Fast image classification by vocabulary tree based image retrieval
    1.
    发明授权
    Fast image classification by vocabulary tree based image retrieval 有权
    基于词汇树的图像检索快速图像分类

    公开(公告)号:US08787682B2

    公开(公告)日:2014-07-22

    申请号:US13339240

    申请日:2011-12-28

    IPC分类号: G06K9/62 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。

    FAST IMAGE CLASSIFICATION BY VOCABULARY TREE BASED IMAGE RETRIEVAL
    2.
    发明申请
    FAST IMAGE CLASSIFICATION BY VOCABULARY TREE BASED IMAGE RETRIEVAL 有权
    基于VOCABULARY TREE的图像检索快速图像分类

    公开(公告)号:US20120243789A1

    公开(公告)日:2012-09-27

    申请号:US13339240

    申请日:2011-12-28

    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。

    3D CONVOLUTIONAL NEURAL NETWORKS FOR AUTOMATIC HUMAN ACTION RECOGNITION
    5.
    发明申请
    3D CONVOLUTIONAL NEURAL NETWORKS FOR AUTOMATIC HUMAN ACTION RECOGNITION 有权
    3D自动人工神经网络识别

    公开(公告)号:US20110182469A1

    公开(公告)日:2011-07-28

    申请号:US12814328

    申请日:2010-06-11

    IPC分类号: G06K9/00

    CPC分类号: G06K9/00335 G06K9/4628

    摘要: Systems and methods are disclosed to recognize human action from one or more video frames by performing 3D convolutions to capture motion information encoded in multiple adjacent frames and extracting features from spatial and temporal dimensions therefrom; generating multiple channels of information from the video frames, combining information from all channels to obtain a feature representation for a 3D CNN model; and applying the 3D CNN model to recognize human actions.

    摘要翻译: 公开了系统和方法,以通过执行3D卷积来识别来自一个或多个视频帧的人类动作来捕获在多个相邻帧中编码的运动信息并从其空间和时间维度提取特征; 从视频帧生成多个信道,组合来自所有信道的信息以获得3D CNN模型的特征表示; 并应用3D CNN模型来识别人类的行为。

    Systems and methods for determining personal characteristics
    6.
    发明授权
    Systems and methods for determining personal characteristics 有权
    用于确定个人特征的系统和方法

    公开(公告)号:US08582807B2

    公开(公告)日:2013-11-12

    申请号:US12790979

    申请日:2010-05-31

    IPC分类号: G06K9/00

    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。

    SYSTEMS AND METHODS FOR DETERMINING PERSONAL CHARACTERISTICS
    7.
    发明申请
    SYSTEMS AND METHODS FOR DETERMINING PERSONAL CHARACTERISTICS 有权
    用于确定个人特征的系统和方法

    公开(公告)号:US20110222724A1

    公开(公告)日:2011-09-15

    申请号:US12790979

    申请日:2010-05-31

    IPC分类号: G06K9/62

    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。

    3D convolutional neural networks for automatic human action recognition
    9.
    发明授权
    3D convolutional neural networks for automatic human action recognition 有权
    3D卷积神经网络,用于自动人体行动识别

    公开(公告)号:US08345984B2

    公开(公告)日:2013-01-01

    申请号:US12814328

    申请日:2010-06-11

    IPC分类号: G06K9/00 G06K9/62 G06K9/64

    CPC分类号: G06K9/00335 G06K9/4628

    摘要: Systems and methods are disclosed to recognize human action from one or more video frames by performing 3D convolutions to capture motion information encoded in multiple adjacent frames and extracting features from spatial and temporal dimensions therefrom; generating multiple channels of information from the video frames, combining information from all channels to obtain a feature representation for a 3D CNN model; and applying the 3D CNN model to recognize human actions.

    摘要翻译: 公开了系统和方法,以通过执行3D卷积来识别来自一个或多个视频帧的人类动作来捕获在多个相邻帧中编码的运动信息并从其空间和时间维度提取特征; 从视频帧生成多个信道,组合来自所有信道的信息以获得3D CNN模型的特征表示; 并应用3D CNN模型来识别人类的行为。