-
公开(公告)号：US09965719B2
公开(公告)日：2018-05-08
申请号：US15342823
申请日：2016-11-03
Applicant: NEC Laboratories America, Inc.
Inventor： Wongun Choi , Yuanqing Lin , Yu Xiang , Silvio Savarese
CPC classification number: G06N3/08 , G06K9/3233 , G06K9/628 , G06N3/0454 , G06T7/73 , G06T2207/20084 , G06T2207/30261
Abstract: A computer-implemented method for detecting objects by using subcategory-aware convolutional neural networks (CNNs) is presented. The method includes generating object region proposals from an image by a region proposal network (RPN) which utilizes subcategory information, and classifying and refining the object region proposals by an object detection network (ODN) that simultaneously performs object category classification, subcategory classification, and bounding box regression. The image is an image pyramid used as input to the RPN and the ODN. The RPN and the ODN each include a feature extrapolating layer to detect object categories with scale variations among the objects.
-
2.发明授权Window dependent feature regions and strict spatial layout for object detection 有权窗口依赖特征区域和严格的物体检测空间布局公开(公告)号：US09020248B2
公开(公告)日：2015-04-28
申请号：US14108280
申请日：2013-12-16
Applicant: NEC Laboratories America, Inc.
Inventor： Xiaoyu Wang , Ming Yang , Yuanqing Lin , Shenghuo Zhu
CPC classification number: G06K9/6257 , G06K9/4642 , G06K9/6202 , G06K9/6211 , G06K9/6212
Abstract: Systems and methods for object detection by receiving an image; segmenting the image and identifying candidate bounding boxes which may contain an object; for each candidate bounding box, dividing the box into overlapped small patches, and extracting dense features from the patches; during a training phase, applying a learning process to learn one or more discriminative classification models to classify negative boxes and positive boxes; and during an operational phase, for a new box generated from the image, applying the learned classification model to classify whether the box contains an object.
Abstract translation: 通过接收图像进行物体检测的系统和方法; 分割图像并识别可能包含对象的候选边界框; 对于每个候选边界框，将框分成重叠的小块，并从补丁中提取密集特征; 在培训阶段，应用学习过程学习一个或多个歧视性分类模型，以对负面框和正面框进行分类; 并且在操作阶段期间，对于从图像生成的新框，应用所学习的分类模型来分类所述框是否包含对象。
-
3.发明申请公开(公告)号：US20160307072A1
公开(公告)日：2016-10-20
申请号：US15095260
申请日：2016-04-11
Applicant: NEC Laboratories America, Inc.
Inventor： Feng Zhou , Yuanqing Lin
CPC classification number: G06K9/66 , G06K9/469 , G06K9/6267 , G06K9/627 , G06N3/0454 , G06N3/0472 , G06N3/084 , G06T7/0004 , G06T2207/30128
Abstract: Systems and methods are disclosed for deep learning and classifying images of objects by receiving images of objects for training or classification of the objects; producing fine-grained labels of the objects; providing object images to a multi-class convolutional neural network (CNN) having a softmax layer and a final fully connected layer to explicitly model bipartite-graph labels (BGLs); and optimizing the CNN with global back-propagation.
Abstract translation: 公开了用于深入学习和分类对象的图像的系统和方法，通过接收用于对象的训练或分类的对象的图像; 生产物品的细粒标签; 向具有softmax层和最终完全连接的层的多级卷积神经网络（CNN）提供对象图像以明确地模拟二分图标签（BGL）; 并利用全局反向传播优化CNN。
-
公开(公告)号：US09202144B2
公开(公告)日：2015-12-01
申请号：US14517211
申请日：2014-10-17
Applicant: NEC Laboratories America, Inc.
Inventor： Xiaoyu Wang , Yuanqing Lin , Will Zou , Miao Sun
CPC classification number: G06K9/66 , G06K9/4628
Abstract: Systems and methods are disclosed for detecting an object in an image by determining convolutional neural network responses on the image; mapping the responses back to their spatial locations in the image; and constructing features densely extract shift invariant activations of a convolutional neural network to produce dense features for the image.
Abstract translation: 公开了通过确定图像上的卷积神经网络响应来检测图像中的对象的系统和方法; 将响应映射回图像中的空间位置; 并且构造特征密集地提取卷积神经网络的移位不变激活以产生图像的密集特征。
-
公开(公告)号：US09042601B2
公开(公告)日：2015-05-26
申请号：US14108295
申请日：2013-12-16
Applicant: NEC Laboratories America, Inc.
Inventor： Xiaoyu Wang , Shenghuo Zhu , Ming Yang , Yuanqing Lin
CPC classification number: G06K9/66 , G06K9/4614 , G06K9/4676 , G06K9/6257
Abstract: Systems and methods are disclosed for object detection by receiving an image and extracting features therefrom; applying a learning process to determine sub-regions and select predetermined pooling regions; and performing selective max-pooling to choose one or more feature regions without noises.
Abstract translation: 公开了通过接收图像并从其中提取特征的对象检测的系统和方法; 应用学习过程来确定子区域并选择预定的汇集区域; 并执行选择性最大化池选择一个或多个没有噪声的特征区域。
-
Patent Agency Ranking
6.发明授权Dimension-wise spatial layout importance selection: an alternative way to handle object deformation 有权维度空间布局重要性选择：处理对象变形的替代方法公开(公告)号：US09020198B2
公开(公告)日：2015-04-28
申请号：US14108212
申请日：2013-12-16
Applicant: NEC Laboratories America, Inc.
Inventor： Xiaoyu Wang , Shenghuo Zhu , Ming Yang , Yuanqing Lin
CPC classification number: G06K9/6232 , G06K9/4614 , G06K9/4676 , G06K9/6257
Abstract: Systems and methods are disclosed for object detection by receiving an image; segmenting the image; extracting features from the image; and performing a dimension-wise spatial layout selection to pick up dimensions inside a discriminative spatial region for classification.
Abstract translation: 公开了通过接收图像进行物体检测的系统和方法; 分割图像; 从图像中提取特征; 以及执行维度空间布局选择以拾取鉴别空间区域内的维度以进行分类。
-
公开(公告)号：US08891908B2
公开(公告)日：2014-11-18
申请号：US14077424
申请日：2013-11-12
Applicant: NEC Laboratories America, Inc.
Inventor： Ming Yang , Xiaoyu Wang , Yuanqing Lin , Qi Tian
CPC classification number: G06F17/30247
Abstract: An image retrieval method includes learning multiple object category classifiers with a processor offline and generating classifications scores of images as the semantic attributes; performing vocabulary tree based image retrieval using local features with semantic-aware co-indexing to jointly embed two distinct cues offline for near-duplicate image retrieval; and identifying top similar or dissimilar images using multiple semantic attributes.
Abstract translation: 图像检索方法包括：处理器离线学习多个对象类别分类器，并生成分类图像分数作为语义属性; 使用具有语义感知共同索引的局部特征来执行基于词汇树的图像检索，以共同嵌入两个不同的线索以进行近似重复的图像检索; 并使用多个语义属性来识别顶部相似或不相似的图像。
-
公开(公告)号：US20140270367A1
公开(公告)日：2014-09-18
申请号：US14108295
申请日：2013-12-16
Applicant: NEC Laboratories America, Inc.
Inventor： Xiaoyu Wang , Shenghuo Zhu , Ming Yang , Yuanqing Lin
IPC: G06K9/66
CPC classification number: G06K9/66 , G06K9/4614 , G06K9/4676 , G06K9/6257
Abstract: Systems and methods are disclosed for object detection by receiving an image and extracting features therefrom; applying a learning process to determine sub-regions and select predetermined pooling regions; and performing selective max-pooling to choose one or more feature regions without noises.
Abstract translation: 公开了通过接收图像并从其中提取特征的对象检测的系统和方法; 应用学习过程来确定子区域并选择预定的汇集区域; 并执行选择性最大化池选择一个或多个没有噪声的特征区域。
-
公开(公告)号：US20170124409A1
公开(公告)日：2017-05-04
申请号：US15343017
申请日：2016-11-03
Applicant: NEC Laboratories America, Inc.
Inventor： Wongun Choi , Fan Yang , Yuanqing Lin
CPC classification number: G06K9/4628 , G06K9/3233 , G06N3/0454
Abstract: A computer-implemented method for training a convolutional neural network (CNN) is presented. The method includes receiving regions of interest from an image, generating one or more convolutional layers from the image, each of the one or more convolutional layers having at least one convolutional feature within a region of interest, applying at least one cascaded rejection classifier to the regions of interest to generate a subset of the regions of interest, and applying scale dependent pooling to convolutional features within the subset to determine a likelihood of an object category.
-
公开(公告)号：US09235904B1
公开(公告)日：2016-01-12
申请号：US14716435
申请日：2015-05-19
Applicant: NEC Laboratories America, Inc.
Inventor： Xiaoyu Wang , Yuanqing Lin
CPC classification number: G06K9/6256 , G01B11/14 , G06K9/3241 , G06K9/4652 , G06T7/77 , G06T2207/10024 , G06T2207/20081
Abstract: An object detector includes a bottom-up object hypotheses generation unit; a top-down object search with supervised descent unit; and an object re-localization unit with a localization model.
Abstract translation: 物体检测器包括自底向上物体假设生成单元; 从下到上的对象搜索与监督下降单位; 以及具有定位模型的对象重新定位单元。
-
-
-
-
-
-
-
-
-
Search Results
30
records
(took 0.034 seconds)
