中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

73 records (took 0.019 seconds)

    Font recognition and font similarity learning using a deep neural network
    21.
    发明授权
    Font recognition and font similarity learning using a deep neural network 有权
    使用深层神经网络的字体识别和字体相似性学习

    公开(公告)号：US09501724B1

    公开(公告)日：2016-11-22

    申请号：US14734466

    申请日：2015-06-09

    Applicant: ADOBE SYSTEMS INCORPORATED

    Inventor： Jianchao Yang ,  Zhangyang Wang ,  Jonathan Brandt ,  Hailin Jin ,  Elya Shechtman ,  Aseem Omprakash Agarwala

    IPC: G06K9/00 ,  G06K9/36 ,  G06K9/66 ,  G06K9/62 ,  G06K9/68 ,  G06T3/40

    CPC classification number: G06T3/40 ,  G06K9/6255 ,  G06K9/6828

    Abstract: A convolutional neural network (CNN) is trained for font recognition and font similarity learning. In a training phase, text images with font labels are synthesized by introducing variances to minimize the gap between the training images and real-world text images. Training images are generated and input into the CNN. The output is fed into an N-way softmax function dependent on the number of fonts the CNN is being trained on, producing a distribution of classified text images over N class labels. In a testing phase, each test image is normalized in height and squeezed in aspect ratio resulting in a plurality of test patches. The CNN averages the probabilities of each test patch belonging to a set of fonts to obtain a classification. Feature representations may be extracted and utilized to define font similarity between fonts, which may be utilized in font suggestion, font browsing, or font recognition applications.

    Abstract translation: 对卷积神经网络（CNN）进行字体识别和字体相似学习。 在训练阶段，通过引入差异来合成具有字体标签的文本图像，以最小化训练图像与真实世界文本图像之间的差距。 生成训练图像并将其输入到CNN中。 根据CNN正在训练的字体数量，输出被输入到N-way softmax函数中，产生N类标签上分类文本图像的分布。 在测试阶段，每个测试图像的高度被标准化，并以纵横比挤压，从而产生多个测试贴片。 CNN对属于一组字体的每个测试补丁的概率进行平均，以获得分类。 可以提取和利用特征表示来定义可以在字体建议，字体浏览或字体识别应用中使用的字体之间的字体相似性。

    Patch-based synthesis techniques using color and color gradient voting
    22.
    发明授权
    Patch-based synthesis techniques using color and color gradient voting 有权
    使用颜色和颜色渐变投票的基于贴片的综合技术

    公开(公告)号：US09317773B2

    公开(公告)日：2016-04-19

    申请号：US14478932

    申请日：2014-09-05

    Applicant: Adobe Systems Incorporated

    Inventor： Elya Shechtman ,  Daniel R. Goldman ,  Aliakbar Darabi

    IPC: G06K9/00 ,  G06K9/46 ,  G06K9/68 ,  G06T11/60 ,  G06K9/62

    CPC classification number: G06K9/4652 ,  G06K9/00 ,  G06K9/6202 ,  G06K9/68 ,  G06T11/60

    Abstract: Methods, apparatus, and computer-readable storage media for patch-based image synthesis using color and color gradient voting. A patch matching technique provides an extended patch search space that encompasses geometric and photometric transformations, as well as color and color gradient domain features. The photometric transformations may include gain and bias. The patch-based image synthesis techniques may also integrate image color and color gradients into the patch representation and replace conventional color averaging with a technique that performs voting for colors and color gradients and then solves a screened Poisson equation based on values for colors and color gradients when blending patch(es) with a target image.

    Abstract translation: 用于使用颜色和颜色梯度投票的基于贴片的图像合成的方法，装置和计算机可读存储介质。 补丁匹配技术提供了包含几何和光度变换以及颜色和颜色渐变域特征的扩展补丁搜索空间。 光度变换可以包括增益和偏置。 基于补丁的图像合成技术还可以将图像颜色和色彩渐变集成到贴片表示中，并且用对颜色和颜色渐变进行投票的技术代替常规颜色平均化，然后基于颜色和颜色梯度的值来解决筛选的泊松方程 当将补丁与目标图像混合时。

    Low memory content aware fill
    23.
    发明授权
    Low memory content aware fill 有权
    低内存含量意识填充

    公开(公告)号：US09305329B2

    公开(公告)日：2016-04-05

    申请号：US14339161

    申请日：2014-07-23

    Applicant: Adobe Systems Incorporated

    Inventor： Dan Goldman ,  Elya Shechtman

    IPC: G06K9/32 ,  G06T1/60 ,  G06T5/00

    CPC classification number: G06T1/60 ,  G06T5/005 ,  G06T2207/20021 ,  G06T2207/20172

    Abstract: A first image at a first resolution is received, the first image having a first hole therein. Based on the first image, a second image is generated at a second resolution lower than the first resolution, the second image having a second hole therein corresponding to the first hole. In the second image, one or more second-image source patches for the second hole are identified. At least one first-image source patch in the first image is identified based on a location of the identified second-image source patch. The identified at least one first-image source patch are stored in memory. Fill content are identified in the at least one first-image source patch stored in the memory. The identified fill content are placed in the first hole.

    Abstract translation: 接收第一分辨率的第一图像，第一图像在其中具有第一孔。 基于第一图像，以比第一分辨率低的第二分辨率产生第二图像，第二图像在其中具有与第一孔相对应的第二孔。 在第二图像中，识别用于第二孔的一个或多个第二图像源补丁。 基于所识别的第二图像源补丁的位置来识别第一图像中的至少一个第一图像源补丁。 所识别的至少一个第一图像源补丁存储在存储器中。 填充内容在存储在存储器中的至少一个第一图像源补丁中被识别。 所识别的填充内容被放置在第一个孔中。

    Visual pattern recognition in an image
    24.
    发明授权
    Visual pattern recognition in an image 有权
    图像中的视觉模式识别

    公开(公告)号：US09141885B2

    公开(公告)日：2015-09-22

    申请号：US13953394

    申请日：2013-07-29

    Applicant: Adobe Systems Incorporated

    Inventor： Jianchao Yang ,  Guang Chen ,  Jonathan Brandt ,  Hailin Jin ,  Elya Shechtman ,  Aseem Omprakash Agarwala

    IPC: G06K9/00 ,  G06K9/62

    CPC classification number: G06K9/627 ,  G06K9/4642

    Abstract: A system may be configured as an image recognition machine that utilizes an image feature representation called local feature embedding (LFE). LFE enables generation of a feature vector that captures salient visual properties of an image to address both the fine-grained aspects and the coarse-grained aspects of recognizing a visual pattern depicted in the image. Configured to utilize image feature vectors with LFE, the system may implement a nearest class mean (NCM) classifier, as well as a scalable recognition algorithm with metric learning and max margin template selection. Accordingly, the system may be updated to accommodate new classes with very little added computational cost. This may have the effect of enabling the system to readily handle open-ended image classification problems.

    Abstract translation: 系统可以被配置为利用称为局部特征嵌入（LFE）的图像特征表示的图像识别机器。 LFE能够生成捕获图像的显着视觉特性的特征向量，以解决识别图像中描绘的视觉图案的细粒度方面和粗粒度方面。 配置为利用具有LFE的图像特征向量，系统可以实现最近的等级均值（NCM）分类器，以及具有度量学习和最大边距模板选择的可缩放识别算法。 因此，可以更新系统以容纳新类别，而且增加了很少的计算成本。 这可能具有使系统能够容易地处理开放式图像分类问题的效果。

    Image Classification for Adjustment
    25.
    发明申请
    Image Classification for Adjustment 审中-公开
    调整图像分类

    公开(公告)号：US20140212054A1

    公开(公告)日：2014-07-31

    申请号：US13755214

    申请日：2013-01-31

    Applicant: ADOBE SYSTEMS INCORPORATED

    Inventor： Hyunjoon Lee ,  Elya Shechtman ,  Jue Wang

    IPC: G06K9/62

    CPC classification number: G06K9/3275 ,  G06K9/00664

    Abstract: Image classification techniques are described for adjustment of an image. In one or more implementations, an image is classified by one or more computing device based on suitability of the image for adjustment to correct perspective distortion of the image. Responsive to a classification of the image as not suitable for the adjustment, suitability of the image is detected for processing by a different image adjustment technique by the one or more computing devices.

    Abstract translation: 描述图像分类技术用于调整图像。 在一个或多个实现中，基于用于调整的图像的适合性来校正图像的透视失真，由一个或多个计算设备对图像进行分类。 响应于图像的分类不适于调整，通过一个或多个计算设备的不同图像调整技术来检测图像的适合性以进行处理。

    Image distractor detection and processing
    26.
    发明授权
    Image distractor detection and processing 有权

    公开(公告)号：US10134165B2

    公开(公告)日：2018-11-20

    申请号：US15597911

    申请日：2017-05-17

    Applicant: Adobe Systems Incorporated

    Inventor： Ohad I. Fried ,  Elya Shechtman ,  Daniel R. Goldman

    IPC: G06K9/40 ,  G09G5/00 ,  G06T11/60 ,  G06T7/11 ,  G06T5/00 ,  G06K9/46

    Abstract: Image distractor detection and processing techniques are described. In one or more implementations, a digital medium environment is configured for image distractor detection that includes detecting one or more locations within the image automatically and without user intervention by the one or more computing devices that include one or more distractors that are likely to be considered by a user as distracting from content within the image. The detection includes forming a plurality of segments from the image by the one or more computing devices and calculating a score for each of the plurality of segments that is indicative of a relative likelihood that a respective said segment is considered a distractor within the image. The calculation is performed using a distractor model trained using machine learning as applied to a plurality images having ground truth distractor locations.

Patent Agency Ranking