公开(公告)号：US09965673B2

公开(公告)日：2018-05-08

申请号：US13977684

申请日：2011-04-11

申请人： Yangzhou Du ,  Jianguo Li ,  Ang Liu ,  Tao Wang ,  Yimin Zhang

发明人： Yangzhou Du ,  Jianguo Li ,  Ang Liu ,  Tao Wang ,  Yimin Zhang

IPC分类号： G06K9/00 ,  G06K9/46 ,  G06K9/32 ,  G06T7/246 ,  G06K9/68

CPC分类号： G06K9/00261 ,  G06K9/00221 ,  G06K9/00228 ,  G06K9/00248 ,  G06K9/00288 ,  G06K9/46 ,  G06K9/6857 ,  G06T7/248 ,  G06T2207/20016 ,  G06T2207/30201

摘要： Techniques are disclosed that involve face detection. For instance, face detection tasks may be decomposed into sets of one or more sub-tasks. In turn the sub-tasks of the sets may be allocated across multiple image frames. This allocation may be based on a multiple layer, quad-tree approach. In addition, face tracking tasks may be performed.

公开(公告)号：US09471829B2

公开(公告)日：2016-10-18

申请号：US13997296

申请日：2011-03-31

申请人： Ang Liu ,  Yangzhou Du ,  Tao Wang ,  Jianguo Li ,  Qiang Li ,  Yimin Zhang

发明人： Ang Liu ,  Yangzhou Du ,  Tao Wang ,  Jianguo Li ,  Qiang Li ,  Yimin Zhang

IPC分类号： G06K9/00 ,  G06K9/46 ,  G06K9/62

CPC分类号： G06K9/00248 ,  G06K9/00281 ,  G06K9/6209

摘要： Detecting facial landmarks in a face detected in an image may be performed by first cropping a face rectangle region of the detected face in the image and generating an integral image based at least in part on the face rectangle region. Next, a cascade classifier may be executed for each facial landmark of the face rectangle region to produce one response image for each facial landmark based at least in part on the integral image. A plurality of Active Shape Model (ASM) initializations may be set up. ASM searching may be performed for each of the ASM initializations based at least in part on the response images, each ASM search resulting in a search result having a cost. Finally, a search result of the ASM searches having a lowest cost function may be selected, the selected search result indicating locations of the facial landmarks in the image.

摘要翻译： 可以通过首先裁剪图像中检测到的面部的面部矩形区域并且至少部分地基于面部矩形区域来生成整体图像来检测在图像中检测到的面部中的面部界标。 接下来，可以对面部矩形区域的每个面部标记执行级联分类器，以至少部分地基于积分图像来为每个面部标记产生一个响应图像。 可以设置多个活动形状模型（ASM）初始化。 可以至少部分地基于响应图像来对每个ASM初始化执行ASM搜索，每个ASM搜索导致搜索结果具有成本。 最后，可以选择具有最低成本函数的ASM搜索的搜索结果，所选择的搜索结果指示图像中的面部地标的位置。

公开(公告)号：US09489567B2

公开(公告)日：2016-11-08

申请号：US13996496

申请日：2011-04-11

申请人： Tao Wang ,  Jianguo Li ,  Yangzhou Du ,  Qiang Li ,  Yimin Zhang

发明人： Tao Wang ,  Jianguo Li ,  Yangzhou Du ,  Qiang Li ,  Yimin Zhang

IPC分类号： G06K9/00

CPC分类号： G06K9/00288 ,  G06K9/00221

摘要： Methods, apparatuses, and articles associated with facial tracking and recognition are disclosed. In embodiments, facial images may be detected in video or still images and tracked. After normalization of the facial images, feature data may be extracted from selected regions of the faces to compare to associated feature data in known faces. The selected regions may be determined using a boosting machine learning processes over a set of known images. After extraction, individual two-class comparisons may be performed between corresponding feature data from regions on the tested facial images and from the known facial image. The individual two-class classifications may then be combined to determine a similarity score for the tested face and the known face. If the similarity score exceeds a threshold, an identification of the known face may be output or otherwise used. Additionally, tracking with voting may be performed on faces detected in video. After a threshold of votes is reached, a given tracked face may be associated with a known face.

摘要翻译： 公开了与面部跟踪和识别有关的方法，装置和文章。 在实施例中，可以在视频或静止图像中检测面部图像并进行跟踪。 在面部图像归一化之后，可以从面部的选定区域提取特征数据，以与已知面部中的相关特征数据进行比较。 可以使用一组已知图像上的升压机学习处理来确定所选择的区域。 提取后，可以在来自所测试的面部图像上的区域和来自已知面部图像的相应特征数据之间进行单独的两类比较。 然后可以组合个体两类分类以确定测试面部和已知面部的相似性得分。 如果相似性分数超过阈值，则可以输出或以其他方式使用已知面部的识别。 此外，可以对在视频中检测到的脸部进行投票跟踪。 在达成一个选票之后，一个给定的被追踪的面孔可能与一个已知的面孔相关联。

公开(公告)号：US20140369608A1

公开(公告)日：2014-12-18

申请号：US14124233

申请日：2013-06-14

申请人： Tao Wang ,  Lianrui Fu ,  Qiang Li ,  Jianguo Li ,  Yurong Chen ,  Ya-ti Peng ,  Yimin Zhang ,  Anbang Yao

发明人： Tao Wang ,  Lianrui Fu ,  Qiang Li ,  Jianguo Li ,  Yurong Chen ,  Ya-ti Peng ,  Yimin Zhang ,  Anbang Yao

IPC分类号： G06K9/46

CPC分类号： G06K9/6211 ,  G06K9/68

摘要： Apparatuses, methods and storage medium associated with processing an image are disclosed herein. In embodiments, a method for processing one or more images may include generating a plurality of pairs of keypoint features for a pair of images. Each pair of keypoint features may include a keypoint feature from each image. Further, for each pair of keypoint features, corresponding adjoin features may be generated. Additionally, for each pair of keypoint features, whether the adjoin features are similar may be determined. Whether the pair of images have at least one similar object may also be determined, based at least in part on a result of the determination of similarity between the corresponding adjoin features. Other embodiments may be described and claimed.

摘要翻译： 本文公开了与处理图像相关联的装置，方法和存储介质。 在实施例中，用于处理一个或多个图像的方法可以包括为一对图像生成多对关键点特征。 每对关键点特征可以包括来自每个图像的关键点特征。 此外，对于每对关键点特征，可以生成相应的邻接特征。 此外，对于每对关键点特征，可以确定邻接特征是否相似。 至少部分地基于相应邻接特征之间的相似性的确定结果，也可以确定该对图像是否具有至少一个相似对象。 可以描述和要求保护其他实施例。

公开(公告)号：US08605795B2

公开(公告)日：2013-12-10

申请号：US12211883

申请日：2008-09-17

申请人： Patricia P. Wang ,  Yimin Zhang ,  Tao Wang ,  Jianguo Li

发明人： Patricia P. Wang ,  Yimin Zhang ,  Tao Wang ,  Jianguo Li

IPC分类号： H04N7/18

CPC分类号： G11B27/034

摘要： Video editing methods and systems, including methods and systems to identify video clips having similar visual characteristics. Video clips may correspond to first and second videos, which may include a professional music video and a personal video, respectively. Identified video clips of the personal video may be combined into a new video clip, and music corresponding to visually similar video clips of the music video may be associated with the corresponding video clips of the new video. Video frames of the video clips may be characterized with respect to one or more visual features, which may include one or more of facial and/or body features, salient objects, camera motion, and image quality. Characterizations may be compared between video clips on an incremental basis. Characterization of a music video may implicitly model an underlying correlation between music rhythm and changes in visual appearance.

摘要翻译： 视频编辑方法和系统，包括识别具有相似视觉特征的视频剪辑的方法和系统。 视频剪辑可以分别对应于第一和第二视频，其可以分别包括专业音乐视频和个人视频。 识别的个人视频的视频剪辑可以被组合成新的视频剪辑，并且对应于音乐视频的视觉上相似的视频剪辑的音乐可以与新视频的相应视频剪辑相关联。 视频剪辑的视频帧可以相对于一个或多个视觉特征来表征，其可以包括面部和/或身体特征，显着对象，照相机运动和图像质量中的一个或多个。 可以在增量的基础上比较视频片段之间的特征。 音乐视频的表征可以隐含地模拟音乐节奏和视觉外观变化之间的底层相关性。

公开(公告)号：US08131065B2

公开(公告)日：2012-03-06

申请号：US11961659

申请日：2007-12-20

申请人： Jianguo Li ,  Tao Wang ,  Yimin Zhang

发明人： Jianguo Li ,  Tao Wang ,  Yimin Zhang

IPC分类号： G06K9/62

CPC分类号： G06K9/4619 ,  G06K9/6256

摘要： Machine-readable media, methods, apparatus and system for obtaining and processing image features are described. In some embodiments, groups of training features derived from regions of training images may be trained to obtain a plurality of classifiers, each classifier corresponding to each group of training features. The plurality of classifiers may be used to classify groups of validation features derived from regions of validation images to obtain a plurality of weights, wherein each weight corresponds to each region of the validation images and indicates how important the each region of the validation images is. Then, a weight may be discarded from the plurality of weights based upon a certain criterion.

摘要翻译： 描述了用于获得和处理图像特征的机器可读介质，方法，装置和系统。 在一些实施例中，可以训练从训练图像的区域导出的训练特征的组以获得多个分类器，每个分类器对应于每组训练特征。 多个分类器可以用于对从验证图像的区域导出的验证特征的组进行分类以获得多个权重，其中每个权重对应于验证图像的每个区域，并且指示验证图像的每个区域有多重要。 然后，可以基于一定的标准从多个权重中舍弃权重。

公开(公告)号：US20090169097A1

公开(公告)日：2009-07-02

申请号：US11967208

申请日：2007-12-30

申请人： Jianguo Li ,  Tao Wang ,  Yurong Chen ,  Yimin Zhang

发明人： Jianguo Li ,  Tao Wang ,  Yurong Chen ,  Yimin Zhang

IPC分类号： G06K9/46

CPC分类号： G06K9/4652 ,  G06K9/4638

摘要： A method for generating a Markov stationary color (MSC) descriptor is disclosed. The MSC descriptor may be used for image/video content representation, which characterizes both intra-color and inter-color spatial relationships in images. The MSC descriptor has a low storage requirement, relative to some other color descriptors.

摘要翻译： 公开了一种用于生成马尔科夫静止颜色（MSC）描述符的方法。 MSC描述符可以用于图像/视频内容表示，其表征图像中的色内和色间空间关系。 相对于其他一些颜色描述符，MSC描述符具有较低的存储要求。

公开(公告)号：US20090161967A1

公开(公告)日：2009-06-25

申请号：US11961649

申请日：2007-12-20

申请人： Jianguo Li ,  Tao Wang ,  Yimin Zhang

发明人： Jianguo Li ,  Tao Wang ,  Yimin Zhang

IPC分类号： G06K9/76

CPC分类号： G06K9/00248 ,  G06K9/4619 ,  G06K9/6217 ,  G06K9/6262

摘要： Machine-readable media, methods, apparatus and system for obtaining and processing image features are described. In some embodiments, a Gabor representation of an image may be obtained by using a Gabor filter. A region may be determined from the Gabor representation, wherein the region comprises a plurality of Gabor pixels of the Gabor representation; and, a sub-region may be determined from the region, wherein the sub-region comprises more than one of the plurality of Gabor pixels. Then, a Gabor feature may be calculated based upon a magnitude calculation related to the sub-region and the region.

摘要翻译： 描述了用于获得和处理图像特征的机器可读介质，方法，装置和系统。 在一些实施例中，可以通过使用Gabor滤波器来获得图像的Gabor表示。 可以从Gabor表示确定区域，其中该区域包括Gabor表示的多个Gabor像素; 并且可以从所述区域确定子区域，其中所述子区域包括所述多个Gabor像素中的多于一个。 然后，可以基于与子区域和区域相关的幅度计算来计算Gabor特征。

公开(公告)号：US08705810B2

公开(公告)日：2014-04-22

申请号：US12006019

申请日：2007-12-28

申请人： Tao Wang ,  Jianguo Li ,  Du Yangzhou ,  Yimin Zhang

发明人： Tao Wang ,  Jianguo Li ,  Du Yangzhou ,  Yimin Zhang

IPC分类号： G06K9/00

CPC分类号： G06K9/00268 ,  G06F17/30781

摘要： Apparatuses, systems, and computer program products that detect and/or index characters of videos are disclosed. One or more embodiments comprise an apparatus an apparatus having a feature extraction module and a cast indexing module. The feature extraction module may extract features of a scale invariant feature transform (SIFT) for face sets of a video and the cast indexing module may detect one or more characters of the video via one or more associations of clusters of the features. Some alternative embodiments may include a cast ranking module to sort characters of the video, considering such factors as appearance times of the characters, appearance frequencies of the characters, and page rankings of the characters. The apparatus may associate or partition the clusters based on a normalized cut process, as well as detect the characters based on measures of distances of nodes associated with the features. Numerous embodiments may detect the characters based upon partitioning the clusters via solutions for eigenvalue systems for matrices of nodes of the clusters.

摘要翻译： 公开了检测和/或索引视频字符的装置，系统和计算机程序产品。 一个或多个实施例包括具有特征提取模块和铸造分度模块的装置的装置。 特征提取模块可以提取视频的面部集合的尺度不变特征变换（SIFT）的特征，并且投射索引模块可以经由特征的聚类的一个或多个关联来检测视频的一个或多个字符。 考虑到字符的出现时间，字符的出现频率以及字符的页面排序等因素，一些备选实施例可以包括用于对视频的字符进行排序的投射排序模块。 该设备可以基于归一化的切割过程来关联或分割群集，并且基于与特征相关联的节点的距离的测量来检测角色。 许多实施例可以基于通过针对簇的节点的矩阵的特征值系统的解决方案来划分簇来检测字符。

公开(公告)号：US08620111B2

公开(公告)日：2013-12-31

申请号：US13431524

申请日：2012-03-27

申请人： Jianguo Li ,  Tao Wang ,  Yimin Zhang

发明人： Jianguo Li ,  Tao Wang ,  Yimin Zhang

IPC分类号： G06K9/60 ,  G06K9/62

CPC分类号： H04N9/8205

摘要： Methods and apparatus are disclosed to index digital frames. An example method includes identifying channel types associated with a plurality of image frames, splitting each one of the plurality of image frames into a respective color channel based on the identified channel types, applying a local binary pattern to each of the respective color channels to generate a respective pattern number, generating a spatial representation of each respective pattern number to determine transition probabilities for each channel type, and identifying a degree of similarity between the plurality of image frames based on the transition probabilities.

摘要翻译： 公开了索引数字帧的方法和装置。 示例性方法包括识别与多个图像帧相关联的频道类型，基于所识别的频道类型将多个图像帧中的每一个分成相应的颜色频道，将本地二进制模式应用于各个颜色通道中的每一个以产生 相应的模式号，产生每个相应模式号的空间表示，以确定每个信道类型的转换概率，并且基于转换概率来识别多个图像帧之间的相似度。