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

US20170344848A1 GENERATING IMAGE FEATURES BASED ON ROBUST FEATURE-LEARNING 有权
公开(公告)号：US20170344848A1
公开(公告)日：2017-11-30
申请号：US15166164
申请日：2016-05-26
申请人： Adobe Systems Incorporated
发明人： Zhe Lin , Xiaohui Shen , Jonathan Brandt , Jianming Zhang
IPC分类号： G06K9/46 , G06K9/62 , G06K9/48 , G06N3/08 , G06N3/04
CPC分类号： G06K9/4628 , G06K9/6298 , G06K2009/00322 , G06N3/04 , G06N3/08
摘要： Techniques for increasing robustness of a convolutional neural network based on training that uses multiple datasets and multiple tasks are described. For example, a computer system trains the convolutional neural network across multiple datasets and multiple tasks. The convolutional neural network is configured for learning features from images and accordingly generating feature vectors. By using multiple datasets and multiple tasks, the robustness of the convolutional neural network is increased. A feature vector of an image is used to apply an image-related operation to the image. For example, the image is classified, indexed, or objects in the image are tagged based on the feature vector. Because the robustness is increased, the accuracy of the generating feature vectors is also increased. Hence, the overall quality of an image service is enhanced, where the image service relies on the image-related operation.

2. 发明申请

US20160132750A1 LOCAL FEATURE REPRESENTATION FOR IMAGE RECOGNITION 审中-公开
标题翻译：本地特征表征图像识别
公开(公告)号：US20160132750A1
公开(公告)日：2016-05-12
申请号：US14535963
申请日：2014-11-07
申请人： ADOBE SYSTEMS INCORPORATED
发明人： Jianchao Yang , Jonathan Brandt
IPC分类号： G06K9/52 , G06K9/46 , G06K9/62
CPC分类号： G06K9/52 , G06F17/30271 , G06K9/46 , G06K9/4642 , G06K9/6201 , G06K9/621 , G06K9/6226 , G06K9/6267 , G06K2009/4666
摘要： Techniques are disclosed for image feature representation. The techniques exhibit discriminative power that can be used in any number of classification tasks, and are particularly effective with respect to fine-grained image classification tasks. In an embodiment, a given image to be classified is divided into image patches. A vector is generated for each image patch. Each image patch vector is compared to the Gaussian mixture components (each mixture component is also a vector) of a Gaussian Mixture Model (GMM). Each such comparison generates a similarity score for each image patch vector. For each Gaussian mixture component, the image patch vectors associated with a similarity score that is too low are eliminated. The selectively pooled vectors from all the Gaussian mixture components are then concatenated to form the final image feature vector, which can be provided to a classifier so the given input image can be properly categorized.
摘要翻译：公开了用于图像特征表示的技术。该技术表现出可以在任意数量的分类任务中使用的辨别力，并且在细粒度图像分类任务方面特别有效。在一个实施例中，要分类的给定图像被分成图像斑块。为每个图像补丁生成一个向量。将每个图像块向量与高斯混合模型（GMM）的高斯混合分量（每个混合分量也是向量）进行比较。每个这样的比较生成每个图像块向量的相似性得分。对于每个高斯混合分量，消除了与相似度得分相关的图像块向量太低。然后将来自所有高斯混合分量的选择性汇集的向量连接起来以形成最终图像特征向量，其可以提供给分类器，从而可以对给定的输入图像进行适当的分类。

3. 发明申请

US20160062731A1 SHORTLIST COMPUTATION FOR SEARCHING HIGH-DIMENSIONAL SPACES 有权
标题翻译：搜索高维空间的列表计划
公开(公告)号：US20160062731A1
公开(公告)日：2016-03-03
申请号：US14473104
申请日：2014-08-29
申请人： Adobe Systems Incorporated
发明人： Zhe Lin , Jonathan Brandt , Xiaohui Shen , Jae-Pil Heo
IPC分类号： G06F7/24 , G06F17/30
CPC分类号： G06F7/24 , G06F17/30268 , G06F17/30271 , G06F17/30622
摘要： Techniques are disclosed for indexing and searching high-dimensional data using inverted file structures and product quantization encoding. An image descriptor is quantized using a form of product quantization to determine which of several inverted lists the image descriptor is to be stored. The image descriptor is appended to the corresponding inverted list with a compact coding using a product quantization encoding scheme. When processing a query, a shortlist is computed that includes a set of candidate search results. The shortlist is based on the orthogonality between two random vectors in high-dimensional spaces. The inverted lists are traversed in the order of the distance between the query and the centroid of a coarse quantizer corresponding to each inverted list. The shortlist is ranked according to the distance estimated by a form of product quantization, and the top images referred to by the ranked shortlist are reported as the search results.
摘要翻译：公开了用于使用反向文件结构和乘积量化编码索引和搜索高维数据的技术。使用产品量化的形式量化图像描述符，以确定要存储图像描述符的几个反转列表中的哪一个。使用产品量化编码方案，使用紧凑编码将图像描述符附加到相应的反转列表。在处理查询时，计算包括一组候选搜索结果的候选清单。该候选清单基于高维空间中的两个随机向量之间的正交性。按照与每个反向列表对应的粗略量化器的查询和质心之间的距离的顺序遍历反向列表。根据通过产品量化形式估计的距离对候选名单进行排序，并将由排名的候选名单引用的顶部图像报告为搜索结果。

4. 发明授权

US09053392B2 Generating a hierarchy of visual pattern classes 有权
标题翻译：生成视觉模式类的层次结构
公开(公告)号：US09053392B2
公开(公告)日：2015-06-09
申请号：US14012770
申请日：2013-08-28
申请人： Adobe Systems Incorporated
发明人： Jianchao Yang , Guang Chen , Hailin Jin , Jonathan Brandt , Elya Shechtman
IPC分类号： G06K9/62 , G06K9/68
CPC分类号： G06K9/6267 , G06K9/00979 , G06K9/4671 , G06K9/6219 , G06K9/6298 , G06K9/6828
摘要： A hierarchy machine may be configured as a clustering machine that utilizes local feature embedding to organize visual patterns into nodes that each represent one or more visual patterns. These nodes may be arranged as a hierarchy in which a node may have a parent-child relationship with one or more other nodes. The hierarchy machine may implement a node splitting and tree-learning algorithm that includes hard-splitting of nodes and soft-assignment of nodes to perform error-bounded splitting of nodes into clusters. This may enable the hierarchy machine, which may form all or part of a visual pattern recognition system, to perform large-scale visual pattern recognition, such as font recognition or facial recognition, based on a learned error-bounded tree of visual patterns.
摘要翻译：层次机器可以被配置为利用局部特征嵌入将可视图案组织成每个表示一个或多个视觉图案的节点的聚类机器。这些节点可以被布置为其中节点可以与一个或多个其他节点具有父子关系的层级。层次机器可以实现节点分割和树学习算法，其包括节点的硬分割和节点的软分配，以执行节点到分簇的有界限制的分割。这可以使得可以形成视觉图案识别系统的全部或一部分的层次机器基于学习的有界错误的视觉图案树来执行诸如字体识别或面部识别的大规模视觉模式识别。

5. 发明申请

US20150131873A1 EXEMPLAR-BASED FEATURE WEIGHTING 有权
标题翻译：基于EXEMPLAR的特征加权
公开(公告)号：US20150131873A1
公开(公告)日：2015-05-14
申请号：US14080010
申请日：2013-11-14
申请人： Adobe Systems Incorporated
发明人： Jonathan Brandt , Zhe Lin , Brandon M. Smith
IPC分类号： G06K9/00 , G06K9/62
CPC分类号： G06K9/00295 , G06K9/00 , G06K9/00268 , G06K9/00275 , G06K9/4633 , G06K9/6215
摘要： In an example embodiment, for each of the image exemplars, a first location offset between an actual landmark location for a first landmark in the image exemplar and a predicted landmark location for the first landmark in the image exemplar is determined. Then, a probability that the image recognition process applied using the first feature produces an accurate identification of the first landmark in the image exemplars is determined based on the first location offsets for each of the image exemplars. A weight may then be assigned to the first feature based on the derived probability. An image recognition process may then be performed on an image, the image recognition process utilizing a voting process, for each of one or more features, for one or more landmarks in the plurality of image exemplars, the voting process for the first feature weighted according to the weight assigned to the first feature.
摘要翻译：在示例实施例中，对于每个图像样本，确定在图像样本中的第一地标的实际地标位置与图像样本中的第一地标的预测地标位置之间的第一位置偏移。然后，基于每个图像样本的第一位置偏移来确定使用第一特征应用的图像识别处理产生图像样本中的第一地标的精确识别的概率。然后可以基于导出的概率将权重分配给第一特征。然后可以对图像执行图像识别处理，对于多个图像样本中的一个或多个地标，针对一个或多个特征中的每一个利用投票处理的图像识别处理，对第一特征的投票处理根据分配给第一个特征的权重。

6. 发明申请

US20150120760A1 IMAGE TAGGING 有权
标题翻译：图像标记
公开(公告)号：US20150120760A1
公开(公告)日：2015-04-30
申请号：US14068238
申请日：2013-10-31
申请人： Adobe Systems Incorporated
发明人： Zhaowen Wang , Jianchao Yang , Zhe Lin , Jonathan Brandt
IPC分类号： G06F17/30
CPC分类号： G06F17/30265 , G06K9/6263 , G06K2209/27
摘要： A system is configured to annotate an image with tags. As configured, the system accesses an image and generates a set of vectors for the image. The set of vectors may be generated by mathematically transforming the image, such as by applying a mathematical transform to predetermined regions of the image. The system may then query a database of tagged images by submitting the set of vectors as search criteria to a search engine. The querying of the database may obtain a set of tagged images. Next, the system may rank the obtained set of tagged images according to similarity scores that quantify degrees of similarity between the image and each tagged image obtained. Tags from a top-ranked subset of the tagged images may be extracted by the system, which may then annotate the image with these extracted tags.
摘要翻译：系统配置为使用标签注释图像。如所配置的，系统访问图像并生成图像的一组向量。可以通过数学变换图像来生成向量集合，例如通过对图像的预定区域应用数学变换。然后，系统可以通过将搜索标准的向量集合提交给搜索引擎来查询标记图像的数据库。数据库的查询可以获得一组标记的图像。接下来，系统可以根据量化图像和所获得的每个标记图像之间的相似度的相似度分数来对获得的标记图像集进行排序。来自标记图像的顶级子集的标签可以由系统提取，然后系统可以利用这些提取的标签来注释图像。

7. 发明授权

US08948517B2 Landmark localization via visual search 有权
标题翻译：通过视觉搜索进行地标定位
公开(公告)号：US08948517B2
公开(公告)日：2015-02-03
申请号：US13782804
申请日：2013-03-01
申请人： Adobe Systems Incorporated
发明人： Zhe Lin , Jonathan Brandt , Xiaohui Shen
IPC分类号： G06K9/46 , G06T7/00
CPC分类号： G06T7/0079 , G06K9/00281
摘要： One exemplary embodiment involves identifying feature matches between each of a plurality of object images and a test image, each of the feature matches between a feature of a respective object image and a matching feature of the test image, wherein there is a spatial relationship between each respective object image feature and a first landmark of the object image, the first landmark at a known location in the object image. The embodiment additionally involves estimating a plurality of locations for a second landmark for the test image, the estimated locations based at least in part on the feature matches and the spatial relationships, and estimating a final location for the second landmark from the plurality of locations for the second landmark for the test image.
摘要翻译：一个示例性实施例涉及识别多个对象图像中的每一个与测试图像之间的特征匹配，每个特征在相应对象图像的特征与测试图像的匹配特征之间匹配，其中每个对象图像之间存在空间关系对象图像的相应对象图像特征和第一界标，在对象图像中的已知位置处的第一界标。该实施例另外包括估计用于测试图像的第二地标的多个位置，至少部分地基于特征匹配和空间关系估计位置，以及从多个位置估计第二地标的最终位置，测试图像的第二个里程碑。

8. 发明申请

US20140247992A1 ATTRIBUTE RECOGNITION VIA VISUAL SEARCH 有权
标题翻译：通过视觉搜索进行属性识别
公开(公告)号：US20140247992A1
公开(公告)日：2014-09-04
申请号：US13782181
申请日：2013-03-01
申请人： ADOBE SYSTEMS INCORPORATED
发明人： Zhe Lin , Jonathan Brandt , Xiaohui Shen
IPC分类号： G06K9/62
CPC分类号： G06K9/6202 , G06K9/00228 , G06K2009/00328
摘要： One exemplary embodiment involves identifying feature matches between each of a plurality of object images and a test image, each feature matches between a feature of a respective object image and a matching feature of the test image, wherein there is a spatial relationship between each respective object image feature and a test image feature, and wherein the object depicted in the test image comprises a plurality of attributes. Additionally, the embodiment involves estimating, for each attribute in the test image, an attribute value based at least in part on information stored in a metadata associated with each of the object images.
摘要翻译：一个示例性实施例涉及识别多个对象图像和测试图像中的每一个之间的特征匹配，每个特征在相应对象图像的特征与测试图像的匹配特征之间匹配，其中每个相应对象之间存在空间关系图像特征和测试图像特征，并且其中测试图像中描绘的对象包括多个属性。另外，该实施例涉及至少部分地基于存储在与每个对象图像相关联的元数据中的信息来估计测试图像中的每个属性的属性值。

9. 发明申请

US20180336401A1 IDENTIFYING UNKNOWN PERSON INSTANCES IN IMAGES 审中-公开
公开(公告)号：US20180336401A1
公开(公告)日：2018-11-22
申请号：US16049322
申请日：2018-07-30
申请人： Adobe Systems Incorporated
发明人： Jonathan Brandt , Zhe Lin , Xiaohui Shen , Haoxiang Li
IPC分类号： G06K9/00 , G06K9/66 , G06K9/62
CPC分类号： G06K9/00295 , G06K9/00369 , G06K9/00677 , G06K9/6218 , G06K9/66
摘要： Methods and systems for recognizing people in images with increased accuracy are disclosed. In particular, the methods and systems divide images into a plurality of clusters based on common characteristics of the images. The methods and systems also determine an image cluster to which an image with an unknown person instance most corresponds. One or more embodiments determine a probability that the unknown person instance is each known person instance in the image cluster using a trained cluster classifier of the image cluster. Optionally, the methods and systems determine context weights for each combination of an unknown person instance and each known person instance using a conditional random field algorithm based on a plurality of context cues associated with the unknown person instance and the known person instances. The methods and systems calculate a contextual probability based on the cluster-based probabilities and context weights to identify the unknown person instance.

10. 发明授权

US10043101B2 Local feature representation for image recognition 有权
公开(公告)号：US10043101B2
公开(公告)日：2018-08-07
申请号：US14535963
申请日：2014-11-07
申请人： Adobe Systems Incorporated
发明人： Jianchao Yang , Jonathan Brandt
IPC分类号： G06K9/52 , G06K9/46 , G06K9/62 , G06F17/30
摘要： Techniques are disclosed for image feature representation. The techniques exhibit discriminative power that can be used in any number of classification tasks, and are particularly effective with respect to fine-grained image classification tasks. In an embodiment, a given image to be classified is divided into image patches. A vector is generated for each image patch. Each image patch vector is compared to the Gaussian mixture components (each mixture component is also a vector) of a Gaussian Mixture Model (GMM). Each such comparison generates a similarity score for each image patch vector. For each Gaussian mixture component, the image patch vectors associated with a similarity score that is too low are eliminated. The selectively pooled vectors from all the Gaussian mixture components are then concatenated to form the final image feature vector, which can be provided to a classifier so the given input image can be properly categorized.

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