-
1.发明授权公开(公告)号：US08682086B2
公开(公告)日：2014-03-25
申请号：US13109997
申请日：2011-05-18
申请人： Yuanqing Lin , Kai Yu
发明人： Yuanqing Lin , Kai Yu
CPC分类号： G06K9/6244 , G06K9/4676
摘要： Systems and methods process an image having a plurality of pixels includes an image sensor to capture an image; a first-layer to encode local patches on an image region; and a second layer to jointly encode patches from the same image region.
摘要翻译： 处理具有多个像素的图像的系统和方法包括用于捕获图像的图像传感器; 编码图像区域上的局部斑块的第一层; 以及第二层，以共同编码来自相同图像区域的斑块。
-
公开(公告)号：US08787682B2
公开(公告)日：2014-07-22
申请号：US13339240
申请日：2011-12-28
申请人： Ming Yang , Xiaoyu Wang , Yuanqing Lin , Kai Yu
发明人： Ming Yang , Xiaoyu Wang , Yuanqing Lin , Kai Yu
CPC分类号： G06K9/4676 , G06K2009/4695
摘要： Systems and methods are disclosed to categorize images by detecting local features for each image; applying a tree structure to index local features in the images; and extracting a rank list of candidate images with category tags based on a tree indexing structure to estimate a label of a query image.
摘要翻译： 公开了通过检测每个图像的局部特征来对图像进行分类的系统和方法; 应用树结构来索引图像中的局部特征; 以及基于树索引结构提取具有类别标签的候选图像的等级列表以估计查询图像的标签。
-
3.发明申请公开(公告)号：US20120243789A1
公开(公告)日：2012-09-27
申请号：US13339240
申请日：2011-12-28
申请人： Ming Yang , Xiaoyu Wang , Yuanqing Lin , Kai Yu
发明人： Ming Yang , Xiaoyu Wang , Yuanqing Lin , Kai Yu
IPC分类号： G06K9/46
CPC分类号： G06K9/4676 , G06K2009/4695
摘要： Systems and methods are disclosed to categorize images by detecting local features for each image; applying a tree structure to index local features in the images; and extracting a rank list of candidate images with category tags based on a tree indexing structure to estimate a label of a query image.
摘要翻译： 公开了通过检测每个图像的局部特征来对图像进行分类的系统和方法; 应用树结构来索引图像中的局部特征; 以及基于树索引结构提取具有类别标签的候选图像的等级列表以估计查询图像的标签。
-
4.发明申请公开(公告)号：US20110299789A1
公开(公告)日：2011-12-08
申请号：US13109997
申请日：2011-05-18
申请人： Yuanqing Lin , Kai Yu
发明人： Yuanqing Lin , Kai Yu
IPC分类号： G06K9/36
CPC分类号： G06K9/6244 , G06K9/4676
摘要： Systems and methods process an image having a plurality of pixels includes an image sensor to capture an image; a first-layer to encode local patches on an image region; and a second layer to jointly encode patches from the same image region.
摘要翻译： 处理具有多个像素的图像的系统和方法包括用于捕获图像的图像传感器; 编码图像区域上的局部斑块的第一层; 以及第二层，以共同编码来自相同图像区域的斑块。
-
公开(公告)号：US08639510B1
公开(公告)日：2014-01-28
申请号：US12341160
申请日：2008-12-22
申请人： Kai Yu , Rob A. Rutenbar
发明人： Kai Yu , Rob A. Rutenbar
CPC分类号： G10L15/08 , G10L15/32 , G10L2015/025 , G10L2015/085
摘要： A hardware acoustic scoring unit for a speech recognition system and a method of operation thereof are provided. Rather than scoring all senones in an acoustic model used for the speech recognition system, acoustic scoring logic first scores a set of ciphones based on acoustic features for one frame of sampled speech. The acoustic scoring logic then scores senones associated with the N highest scored ciphones. In one embodiment, the number (N) is three. While the acoustic scoring logic scores the senones associated with the N highest scored ciphones, high score ciphone identification logic operates in parallel with the acoustic scoring unit to identify one or more additional ciphones that have scores greater than a threshold. Once the acoustic scoring unit finishes scoring the senones for the N highest scored ciphones, the acoustic scoring unit then scores senones associated with the one or more additional ciphones.
摘要翻译： 提供了一种用于语音识别系统的硬件声学评分单元及其操作方法。 声学评分逻辑不是在用于语音识别系统的声学模型中对所有的声音进行评分，而是首先根据一帧采样语音的声学特征对一组虹吸进行评分。 声学评分逻辑然后评分与N个最高得分的虹吸管相关联的声调。 在一个实施例中，数（N）为3。 虽然声学评分逻辑评分与N个最高得分的虹吸管相关联的声音，但高分数人声识别逻辑与声学评分单元并行操作，以识别具有大于阈值的分数的一个或多个附加虹吸。 一旦声学评分单元完成对N个最高得分的虹吸分数的声分数，则声学评分单元然后评分与一个或多个附加虹吸管相关联的声调。
-
公开(公告)号：US08582807B2
公开(公告)日：2013-11-12
申请号：US12790979
申请日：2010-05-31
申请人： Ming Yang , Shenghuo Zhu , Fengjun Lv , Kai Yu
发明人： Ming Yang , Shenghuo Zhu , Fengjun Lv , Kai Yu
IPC分类号： G06K9/00
CPC分类号： G06K9/00248 , G06K9/4628 , G06K2009/00322
摘要： Systems and methods are disclosed for determining personal characteristics from images by generating a baseline gender model and an age estimation model using one or more convolutional neural networks (CNNs); capturing correspondences of faces by face tracking, and applying incremental learning to the CNNs and enforcing correspondence constraint such that CNN outputs are consistent and stable for one person.
摘要翻译： 公开了用于通过生成基线性别模型和使用一个或多个卷积神经网络（CNN）的年龄估计模型来确定来自图像的个人特征的系统和方法; 通过面部跟踪获取面部对应关系，并将增量学习应用于CNN，并执行对应约束，使得CNN输出对于一个人来说是一致和稳定的。
-
公开(公告)号：US08447119B2
公开(公告)日：2013-05-21
申请号：US12818156
申请日：2010-06-18
申请人： Kai Yu
发明人： Kai Yu
IPC分类号： G06K9/62
CPC分类号： G06K9/00664 , G06K9/6223 , G06K9/6269
摘要： Methods and systems are disclosed for image classification coding an image by nonlinearly mapping an image descriptor to form a high-dimensional sparse vector; spatially pooling each local region to form an image-level feature vector using a probability kernel incorporating a similarity metric of local descriptors; and classifying the image.
摘要翻译： 公开了通过非线性映射图像描述符以形成高维稀疏矢量的图像分类编码图像的方法和系统; 使用包含局部描述符的相似性度量的概率内核在空间上汇集每个局部区域以形成图像级特征向量; 并分类图像。
-
公开(公告)号：US20130011262A1
公开(公告)日：2013-01-10
申请号：US13634873
申请日：2011-05-22
申请人： Xuegen Xu , Kai Yu , Guorong He
发明人： Xuegen Xu , Kai Yu , Guorong He
IPC分类号： F03D1/06
CPC分类号： F03D7/0224 , F05B2240/2213 , F05B2250/311 , F05B2260/79 , Y02E10/721 , Y02E10/723
摘要： The invention designs a downwind variable pitch wind turbine generator which mainly comprises a wheel hub, a main shaft, a generator, a frame and a base, wherein a synchronizing disc is arranged in the wheel hub, the synchronizing disc is connected with a fan blade connection slewing bearing through two joint bearings which are in mutual butt joint, one end of a reciprocating rod arranged in the main shaft is connected with the synchronizing disc, the other end of the reciprocating rod is connected with a displacement sleeve, an electric push rod is arranged in the frame, and the displacement sleeve is connected with the electric push rod. The downward variable pitch wind turbine generator is mainly suitable for medium and small-sized fans, and a pitch control mechanism of the downward variable pitch wind turbine generator can control three blades to change pitch simultaneously and keep the pitch-changing amplitude consistent; and the pitch control mechanism can enable each blade to change the pitch to a certain negative angle, realize the shutdown of the downward variable pitch wind turbine generator and keep the shutdown state under harsh conditions, such as hurricane and the like. The frame can rotate 360 degrees relative to the base so as to adapt to all wind directions. In addition, the downwind variable pitch wind turbine generator is further provided with an anemometer and a controller, so that the reaction is more sensitive and the downwind variable pitch wind turbine generator can be suitable for wind direction and wind speed, which change suddenly.
摘要翻译： 本发明设计了一种顺风变桨距风力发电机，其主要包括轮毂，主轴，发电机，框架和基座，其中同步盘布置在轮毂中，同步盘与风扇叶片 连接通过两个相互对接的关节轴承的回转支承，布置在主轴上的往复杆的一端与同步盘连接，往复杆的另一端与排量套筒连接，电动推杆 布置在框架中，并且位移套筒与电动推杆连接。 向下变桨距风力发电机主要适用于中小型风机，向下变桨距风力发电机的俯仰控制机构可同时控制三台叶片同时变桨距，保持变桨幅度一致; 并且俯仰控制机构能够使各叶片将桨距改变为一定的负角度，实现向下变桨距风力发电机的关闭，并在诸如飓风等恶劣条件下保持关闭状态。 框架可以相对于基座旋转360度，以适应所有的风向。 此外，顺风变桨距风力发电机还设置有风速计和控制器，使得反应更加灵敏，而风向变风量风力发电机可以适应风向和风速的突然变化。
-
9.发明授权公开(公告)号：US08131732B2
公开(公告)日：2012-03-06
申请号：US12331346
申请日：2008-12-09
申请人： Kai Yu , Shenghuo Zhu , Yihong Gong
发明人： Kai Yu , Shenghuo Zhu , Yihong Gong
CPC分类号： G06F17/30867 , G06F17/16
摘要： A system is disclosed with a collaborative filtering engine to predict an active user's ratings/interests/preferences on a set of new products/items. The predictions are based on an analysis the database containing the historical data of many users' ratings/interests/preferences on a large set of products/items.
摘要翻译： 公开了一种具有协作过滤引擎的系统，以预测一组新产品/项目上的主动用户的评级/兴趣/偏好。 这些预测是基于对包含大量产品/项目的许多用户评级/兴趣/偏好的历史数据的数据库的分析。
-
公开(公告)号：US20110222724A1
公开(公告)日：2011-09-15
申请号：US12790979
申请日：2010-05-31
申请人： Ming Yang , Shenghuo Zhu , Fengjun Lv , Kai Yu
发明人： Ming Yang , Shenghuo Zhu , Fengjun Lv , Kai Yu
IPC分类号： G06K9/62
CPC分类号： G06K9/00248 , G06K9/4628 , G06K2009/00322
摘要： Systems and methods are disclosed for determining personal characteristics from images by generating a baseline gender model and an age estimation model using one or more convolutional neural networks (CNNs); capturing correspondences of faces by face tracking, and applying incremental learning to the CNNs and enforcing correspondence constraint such that CNN outputs are consistent and stable for one person.
摘要翻译： 公开了用于通过生成基线性别模型和使用一个或多个卷积神经网络（CNN）的年龄估计模型来确定来自图像的个人特征的系统和方法; 通过面部跟踪获取面部对应关系，并将增量学习应用于CNN，并执行对应约束，使得CNN输出对于一个人来说是一致和稳定的。
-
-
-
-
-
-
-
-
-
搜索结果
56 条记录 (耗时 0.02 秒)
