公开(公告)号：US20170061254A1

公开(公告)日：2017-03-02

申请号：US15249449

申请日：2016-08-28

Applicant: THOMSON LICENSING

Inventor： Xavier BURGOS ,  Julien FLEUREAU ,  Francois LE CLERC

IPC: G06K9/62 ,  G06K9/66 ,  G06T7/00 ,  G06K9/52

CPC classification number: G06K9/6267 ,  G06K9/00281 ,  G06K9/3241 ,  G06K9/52 ,  G06K9/6282 ,  G06K9/66 ,  G06T7/0002 ,  G06T7/77 ,  G06T2207/20021 ,  G06T2207/20076 ,  G06T2207/20081

Abstract: A method for processing an image of pixels is described. The method includes determining a plurality of features representative of the image, obtaining both information representative of a probability of presence of an object in the image and information representative of a shape of the object by implementing a unique multivariate regression applied to at least a part of the features.

Abstract translation: 描述用于处理像素的图像的方法。 该方法包括确定表示图像的多个特征，通过实施应用于至少一部分图像的唯一多变量回归，获得表示图像中的对象的存在概率的信息和表示对象的形状的信息 特点。

公开(公告)号：US20160180545A1

公开(公告)日：2016-06-23

申请号：US14975665

申请日：2015-12-18

Applicant: THOMSON LICENSING

Inventor： Francois LE CLERC ,  Philippe ROBERT ,  Kiran VARANASI

IPC: G06T7/20 ,  G06K9/62

CPC classification number: G06T7/251 ,  G06K9/00771 ,  G06K9/6202 ,  G06K9/621 ,  G06T2207/10016 ,  G06T2207/10052 ,  G06T2207/20121

Abstract: A method and an electronic device for object tracking in a sequence of light-field captures. A data acquisition unit acquires a sequence of light-field captures, wherein each light-field capture comprises a plurality of views. A feature determining unit determines features of an initial visual appearance model for an object of interest in a reference view of a first light-field capture. A feature matching unit matches the features in the reference view and in the further views of the first light-field capture. A feature discarding unit discards features that cannot be well matched in all views of the first light-field capture. An appearance model building unit builds an updated visual appearance model for the object of interest based on the remaining features. Finally, a movement tracking unit tracks the movement of the object of interest in the sequence of light-field captures using the visual appearance model.

Abstract translation: 一种用于在光场捕获序列中进行物体跟踪的方法和电子装置。 数据采集​​单元获取光场捕获序列，其中每个光场捕获包括多个视图。 特征确定单元在第一光场捕获的参考视图中确定感兴趣对象的初始视觉外观模型的特征。 特征匹配单元与参考视图中的特征以及第一光场捕获的进一步视图相匹配。 特征丢弃单元丢弃在第一光场捕获的所有视图中不能很好匹配的特征。 外观模型构建单元基于剩余的特征为感兴趣的对象构建更新的视觉外观模型。 最后，运动跟踪单元使用视觉外观模型跟踪光场捕获序列中感兴趣对象的移动。

公开(公告)号：US20150339828A1

公开(公告)日：2015-11-26

申请号：US14404578

申请日：2013-05-30

Applicant: THOMSON LICENSING

Inventor： Abdelaziz DJELOUAH ,  Patrick PEREZ ,  Francois LE CLERC ,  Jean-Sebastien FRANCO ,  Edmond BOYER

IPC: G06T7/00

CPC classification number: G06T7/11 ,  G06T7/143 ,  G06T7/194 ,  G06T2207/10012 ,  G06T2207/10024 ,  G06T2207/30196

Abstract: A method and a module for segmenting a foreground region from a background region in a 3D scene captured by n capturing devices. A reduced number of 3D samples are selected in the scene. These 3D samples are projected in each captured image. Foreground probability and background probabilities are computed for each selected 3D sample based on color models and the projection of these samples in the images These probabilities are used to update the color models. These probabilities are then re-computed based on the updated color models. These steps are reiterated until the color models or the foreground and background probabilities of the selected 3D samples converge. A final segmentation is computed using foreground color models and foreground and background probabilities.

Abstract translation: 一种用于从由n个捕获设备捕获的3D场景中的背景区域分割前景区域的方法和模块。 在场景中选择减少数量的3D样本。 这些3D样本被投影在每个捕获的图像中。 基于颜色模型和这些样本在图像中的投影，为每个选定的3D样本计算前景概率和背景概率。这些概率用于更新颜色模型。 然后基于更新的颜色模型重新计算这些概率。 重复这些步骤，直到选定的3D样本的颜色模型或前景和背景概率收敛。 使用前景颜色模型和前景和背景概率计算最终分割。

公开(公告)号：US20180144212A1

公开(公告)日：2018-05-24

申请号：US15574299

申请日：2016-05-20

Applicant: THOMSON Licensing

Inventor： Xavier BURGOS ,  Cristina BORDEI ,  Francois LE CLERC

IPC: G06K9/62 ,  G06K9/00 ,  G06T5/50

CPC classification number: G06K9/6223 ,  G06K9/00201 ,  G06K9/00275 ,  G06K9/00281 ,  G06K9/00308 ,  G06K9/00315 ,  G06T5/50 ,  G06T11/001 ,  G06T2207/20221

Abstract: A method of generating a first image, the method including, for each second image of a set of second images representative of an object, obtaining an information representative of a shape of the object; clustering the second images into a plurality of clusters according to the information; for each cluster of the plurality of clusters, generating the first image using the information associated with each second image of the cluster, the first image being representative of the cluster. A corresponding device is also recited.

公开(公告)号：US20170061253A1

公开(公告)日：2017-03-02

申请号：US15249440

申请日：2016-08-28

Applicant: THOMSON LICENSING

Inventor： Xavier BURGOS ,  Nezha KABBAJ ,  Francois LE CLERC

IPC: G06K9/62 ,  G06K9/66 ,  G06T7/00 ,  G06K9/52

CPC classification number: G06K9/6267 ,  G06K9/00248 ,  G06K9/00268 ,  G06K9/00281 ,  G06K9/52 ,  G06K9/6215 ,  G06K9/66 ,  G06T7/70

Abstract: The disclosure relates to a method for determining the shape (St) of an object represented in a current image, a plurality of landmarks being associated with the object. According to the disclosure, the method includes determining a plurality of features (Ft) representative of the current image, and the features being determined from at least a part of the plurality of landmarks. The method further obtains information (St) representative of the shape of the object by using at least one parameter (Pt) representative of an estimation of the pose of the object in the current image, the at least one parameter being obtained by implementing a multivariate regression applied to at least a part of the set of features.

Abstract translation: 本公开涉及一种用于确定当前图像中表示的对象的形状（St）的方法，与对象相关联的多个界标。 根据本公开，该方法包括确定代表当前图像的多个特征（Ft），并且从多个地标的至少一部分确定特征。 该方法通过使用表示当前图像中的对象的姿态的估计的至少一个参数（Pt）来获得表示对象形状的信息（St），所​​述至少一个参数是通过实施多变量 回归应用于该组特征的至少一部分。

公开(公告)号：US20170061252A1

公开(公告)日：2017-03-02

申请号：US15249435

申请日：2016-08-28

Applicant: THOMSON LICENSING

Inventor： Xavier BURGOS ,  Nezha Kabbaj ,  Francois LE CLERC

IPC: G06K9/62 ,  G06K9/66 ,  G06T11/60 ,  G06K9/52

CPC classification number: G06K9/6267 ,  G06K9/00281 ,  G06K9/52 ,  G06K9/66 ,  G06T11/60

Abstract: The disclosure relates to a method for classifying an object of a current image, a plurality of first landmarks representative of the shape of the object being associated with the current image, a first unique identifier being associated with each first landmark. According to the disclosure, the method includes, for at least a first landmark, a step of obtaining at least a first descriptor describing an area of the current image having the at least first selected landmark. Then, the first landmark is selected according to its first identifier, and when its first identifier corresponds to a second identifier of a second landmark, a second descriptor is used in a classifier for classifying the object. Finally, the method determines information representative of confidence of the shape according to the first descriptor and according to weighting information associated with the second descriptor.

Abstract translation: 本公开涉及一种用于分类当前图像的对象的方法，代表与当前图像相关联的对象的形状的多个第一地标，与每个第一地标相关联的第一唯一标识符。 根据本公开，该方法至少包括第一标记，获得至少描述描述具有至少第一选定标记的当前图像的区域的第一描述符的步骤。 然后，根据其第一标识符选择第一标记，并且当其第一标识符对应于第二标记的第二标识符时，在用于分类对象的分类器中使用第二描述符。 最后，该方法确定表示根据第一描述符的形状置信度的信息，并根据与第二描述符相关联的加权信息。

公开(公告)号：US20160275987A1

公开(公告)日：2016-09-22

申请号：US15073544

申请日：2016-03-17

Applicant: THOMSON LICENSING

Inventor： Francois LE CLERC ,  Neus SABATER ,  Philippe SCHMOUKER

IPC: G11B27/031 ,  H04N13/04 ,  H04N13/00

CPC classification number: G11B27/031 ,  H04N5/23216 ,  H04N5/23229 ,  H04N5/23293 ,  H04N13/111

Abstract: A method and an apparatus for displaying light field video data are described. A receiving unit of the apparatus receives light field video data including a plurality of light field video sequences. Each light field video frames of of the light field video sequences has multiple selectable viewpoints. A light field video frame is provided by an operation unit and displayed on a display device with a first viewpoint. An input unit receives an input indicating a viewpoint different from the first viewpoint. The displayed light field video frame is then displayed with the indicated viewpoint. In one embodiment, two light field video frames of two light field video sequences are displayed respectively in main and side windows on the display device.

Abstract translation: 描述用于显示光场视频数据的方法和装置。 该装置的接收单元接收包括多个光场视频序列的光场视频数据。 光场视频序列的每个光场视频帧具有多个可选视点。 光场视频帧由操作单元提供并且以具有第一视点的显示装置显示。 输入单元接收指示与第一视点不同的视点的输入。 然后以指示的视点显示所显示的光场视频帧。 在一个实施例中，两个光场视频序列的两个光场视频帧分别显示在显示设备上的主窗口和侧窗口中。

公开(公告)号：US20140267777A1

公开(公告)日：2014-09-18

申请号：US14203471

申请日：2014-03-10

Applicant: THOMSON LICENSING

Inventor： Francois LE CLERC ,  Caroline BAILLARD ,  Nicolas MOLLET

IPC: H04N5/28

CPC classification number: H04N5/28 ,  G03B15/006 ,  G03B15/07 ,  G05D1/101 ,  H04N5/2224

Abstract: The present invention discloses a method for shooting a performance making use of umanned aerial vehicles, such drones for example, to provide the physical markers that are needed to give a physical actor indications on the positioning of virtual elements to be inserted later in the scene, and with which s/he needs to interact.

Abstract translation: 本发明公开了一种用于拍摄使用无人驾驶飞行器（例如无人机）的演奏的方法，以提供为物理运动员指示在稍后插入场景中的虚拟元素的定位所需的物理标记， 他/他需要与之进行互动。