公开(公告)号：US20090018981A1

公开(公告)日：2009-01-15

申请号：US11777482

申请日：2007-07-13

申请人： Cha Zhang ,  Paul Viola

发明人： Cha Zhang ,  Paul Viola

IPC分类号： G06F15/18

CPC分类号： G06K9/00288 ,  G06K9/6256 ,  G06K9/6282

摘要： A “Classifier Trainer” trains a combination classifier for detecting specific objects in signals (e.g., faces in images, words in speech, patterns in signals, etc.). In one embodiment “multiple instance pruning” (MIP) is introduced for training weak classifiers or “features” of the combination classifier. Specifically, a trained combination classifier and associated final threshold for setting false positive/negative operating points are combined with learned intermediate rejection thresholds to construct the combination classifier. Rejection thresholds are learned using a pruning process which ensures that objects detected by the original combination classifier are also detected by the combination classifier, thereby guaranteeing the same detection rate on the training set after pruning. The only parameter required throughout training is a target detection rate for the final cascade system. In additional embodiments, combination classifiers are trained using various combinations of weight trimming, bootstrapping, and a weak classifier termed a “fat stump” classifier.

摘要翻译： “分类器训练器”训练用于检测信号中的特定对象的组合分类器（例如，图像中的面部，语音中的词，信号中的模式等）。 在一个实施例中，引入了用于训练组合分类器的弱分类器或“特征”的“多实例修剪”（MIP）。 具体来说，将训练有素的组合分类器和用于设置假正/负操作点的相关联的最终阈值与学习的中间拒绝阈值组合以构建组合分类器。 使用修剪过程学习拒绝阈值，确保由组合分类器检测到原始组合分类器检测到的对象，从而保证修剪后训练集上的相同检测率。 训练所需的唯一参数是最终级联系统的目标检测率。 在另外的实施例中，组合分类器使用称为“胖树桩”分类器的重量修剪，自举和弱分类器的各种组合进行训练。

公开(公告)号：US07890443B2

公开(公告)日：2011-02-15

申请号：US11777482

申请日：2007-07-13

申请人： Cha Zhang ,  Paul Viola

发明人： Cha Zhang ,  Paul Viola

IPC分类号： G06N5/00

CPC分类号： G06K9/00288 ,  G06K9/6256 ,  G06K9/6282

摘要： A “Classifier Trainer” trains a combination classifier for detecting specific objects in signals (e.g., faces in images, words in speech, patterns in signals, etc.). In one embodiment “multiple instance pruning” (MIP) is introduced for training weak classifiers or “features” of the combination classifier. Specifically, a trained combination classifier and associated final threshold for setting false positive/negative operating points are combined with learned intermediate rejection thresholds to construct the combination classifier. Rejection thresholds are learned using a pruning process which ensures that objects detected by the original combination classifier are also detected by the combination classifier, thereby guaranteeing the same detection rate on the training set after pruning. The only parameter required throughout training is a target detection rate for the final cascade system. In additional embodiments, combination classifiers are trained using various combinations of weight trimming, bootstrapping, and a weak classifier termed a “fat stump” classifier.

摘要翻译： “分类器训练器”训练用于检测信号中的特定对象的组合分类器（例如，图像中的面部，语音中的词，信号中的模式等）。 在一个实施例中，引入了用于训练组合分类器的弱分类器或“特征”的“多实例修剪”（MIP）。 具体来说，将训练有素的组合分类器和用于设置假正/负操作点的相关联的最终阈值与学习的中间拒绝阈值组合以构建组合分类器。 使用修剪过程学习拒绝阈值，确保由组合分类器检测到原始组合分类器检测到的对象，从而保证修剪后训练集上相同的检测率。 训练所需的唯一参数是最终级联系统的目标检测率。 在另外的实施例中，组合分类器使用称为“胖树桩”分类器的重量修剪，自举和弱分类器的各种组合进行训练。

公开(公告)号：US08010471B2

公开(公告)日：2011-08-30

申请号：US11777464

申请日：2007-07-13

申请人： Cha Zhang ,  Paul Viola

发明人： Cha Zhang ,  Paul Viola

IPC分类号： G06F17/00 ,  G06N5/00

CPC分类号： G06K9/6256 ,  G06K9/00288 ,  G06K9/6282

摘要： A “Classifier Trainer” trains a combination classifier for detecting specific objects in signals (e.g., faces in images, words in speech, patterns in signals, etc.). In one embodiment “multiple instance pruning” (MIP) is introduced for training weak classifiers or “features” of the combination classifier. Specifically, a trained combination classifier and associated final threshold for setting false positive/negative operating points are combined with learned intermediate rejection thresholds to construct the combination classifier. Rejection thresholds are learned using a pruning process which ensures that objects detected by the original combination classifier are also detected by the combination classifier, thereby guaranteeing the same detection rate on the training set after pruning. The only parameter required throughout training is a target detection rate for the final cascade system. In additional embodiments, combination classifiers are trained using various combinations of weight trimming, bootstrapping, and a weak classifier termed a “fat stump” classifier.

摘要翻译： “分类器训练器”训练用于检测信号中的特定对象的组合分类器（例如，图像中的面部，语音中的词，信号中的模式等）。 在一个实施例中，引入了用于训练组合分类器的弱分类器或“特征”的“多实例修剪”（MIP）。 具体来说，将训练有素的组合分类器和用于设置假正/负操作点的相关联的最终阈值与学习的中间拒绝阈值组合以构建组合分类器。 使用修剪过程学习拒绝阈值，确保由组合分类器检测到原始组合分类器检测到的对象，从而保证修剪后训练集上相同的检测率。 训练所需的唯一参数是最终级联系统的目标检测率。 在另外的实施例中，组合分类器使用称为“胖树桩”分类器的重量修剪，自举和弱分类器的各种组合进行训练。

公开(公告)号：US20090018985A1

公开(公告)日：2009-01-15

申请号：US11777471

申请日：2007-07-13

申请人： Cha Zhang ,  Paul Viola

发明人： Cha Zhang ,  Paul Viola

IPC分类号： G06F15/18 ,  G06F17/00

CPC分类号： G06K9/6257 ,  G06K9/00248

摘要： A “Classifier Trainer” trains a combination classifier for detecting specific objects in signals (e.g., faces in images, words in speech, patterns in signals, etc.). In one embodiment “multiple instance pruning” (MIP) is introduced for training weak classifiers or “features” of the combination classifier. Specifically, a trained combination classifier and associated final threshold for setting false positive/negative operating points are combined with learned intermediate rejection thresholds to construct the combination classifier. Rejection thresholds are learned using a pruning process which ensures that objects detected by the original combination classifier are also detected by the combination classifier, thereby guaranteeing the same detection rate on the training set after pruning. The only parameter required throughout training is a target detection rate for the final cascade system. In additional embodiments, combination classifiers are trained using various combinations of weight trimming, bootstrapping, and a weak classifier termed a “fat stump” classifier.

摘要翻译： “分类器训练器”训练用于检测信号中的特定对象的组合分类器（例如，图像中的面部，语音中的词，信号中的模式等）。 在一个实施例中，引入了用于训练组合分类器的弱分类器或“特征”的“多实例修剪”（MIP）。 具体来说，将训练有素的组合分类器和用于设置假正/负操作点的相关联的最终阈值与学习的中间拒绝阈值组合以构建组合分类器。 使用修剪过程学习拒绝阈值，确保由组合分类器检测到原始组合分类器检测到的对象，从而保证修剪后训练集上的相同检测率。 训练所需的唯一参数是最终级联系统的目标检测率。 在另外的实施例中，组合分类器使用称为“胖树桩”分类器的重量修剪，自举和弱分类器的各种组合进行训练。

公开(公告)号：US07822696B2

公开(公告)日：2010-10-26

申请号：US11777471

申请日：2007-07-13

申请人： Cha Zhang ,  Paul Viola

发明人： Cha Zhang ,  Paul Viola

IPC分类号： G06F15/18 ,  G06F17/00

CPC分类号： G06K9/6257 ,  G06K9/00248

摘要： A “Classifier Trainer” trains a combination classifier for detecting specific objects in signals (e.g., faces in images, words in speech, patterns in signals, etc.). In one embodiment “multiple instance pruning” (MIP) is introduced for training weak classifiers or “features” of the combination classifier. Specifically, a trained combination classifier and associated final threshold for setting false positive/negative operating points are combined with learned intermediate rejection thresholds to construct the combination classifier. Rejection thresholds are learned using a pruning process which ensures that objects detected by the original combination classifier are also detected by the combination classifier, thereby guaranteeing the same detection rate on the training set after pruning. The only parameter required throughout training is a target detection rate for the final cascade system. In additional embodiments, combination classifiers are trained using various combinations of weight trimming, bootstrapping, and a weak classifier termed a “fat stump” classifier.

摘要翻译： “分类器训练器”训练用于检测信号中的特定对象的组合分类器（例如，图像中的面部，语音中的词，信号中的模式等）。 在一个实施例中，引入了用于训练组合分类器的弱分类器或“特征”的“多实例修剪”（MIP）。 具体来说，将训练有素的组合分类器和用于设置假正/负操作点的相关联的最终阈值与学习的中间拒绝阈值组合以构建组合分类器。 使用修剪过程学习拒绝阈值，确保由组合分类器检测到原始组合分类器检测到的对象，从而保证修剪后训练集上相同的检测率。 训练所需的唯一参数是最终级联系统的目标检测率。 在另外的实施例中，组合分类器使用称为“胖树桩”分类器的重量修剪，自举和弱分类器的各种组合进行训练。

公开(公告)号：US20090018980A1

公开(公告)日：2009-01-15

申请号：US11777464

申请日：2007-07-13

申请人： Cha Zhang ,  Paul Viola

发明人： Cha Zhang ,  Paul Viola

IPC分类号： G06F15/18

CPC分类号： G06K9/6256 ,  G06K9/00288 ,  G06K9/6282

摘要： A “Classifier Trainer” trains a combination classifier for detecting specific objects in signals (e.g., faces in images, words in speech, patterns in signals, etc.). In one embodiment “multiple instance pruning” (MIP) is introduced for training weak classifiers or “features” of the combination classifier. Specifically, a trained combination classifier and associated final threshold for setting false positive/negative operating points are combined with learned intermediate rejection thresholds to construct the combination classifier. Rejection thresholds are learned using a pruning process which ensures that objects detected by the original combination classifier are also detected by the combination classifier, thereby guaranteeing the same detection rate on the training set after pruning. The only parameter required throughout training is a target detection rate for the final cascade system. In additional embodiments, combination classifiers are trained using various combinations of weight trimming, bootstrapping, and a weak classifier termed a “fat stump” classifier.

摘要翻译： “分类器训练器”训练用于检测信号中的特定对象的组合分类器（例如，图像中的面部，语音中的词，信号中的模式等）。 在一个实施例中，引入了用于训练组合分类器的弱分类器或“特征”的“多实例修剪”（MIP）。 具体来说，将训练有素的组合分类器和用于设置假正/负操作点的相关联的最终阈值与学习的中间拒绝阈值组合以构建组合分类器。 使用修剪过程学习拒绝阈值，确保由组合分类器检测到原始组合分类器检测到的对象，从而保证修剪后训练集上的相同检测率。 训练所需的唯一参数是最终级联系统的目标检测率。 在另外的实施例中，组合分类器使用称为“胖树桩”分类器的重量修剪，自举和弱分类器的各种组合进行训练。

公开(公告)号：US20080187213A1

公开(公告)日：2008-08-07

申请号：US11671760

申请日：2007-02-06

申请人： Cha Zhang ,  Paul Viola ,  Sang Min Oh

发明人： Cha Zhang ,  Paul Viola ,  Sang Min Oh

IPC分类号： G06K9/62

CPC分类号： G06K9/00281

摘要： A landmark detection technique that can quickly detect both objects of interest and landmarks within the objects in an image using regression methods. The present fast landmark detection scheme reuses existing feature values used for object detection (e.g., face detection) to find the landmarks in an object (e.g., the eyes and mouth of the face). Hence, the technique provides landmark detection functionality at almost no cost.

摘要翻译： 一种地标检测技术，可以使用回归方法快速检测图像中对象内的感兴趣对象和地标。 现在的快速地标检测方案重用用于对象检测（例如，面部检测）的现有特征值以找到对象（例如，脸部的眼睛和嘴巴）中的界标。 因此，该技术几乎没有成本地提供地标检测功能。

公开(公告)号：US08873840B2

公开(公告)日：2014-10-28

申请号：US12959383

申请日：2010-12-03

申请人： Eyal Krupka ,  Igor Abramovski ,  Igor Kviatkovsky ,  Jason M. Cahill ,  Timothy R. O'Connor ,  Cha Zhang

发明人： Eyal Krupka ,  Igor Abramovski ,  Igor Kviatkovsky ,  Jason M. Cahill ,  Timothy R. O'Connor ,  Cha Zhang

IPC分类号： G06K9/62 ,  G06K9/00

CPC分类号： G06K9/6257 ,  G06K9/00228 ,  G06K9/6292

摘要： A training set for a post-filter classifier is created from the output of a face detector. The face detector can be a Viola Jones face detector. Face detectors produce false positives and true positives. The regions in the training set are labeled so that false positives are labeled negative and true positives are labeled positive. The labeled training set is used to train a post-filter classifier. The post-filter classifier can be an SVM (Support Vector Machine). The trained face detection classifier is placed at the end of a face detection pipeline comprising a face detector, one or more feature extractors and the trained post-filter classifier. The post-filter reduces the number of false positives in the face detector output while keeping the number of true positives almost unchanged using features different from the Haar features used by the face detector.

摘要翻译： 用于后置滤波器分类器的训练集由脸部检测器的输出创建。 脸部检测器可以是Viola Jones脸部检测器。 脸部检测仪产生假阳性和真阳性。 训练集中的区域被标记为假阳性标记为阴性，真阳性标记为阳性。 标记的训练集用于训练后置过滤分类器。 后置滤波器分类器可以是SVM（支持向量机）。 经训练的人脸检测分类器被放置在包括面部检测器，一个或多个特征提取器和经过训练的后置滤波器分类器的面部检测管线的末端。 后置滤波器减少了脸部检测器输出中的误报数量，同时使用与脸部检测器使用的Haar特征不同的特征保持真正的数量几乎不变。

公开(公告)号：US20130294710A1

公开(公告)日：2013-11-07

申请号：US13463934

申请日：2012-05-04

申请人： Philip Andrew Chou ,  Cha Zhang ,  Zhengyou Zhang ,  Shujie Liu

发明人： Philip Andrew Chou ,  Cha Zhang ,  Zhengyou Zhang ,  Shujie Liu

IPC分类号： G06K9/32

CPC分类号： G06K9/32 ,  G06T7/593

摘要： A temporal information integration dis-occlusion system and method for using historical data to reconstruct a virtual view containing an occluded area. Embodiments of the system and method use temporal information of the scene captured previously to obtain a total history. This total history is warped onto information captured by a camera at a current time in order to help reconstruct the dis-occluded areas. The historical data (or frames) from the total history match only a portion of the frames contained in the captured information. This warping yields warped history information. Warping is performed by using one of two embodiments to match points in an estimation of the current information to points in the captured information. Next, regions of current information are split using a classifier. The warped history information and the captured information then are merged to obtain an estimate for the current information and the reconstructed virtual view.

摘要翻译： 一种用于使用历史数据重建包含遮挡区域的虚拟视图的时间信息整合遮挡系统和方法。 系统和方法的实施例使用先前捕获的场景的时间信息来获得总历史。 这个总历史在当前时间由相机拍摄的信息扭曲，以帮助重建被遮挡的区域。 来自总历史记录的历史数据（或帧）仅匹配捕获信息中包含的帧的一部分。 这种扭曲产生扭曲的历史信息。 通过使用两个实施例中的一个实现扭曲，以将当前信息的估计中的点与捕获的信息中的点进行匹配。 接下来，使用分类器分割当前信息的区域。 然后将翘曲的历史信息和捕获的信息合并，以获得当前信息和重建的虚拟视图的估计。

公开(公告)号：US08510110B2

公开(公告)日：2013-08-13

申请号：US13546153

申请日：2012-07-11

申请人： Cha Zhang ,  Paul A. Viola ,  Pei Yin ,  Ross G. Cutler ,  Xinding Sun ,  Yong Rui

发明人： Cha Zhang ,  Paul A. Viola ,  Pei Yin ,  Ross G. Cutler ,  Xinding Sun ,  Yong Rui

IPC分类号： G10L15/00

CPC分类号： G06K9/6256 ,  G06K9/4614 ,  G10L25/78 ,  G10L2021/02166 ,  H04N7/147 ,  H04N7/15 ,  H04N21/42203 ,  H04N21/4223 ,  H04N21/4394 ,  H04N21/44008 ,  H04N21/44213 ,  H04N21/4788

摘要： Systems and methods for detecting people or speakers in an automated fashion are disclosed. A pool of features including more than one type of input (like audio input and video input) may be identified and used with a learning algorithm to generate a classifier that identifies people or speakers. The resulting classifier may be evaluated to detect people or speakers.

摘要翻译： 公开了以自动方式检测人或扬声器的系统和方法。 可以识别包括多于一种类型的输入（例如音频输入和视频输入）的功能池，并与学习算法一起使用以生成识别人或扬声器的分类器。 可以评估所得分类器以检测人或扬声器。