公开(公告)号：EP3229206A1

公开(公告)日：2017-10-11

申请号：EP17163444.7

申请日：2017-03-28

申请人： Xerox Corporation

发明人： GAIDON, Adrien

IPC分类号： G06T7/246 ,  G06T7/254

CPC分类号： G06K9/00718 ,  G06K9/00805 ,  G06K9/4628 ,  G06K9/6201 ,  G06K9/6256 ,  G06K9/6267 ,  G06K9/78 ,  G06N3/04 ,  G06N3/08 ,  G06T3/0087 ,  G06T7/20 ,  G06T7/248 ,  G06T7/254 ,  G06T2207/10016 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30196 ,  G06T2207/30236 ,  G06T2210/12

摘要： A system for applying video data to a neural network (NN) for online multi-class multi-object tracking includes a computer programed to perform an image classification method including the operations of receiving a video sequence; detecting candidate objects in each of a previous and a current video frame; transforming the previous and current video frames into a temporal difference input image; applying the temporal difference input image to a pre-trained neural network (NN) ( or deep convolutional network ) comprising an ordered sequence of layers; and based on a classification value received by the neural network, associating a pair of detected candidate objects in the previous and current frames as belonging to one of matching objects and different objects.

摘要翻译： 一种用于将视频数据应用于神经网络（NN）以用于在线多类别多对象追踪的系统，包括：计算机，其被编程为执行图像分类方法，所述图像分类方法包括接收视频序列的操作; 检测先前和当前视频帧中的每一个中的候选对象; 将先前和当前视频帧变换成时间差异输入图像; 将时间差异输入图像应用于包括有序序列的层的预先训练的神经网络（NN）（或深度卷积网络） 并且基于神经网络接收到的分类值，将前一帧和当前帧中检测到的一对候选对象关联为属于匹配对象和不同对象中的一个。

公开(公告)号：EP3211596A1

公开(公告)日：2017-08-30

申请号：EP17156383.6

申请日：2017-02-15

申请人： Xerox Corporation

发明人： WANG, Qiao ,  GAIDON, Adrien ,  VIG, Eleonora

IPC分类号： G06T7/20 ,  G06T19/00

CPC分类号： G06K9/6262 ,  G06K9/00711 ,  G06K9/00751 ,  G06T7/20 ,  G06T19/00 ,  G06T2207/10016 ,  G06T2207/30236

摘要： A system and method are suited for assessing video performance analysis. A computer graphics engine clones real-world data in a virtual world by decomposing the real-world data into visual components and objects in one or more object categories and populates the virtual world with virtual visual components and virtual objects. A scripting component controls the virtual visual components and the virtual objects in the virtual world based on the set of real-world data. A synthetic clone of the video sequence is generated based on the script controlling the virtual visual components and the virtual objects. The real-world data is compared with the synthetic clone of the video sequence and a transferability of conclusions from the virtual world to the real-world is assessed based on this comparison.

摘要翻译： 系统和方法适用于评估视频性能分析。 计算机图形引擎通过将真实世界的数据分解为一个或多个对象类别中的可视化组件和对象，并利用虚拟可视化组件和虚拟对象填充虚拟世界，从而克服虚拟世界中的实际数据。 脚本组件基于真实世界数据集来控制虚拟世界中的虚拟可视组件和虚拟对象。 基于控制虚拟可视组件和虚拟对象的脚本来生成视频序列的合成克隆。 将真实世界的数据与视频序列的合成克隆进行比较，并根据此比较评估从虚拟世界到现实世界的结论的可转移性。

公开(公告)号：EP3096292A1

公开(公告)日：2016-11-23

申请号：EP16168460.0

申请日：2016-05-04

申请人： Xerox Corporation

发明人： GAIDON, Adrien ,  VIG, Eleonora

IPC分类号： G06T7/20 ,  G06K9/00 ,  G06K9/62

CPC分类号： G06T7/2033 ,  G06K9/00 ,  G06K9/00771 ,  G06K9/3233 ,  G06K9/4628 ,  G06K9/6271 ,  G06K2009/3291 ,  G06T7/20 ,  G06T2207/10016 ,  G06T2207/20081

摘要： A tracking system and method are suited to tracking multiple of objects of different categories in a video sequence. A sequence of video frames is received and a set of windows is extracted from each frame in turn, based on a computed probability that the respective window contains an object, without reference to any specific category of object. For each of these windows, a feature representation is extracted. A trained detector for a selected category detects windows that constitute targets in that category, based on the respective feature representations. More than one detector can be used when there is more than one category of objects to be tracked. A target-specific appearance model is generated for each of the targets (e.g., learned or updated, if the target is present in a prior frame). The detected targets are tracked over one or more subsequent frames based on the target-specific appearance models of the targets.

摘要翻译： 跟踪系统和方法适合于跟踪视频序列中不同类别的多个对象。 基于所计算出的相应窗口包含对象的概率，而不引用任何特定类别的对象，接收一系列视频帧并依​​次从每个帧提取一组窗口。 对于这些窗口中的每一个，提取特征表示。 所选类别的经过训练的检测器基于相应的特征表示来检测构成该类别中的目标的窗口。 当有多个类别的对象被跟踪时，可以使用多个检测器。 为每个目标生成目标特定的外观模型（例如，如果目标在先前帧中存在，则学习或更新）。 基于目标的目标特定外观模型，检测到的目标在一个或多个后续帧上被跟踪。