公开(公告)号：US20170228618A1

公开(公告)日：2017-08-10

申请号：US15495541

申请日：2017-04-24

Applicant: Huawei Technologies Co., Ltd. ,  Fudan University

Inventor： Yugang JIANG ,  Zuxuan WU ,  Xiangyang XUE ,  Zichen GU ,  Zhenhua CHAI

IPC: G06K9/62 ,  G06N3/04 ,  G06F17/16 ,  G06N3/08

CPC classification number: G06K9/6271 ,  G06F16/70 ,  G06F17/16 ,  G06K9/00718 ,  G06K9/6215 ,  G06K9/629 ,  G06N3/04 ,  G06N3/08 ,  G06N3/084

Abstract: A video classification method and apparatus are provided in embodiments of the present invention. The method includes: establishing a neural network classification model according to a relationship between features of video samples and a semantic relationship of the video samples; obtaining a feature combination of a to-be-classified video file; and classifying the to-be-classified video file by using the neural network classification model and the feature combination of the to-be-classified video file The neural network classification model is established according to the relationship between the features of the video samples and the semantic relationship of the video samples, and the relationship between the features and the semantic relationship are fully considered. Therefore, video classification accuracy are improved.

公开(公告)号：US20150227816A1

公开(公告)日：2015-08-13

申请号：US14581418

申请日：2014-12-23

Applicant: Huawei Technologies Co., Ltd.

Inventor： Xinyu DU ,  Dong WANG ,  Zichen GU ,  Daiyu WEI

IPC: G06K9/62 ,  G06K9/48 ,  G06K9/46 ,  G06T7/40 ,  G06K9/66 ,  G06T7/00

CPC classification number: G06K9/623 ,  G06K9/4623 ,  G06K9/4642 ,  G06K9/4652 ,  G06K9/481 ,  G06K9/6256 ,  G06K9/66 ,  G06K9/80 ,  G06T2207/10024 ,  G06T2207/20081 ,  G06T2207/20164

Abstract: The present invention provides a method and an apparatus for detecting a salient region of an image. Classification processing is performed on a test image according to an image feature vector of the test image by using a classifier obtained by means of pre-training, so as to obtain a classification label, where the classification label is used to indicate a salience detection algorithm for detecting a salient region of the test image. Salience detection is performed on the test image by using the salience detection algorithm indicated by the classification label, so as to obtain the salient region of the test image. Because a salience detection algorithm with the best detection effect is acquired by using the image feature vector of the test image, to detect the salient region of the test image, accuracy of salience detection is improved.

Abstract translation: 本发明提供一种用于检测图像的显着区域的方法和装置。 通过使用通过预训练获得的分类器，根据测试图像的图像特征向量对测试图像执行分类处理，以获得分类标签，其中分类标签用于指示突出检测算法 用于检测测试图像的显着区域。 通过使用由分类标签指示的突出检测算法，对测试图像进​​行显着性检测，以获得测试图像的显着区域。 因为通过使用测试图像的图像特征向量来获取具有最佳检测效果的突出检测算法，以检测测试图像的显着区域，提高了显着性检测的准确性。