专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

21. 发明申请

WO2016037848A1 IMAGE RECOGNITION USING DESCRIPTOR PRUNING 审中-公开
标题翻译：使用描述符预处理的图像识别
公开(公告)号：WO2016037848A1
公开(公告)日：2016-03-17
申请号：PCT/EP2015/069452
申请日：2015-08-25
申请人： THOMSON LICENSING
发明人： ZEPEDA SALVATIERRA, Joaquin , PEREZ, Patrick , RANA, Aakanksha
IPC分类号： G06K9/46 , G06F17/30
CPC分类号： G06T5/20 , G06K9/4676 , G06T2207/20081
摘要： The present disclosure relates to image recognition or image searching. More precisely, the present disclosure relates to pruning local descriptors extracted from an input image. The present disclosure proposes a system, method and device directed to the pruning of local descriptors extracted from image patches of an input image. The present disclosure prunes local descriptors assigned to a codebook cell, based on a relationship of the local descriptor and the assigned codebook cell. The present disclosure includes assigning a weight value for use in pruning based on the relationship of the local descriptor and the assigned codebook cell. This weight value is then used during the encoding of the local descriptors for use in image searching or image recognition.
摘要翻译：本公开涉及图像识别或图像搜索。更准确地说，本公开涉及从输入图像中提取的修剪局部描述符。本公开提出了一种针对从输入图像的图像块提取的局部描述符的修剪的系统，方法和装置。本发明基于本地描述符和分配的码本单元的关系来修剪分配给码本单元的本地描述符。本公开包括基于本地描述符和分配的码本小区的关系来分配用于修剪的权重值。然后在本地描述符的编码期间使用该权重值用于图像搜索或图像识别。

22. 发明申请

WO2015181179A1 METHOD AND APPARATUS FOR OBJECT TRACKING AND SEGMENTATION VIA BACKGROUND TRACKING 审中-公开
标题翻译：通过背景跟踪对象跟踪和分段的方法和装置
公开(公告)号：WO2015181179A1
公开(公告)日：2015-12-03
申请号：PCT/EP2015/061604
申请日：2015-05-26
申请人： THOMSON LICENSING
发明人： CRIVELLI, Tomas Enrique , PEREZ, Patrick , PEREZ RUA, Juan Manuel
IPC分类号： G06T7/20
CPC分类号： G06T7/269 , G06T7/194 , G06T7/215
摘要： A method for tracking an object commences by first establishing the object (12) in a current frame. Thereafter, a background region (202) is established encompassing the object in the current frame. The location for the object (12) is then estimated in a next frame. Next, the propagation of the background region (202) is determined. Finally, the object is segmented from its background based on propagation of the background region, thereby allowing tracking of the object from frame to frame.
摘要翻译：用于跟踪对象的方法通过首先在当前帧中建立对象（12）而开始。此后，建立包含当前帧中的对象的背景区域（202）。然后在下一帧中估计对象（12）的位置。接下来，确定背景区域（202）的传播。最后，基于背景区域的传播，对象从其背景被分割，从而允许从帧到帧跟踪对象。

23. 发明申请

WO2015181161A1 METHODS AND SYSTEMS FOR TOUCH INPUT 审中-公开
标题翻译：触摸输入的方法和系统
公开(公告)号：WO2015181161A1
公开(公告)日：2015-12-03
申请号：PCT/EP2015/061579
申请日：2015-05-26
申请人： THOMSON LICENSING
发明人： VARANASI, Kiran , PEREZ, Patrick
IPC分类号： G06F3/0488
CPC分类号： G06F3/04883 , G06F2203/04808
摘要： Various systems and methods for determining control inputs of a computing system based on trained touch inputs are provided. A set of first touch inputs can be obtained (601) as trained touch inputs. Each first touch input can be paired (602) with a sample control input of a computing system. A second touch input can be obtained (603). Interpolation of the set of first touch inputs based on the second touch input can be performed (604) to determine a runtime control input of the computing system.
摘要翻译：提供了用于基于训练的触摸输入来确定计算系统的控制输入的各种系统和方法。可以获得一组第一触摸输入（601）作为训练的触摸输入。每个第一触摸输入可以与计算系统的样本控制输入配对（602）。可以获得第二触摸输入（603）。可以执行基于第二触摸输入的该组第一触摸输入的插值（604）以确定计算系统的运行时控制输入。

24. 发明申请

WO2015132016A1 DEVICES AND METHODS FOR LEARNING AND APPLYING A DISTANCE METRIC 审中-公开
标题翻译：用于学习和应用距离公制的装置和方法
公开(公告)号：WO2015132016A1
公开(公告)日：2015-09-11
申请号：PCT/EP2015/051169
申请日：2015-01-21
申请人： THOMSON LICENSING
发明人： SHARMA, Gaurav , PEREZ, Patrick
IPC分类号： G06K9/00 , G06K9/62
CPC分类号： G06K9/00288 , G06K9/6256
摘要： A distance metric (63) is learned with a training set (61) including training images and pairwise information indicating whether subsets of those images are pairwise corresponding to same objects. The distance metric is associated (41) with adjustable projection spaces, adapted to projecting the training images into those projection spaces and to computing related local distances. The projection spaces are selected (42) in function of pairs of the subsets of training images, the distance metric corresponding to the selected spaces. Adjusted projection spaces are accordingly determined (43, 44) by reducing offsets between expected values based on the pairwise information and effective values, of the distance metric with respect to the objects for the training images. A distance metric is also applied to study images, thereby deciding whether those images correspond pairwise to same objects. Those are relevant to face verification and to identity based clustering of faces.
摘要翻译：使用包括训练图像的训练集（61）和指示这些图像的子集是否对应于相同对象的成对信息来学习距离度量（63）。距离度量与可调投影空间相关联（41），适于将训练图像投影到这些投影空间中并计算相关的局部距离。在训练图像的子集对中，对应于所选择的空间的距离度量，投影空间被选择（42）。因此，通过基于相对于训练图像的对象的距离度量的成对信息和有效值来减小预期值之间的偏移，从而确定调整的投影空间（43,44）。还应用距离度量来研究图像，从而决定这些图像是否与成对对象成对对应。这些与面部验证和基于身份的面孔聚类相关。

25. 发明申请

WO2014202392A1 METHOD AND APPARATUS FOR CONTROLLING EXAMPLE-BASED IMAGE MANIPULATION 审中-公开
标题翻译：用于控制基于实例的图像处理的方法和装置
公开(公告)号：WO2014202392A1
公开(公告)日：2014-12-24
申请号：PCT/EP2014/061617
申请日：2014-06-04
申请人： THOMSON LICENSING
发明人： HELLIER, Pierre , SABATER, Neus , PEREZ, Patrick
IPC分类号： G06F3/0484 , G06T11/60
CPC分类号： G06F3/038 , G06F3/017 , G06F3/02 , G06F3/04845 , G06F3/167 , G06F2203/0381 , G06T11/60
摘要： Method and apparatus for controlling example-based image manipulation A method and apparatus for controlling example-based image manipulation are disclosed where selection of example imagery and workpiece imagery is done with a pointing device (605)and an act of motion thereof. For increasing the efficiency of controlling, selection of an image characteristic to be transferred is derived from an interaction sensed on a second input device(603, 611, 612, 613, 614) during a time interval of the act of motion of the pointing device(605).
摘要翻译：用于控制基于示例的图像操作的方法和装置公开了一种用于控制基于示例的图像操纵的方法和装置，其中使用指示装置（605）及其运动动作来完成示例图像和工件图像的选择。为了提高控制效率，在指示装置的运动的时间间隔期间从在第二输入装置（603,611,612,613,614）上感测到的交互导出要传送的图像特征的选择（605）。

26. 发明申请

WO2014174058A1 METHOD OF OBTAINING A MEGA-FRAME IMAGE FINGERPRINTS FOR IMAGE FINGERPRINT BASED CONTENT IDENTIFICATION, METHOD OF IDENTIFYING A VIDEO SEQUENCE, AND CORRESPONDING DEVICE 审中-公开
标题翻译：用于基于图像指纹识别的内容识别的MEGA-FRAME图像指纹的识别方法，识别视频序列的方法和相应的设备
公开(公告)号：WO2014174058A1
公开(公告)日：2014-10-30
申请号：PCT/EP2014/058419
申请日：2014-04-25
申请人： THOMSON LICENSING
发明人： LEFEBVRE, Frédéric , ZEPEDA, Joaquin , PEREZ, Patrick
IPC分类号： G06K9/00
CPC分类号： G06K9/00744 , G06K9/6202
摘要： A temporal section that is defined by boundary images is determined in a video sequence. A maximum of k stable image frames are determined in the temporal section of image frames having a lowest temporal activity. Image fingerprints are computed from the determined stable image frames. A mega-frame image fingerprint data structure is constituted from the computed fingerprints and the data structure is stored in memory for use of fingerprint based content identification.
摘要翻译：在视频序列中确定由边界图像定义的时间部分。在具有最低时间活动度的图像帧的时间部分中确定最多k个稳定图像帧。从确定的稳定图像帧计算图像指纹。大图像图像指纹数据结构由计算出的指纹构成，数据结构被存储在存储器中以用于基于指纹的内容识别。

27. 发明申请

WO2014001095A1 METHOD FOR AUDIOVISUAL CONTENT DUBBING 审中-公开
标题翻译：用于音频内容的方法DUBBING
公开(公告)号：WO2014001095A1
公开(公告)日：2014-01-03
申请号：PCT/EP2013/062243
申请日：2013-06-13
申请人： THOMSON LICENSING
发明人： HELLIER, Pierre , OISEL, Lionel , PEREZ, Patrick
IPC分类号： G06K9/00
CPC分类号： G06K9/00281 , G11B27/034 , H04N5/23219 , H04N21/4318 , H04N21/44008
摘要： A method for processing a video shot comprising a set of traceable objects is described. One or more traceable objects of the set of traceable objects are selected. One or more zones respectively encompassing the one or more selected traceable objects are positioned such that the respective zones are kept at the same respective position between subsequent frames of the video shot. Content of the video shot outside the positioned zones is blurred.
摘要翻译：描述了一种用于处理包括一组可跟踪对象的视频镜头的方法。选择一组可跟踪对象中的一个或多个可跟踪对象。分别包围一个或多个所选择的可跟踪对象的一个或多个区域被定位成使得各个区域在视频镜头的后续帧之间保持在相同的相应位置。在定位区域外拍摄的视频的内容模糊。

28. 发明申请

WO2017102549A1 METHOD AND DEVICE FOR REFOCUSING AT LEAST ONE PLENOPTIC VIDEO 审中-公开
标题翻译：至少一个普通视频重新聚焦的方法和设备
公开(公告)号：WO2017102549A1
公开(公告)日：2017-06-22
申请号：PCT/EP2016/080285
申请日：2016-12-08
申请人： THOMSON LICENSING
发明人： HELLIER, Pierre , ALLIE, Valérie , PEREZ, Patrick
IPC分类号： G06T5/00 , G06T5/50
CPC分类号： G06T5/003 , G06T5/50 , G06T2200/21 , G06T2207/10052
摘要： A method for refocusing, on at least one common point of interest, the rendering of one set of plenoptic video data provided by one plenoptic device belonging to a set of plenoptic devices capturing simultaneously a same scene. According to the present disclosure, said method comprises: obtaining (21) a common 3D reference system used for spatially locating said plenoptic device that has provided said set of plenoptic video data and at least one other device of said set of plenoptic devices, from said at least one common point of interest, determining (22) common refocusing plane parameters in said common 3D reference system, refocusing (23) the rendering of said set of plenoptic video data by converting (231) said common refocusing plane parameters into a rendering refocusing plane of a 3D reference system associated with said plenoptic device.
摘要翻译：一种用于在至少一个公共兴趣点上重新聚焦由属于同时捕捉相同场景的一组全景设备的一个全光设备提供的一组全景视频数据的方法。根据本公开内容，所述方法包括：从所述集合中获得（21）用于在空间上定位已经提供所述一组全光视频数据的所述全光设备和所述一组全域设备中的至少一个其他设备的公共3D参考系统至少一个公共重点，在所述公共3D参考系统中确定（22）共同重聚焦平面参数，通过将所述共同重聚焦平面参数转换（231）为渲染重聚焦，重新聚焦（23）所述全光视频数据组的渲染与所述全光设备相关联的3D参考系统的平面。

29. 发明申请

WO2017077076A1 METHOD AND APPARATUS FOR GENERATING CODEBOOKS FOR EFFICIENT SEARCH 审中-公开
标题翻译：用于生成有效搜索的代码簿的方法和设备
公开(公告)号：WO2017077076A1
公开(公告)日：2017-05-11
申请号：PCT/EP2016/076734
申请日：2016-11-04
申请人： THOMSON LICENSING
发明人： JAIN, Himalaya , BILEN, Cagdas , ZEPEDA, Salvatierra Joaquin , PEREZ, Patrick
IPC分类号： G06F17/30
CPC分类号： G06F17/30277 , G06F17/30247
摘要： In a particular implementation, a codebook C can be used for quantizing a feature vector of a database image into a quantization index, and then a different codebook ( B ) can be used to approximate the feature vector based on the quantization index. The codebooks B and C can have different sizes. Before performing image search, a lookup table can be built offline to include distances between the feature vector for a query image and codevectors in codebook B to speed up the image search. Using triplet constraints wherein a first image and a second image are indicated as a matching pair and the first image and a third image as non-matching, the codebooks B and C can be trained for the task of image search. The present principles can be applied to regular vector quantization, product quantization, and residual quantization.
摘要翻译：在特定的实现中，可以使用码本C来将数据库图像的特征向量量化为量化索引，然后使用不同的码本（B ）可以用来近似基于量化索引的特征向量。码本和可以有不同的大小。在执行图像搜索之前，可以离线构建查找表以包括查询图像的特征向量与码本中的代码向量之间的距离，以加速图像搜索。使用三重约束，其中第一图像和第二图像被指示为匹配对并且第一图像和第三图像不匹配，码本B和C可以接受图像搜索任务的培训。本原理可应用于常规矢量量化，产品量化和残余量化。

30. 发明申请

WO2016075293A1 ACCELERATED SUPPORT VECTOR MACHINE (SVM) LEARNING USING CLUSTERING 审中-公开
标题翻译：加速支持向量机（SVM）使用聚类学习
公开(公告)号：WO2016075293A1
公开(公告)日：2016-05-19
申请号：PCT/EP2015/076567
申请日：2015-11-13
申请人： THOMSON LICENSING
发明人： ZEPEDA SALVATIERRA, Joaquin , PEREZ, Patrick
IPC分类号： G06K9/62
CPC分类号： G06K9/6269 , G06F17/30047 , G06N99/005
摘要： An aspect of present principles relate to methods and apparatus for image searching. The method may include determining a classifier based on at least a centroid representing a plurality of negative features and performing image searching based on the classifier. The apparatus may include an apparatus configured to determine a classifier based on centroids representing a plurality of negative features and to perform image searching based on the classifier. The centroids are determined based on a clustering of the plurality of negative features.
摘要翻译：本原理的一个方面涉及用于图像搜索的方法和装置。该方法可以包括基于表示多个负特征的至少一个质心来确定分类器，并且基于分类器执行图像搜索。该装置可以包括被配置为基于表示多个负特征的质心来确定分类器并且基于分类器执行图像搜索的装置。基于多个负特征的聚类来确定质心。

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