15154796
申请日：2016-05-13
申请人： ADOBE SYSTEMS INCORPORATED
发明人： Sylvain Paris , Durga Ganesh Grandhi , Srinivasa Madhava Phaneendra Angara , Robert Land Gager
IPC分类号： G06K9/62 , G06T5/50
CPC分类号： G06K9/6254 , G06K9/6202 , G06K9/6215 , G06T5/00 , G06T5/50 , G06T2200/24 , G06T2200/28 , G06T2207/20081
摘要： In example embodiments, systems and methods for learning and using user preferences for image adjustments are presented. In example embodiments, a new image is received. A correction parameter based on previously stored user adjustments for similar images is determined. A user style that is an adjusted version of the new image is generated by applying the correction parameter. The user style is provided on a user interface. A user adjustment is received. Based on determining that a user sample image is within a predetermined threshold of closeness to the new image, data corresponding to the user sample image is replaced with new adjustment data for the new image in a database of user sample images used to generate the correction parameter. Based on determining that no user sample images are within the predetermined threshold of closeness, new adjustment data is appended to the database used to generate the correction parameter.

10. 发明授权

US09361666B2 Learning user preferences for photo adjustments 有权
标题翻译：学习照片调整的用户偏好
公开(公告)号：US09361666B2
公开(公告)日：2016-06-07
申请号：US14047735
申请日：2013-10-07
申请人： Adobe Systems Incorporated
发明人： Sylvain Paris , Durga Ganesh Grandhi , Srinivasa Madhava Phaneendra Angara , Robert Land Gager
IPC分类号： G06K9/62 , G06T5/00
CPC分类号： G06K9/6254 , G06K9/6202 , G06K9/6215 , G06T5/00 , G06T5/50 , G06T2200/24 , G06T2200/28 , G06T2207/20081
摘要： In example embodiments, systems and methods for learning and using user preferences for image adjustments are presented. In example embodiments, a new image is received. A correction parameter based on previously stored user adjustments for similar images is determined. A user style that is an adjusted version of the new image is generated by applying the correction parameter. The user style is provided on a user interface. A user adjustment is received. Based on determining that a user sample image is within a predetermined threshold of closeness to the new image, data corresponding to the user sample image is replaced with new adjustment data for the new image in a database of user sample images used to generate the correction parameter. Based on determining that no user sample images are within the predetermined threshold of closeness, new adjustment data is appended to the database used to generate the correction parameter.
摘要翻译：在示例实施例中，呈现用于学习和使用用于图像调整的用户偏好的系统和方法。在示例实施例中，接收新的图像。确定基于用于类似图像的先前存储的用户调整的校正参数。通过应用校正参数生成作为新图像的调整版本的用户风格。在用户界面上提供用户风格。接收用户调整。基于确定用户采样图像在与新图像的接近度的预定阈值内，与用户样本图像相对应的数据被替换为用于生成修正参数的用户样本图像的数据库中的新图像的新调整数据。基于确定没有用户样本图像在预定的接近阈值内，新的调整数据被附加到用于生成校正参数的数据库。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式