专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

EP3040884B1 METHOD AND DEVICE FOR CLASSIFYING PICTURES 有权
公开(公告)号：EP3040884B1
公开(公告)日：2019-10-09
申请号：EP15200908.0
申请日：2015-12-17
申请人： Xiaomi Inc.
发明人： ZHANG, Tao , LONG, Fei , CHEN, Zhijun
IPC分类号： G06F16/583

2. 发明公开

EP3179408A2 PICTURE PROCESSING METHOD AND APPARATUS, COMPUTER PROGRAM AND RECORDING MEDIUM 有权
标题翻译：联合国开发计划署副总理兼联合国儿童基金会
公开(公告)号：EP3179408A2
公开(公告)日：2017-06-14
申请号：EP16190953.6
申请日：2016-09-27
申请人： Xiaomi Inc.
发明人： CHEN, Zhijun , ZHANG, Tao , WANG, Pingze
IPC分类号： G06K9/00
CPC分类号： G06K9/00288 , G06F17/30256 , G06F17/30268 , G06K9/00295 , G06K9/00677 , G06K9/6267 , G06K2009/00328 , G06K2209/27 , G06T11/60 , G06T2207/30201
摘要： The present invention is related to a picture processing method and apparatus, a computer program and a recording medium. The method comprises: recognizing a face in a picture of a user; determining a personal identity corresponding to the recognized face, the personal identity comprising at least one of an identification of a person corresponding to the face and a relationship between the person corresponding to the face and the user; acquiring photographing information of the picture, the photographing information comprising at least one of a photographing time and a photographing location of the picture; and generating a description of the picture according to the personal identity and the photographing information. With this technical solution, a description of the picture can be generated more accurately, the intelligence for automatic description of a picture is improved to be closer to human's ability to describe the picture, the user can quickly get an accurate knowledge of every picture, and user experience is improved.
摘要翻译：本发明涉及图像处理方法和装置，计算机程序和记录介质。该方法包括：识别用户的画面中的脸部; 确定对应于所识别的面部的个人身份，所述个人身份包括与所述面部相对应的人的标识中的至少一个以及与所述面部对应的人与所述用户之间的关系中的至少一个; 获取图像的拍摄信息，所述拍摄信息包括图像的拍摄时间和拍摄位置中的至少一个; 以及根据个人身份和拍摄信息生成图片的描述。利用这一技术方案，可以更精确地生成图像的描述，改进了对图像自动描述的智能，以更接近人类描述图像的能力，用户可以快速获得每张图片的准确知识，用户体验得到改善。

3. 发明公开

EP3176735A1 METHOD AND DEVICE FOR ACQUIRING BUSINESS CARD INFORMATION 审中-公开
标题翻译：用于获取商业卡信息的方法和设备
公开(公告)号：EP3176735A1
公开(公告)日：2017-06-07
申请号：EP16200996.3
申请日：2016-11-28
申请人： Xiaomi Inc.
发明人： CHEN, Zhijun , HOU, Wendi , QIN, Qiuping
IPC分类号： G06Q10/00
CPC分类号： G06Q10/109 , G06K9/00228 , G06K9/00281 , G06K9/00288 , G06Q10/00
摘要： The present invention relates to a method and a device for acquiring business card information. The method applied to a server includes: receiving (201, 301, 401) a request for acquiring business card information sent by a terminal of a first user, in which the request for acquiring business card information includes facial feature information of a target user; determining (202) identity information of the target user according to the facial feature information of the target user; verifying (203, 306) the request for acquiring business card information sent by the terminal of the first user; sending (204, 307, 406) business card information corresponding to the identity information of the target user to the terminal of the first user if the request for acquiring business card information sent by the terminal of the first user is verified successfully.
摘要翻译：本发明涉及一种获取名片信息的方法和装置。应用于服务器的方法包括：接收第一用户的终端发送的获取名片信息的请求，获取名片信息的请求包括目标用户的面部特征信息; 根据所述目标用户的面部特征信息确定（202）所述目标用户的身份信息; 验证（203,306）第一用户的终端发送的获取名片信息的请求; 若所述第一用户的终端发送的获取名片信息的请求获得验证成功，则向所述第一用户的终端发送（204,307,406）与所述目标用户的身份信息对应的名片信息。

4. 发明公开

EP3173978A1 METHOD AND DEVICE FOR CHARACTERISTIC EXTRACTION 审中-公开
标题翻译：特征提取方法和装置
公开(公告)号：EP3173978A1
公开(公告)日：2017-05-31
申请号：EP16199678.0
申请日：2016-11-18
申请人： Xiaomi Inc.
发明人： LONG, Fei , CHEN, Zhijun , ZHANG, Tao
IPC分类号： G06K9/46 , G06K9/62
CPC分类号： G06K9/4671 , G06K9/4647 , G06K9/6249 , G06K2009/4695 , G06K2009/485
摘要： Method and device for characteristic extraction are provided in the disclosure, which belongs to the technical field of image processing. The method for characteristic extraction may comprise: dividing (102) the image into several blocks, each block comprising several cells; performing (104) sparse signal decomposition for each cell to get a sparse vector corresponding to each cell; extracting (106) a characteristic of Histogram of Oriented Gradient (HOG) of the image according to the sparse vectors.
摘要翻译：本发明提供了一种特征提取方法和装置，属于图像处理技术领域。用于特征提取的方法可以包括：将图像分成（102）几个块，每个块包括几个单元; 对每个单元执行（104）稀疏信号分解以获得对应于每个单元的稀疏矢量; 根据稀疏矢量提取（106）图像的定向梯度直方图（HOG）的特性。

5. 发明公开

EP3173940A1 METHOD AND DEVICE FOR IDENTIFYING INFORMATION AND COMPUTER-READABLE STORAGE MEDIUM 审中-公开
标题翻译： METHOD AND APPARATUS FOR识别信息和计算机可读存储介质
公开(公告)号：EP3173940A1
公开(公告)日：2017-05-31
申请号：EP16197837.4
申请日：2016-11-08
申请人： Xiaomi Inc.
发明人： WANG, Pingze , ZHANG, Tao , CHEN, Zhijun
IPC分类号： G06F17/27 , G06F17/30 , G06Q10/06
CPC分类号： G06F17/248 , G06F17/2705 , G06F17/2765 , G06F17/30705 , G06Q10/06 , H04W4/14
摘要： A method and device for identifying information and a computer-readable storage medium are provided in the disclosure. The method includes: obtaining at least one clause in target information to be identified, wherein the clause comprises a preset keyword (201); segmenting the clause to obtain one or more words and extracting a specified characteristic set from the one or more words, wherein the specified characteristic set comprises at least one characteristic word (202); identifying a labeling result of the preset keyword in the clause based on the preset keyword, the characteristic word, and a template that is constructed in advance (203). Accuracy of information identification is improved in the disclosure.
摘要翻译：在本公开内容提供了一种用于识别信息和一个计算机可读存储介质，方法和装置。该方法包括：获得在目标信息中的至少一个条款被识别，worin该条包括预设关键字（201）; 分割所述条款以获得一个或多个单词，并从一个或多个单词中提取指定的特征集，worin所指定的特征组包括至少一个特征字（202）; 识别基于预设关键字，特征字的子句中的预设关键字的标记结果，和模板也被预先（203）构成。的信息识别精度在本公开被改善。

6. 发明公开

EP3168776A1 IMAGE RECOGNITION METHOD AND DEVICE FOR GAME 审中-公开
标题翻译：图像识别方法和游戏装置
公开(公告)号：EP3168776A1
公开(公告)日：2017-05-17
申请号：EP16182244.0
申请日：2016-08-01
申请人： Xiaomi Inc.
发明人： CHEN, Zhijun , HOU, Wendi , WANG, Pingze
IPC分类号： G06K9/00 , A63F13/00
CPC分类号： G06K9/00268 , A63F13/213 , A63F13/31 , A63F13/323 , A63F13/35 , A63F13/655 , A63F13/71 , A63F13/79 , A63F13/822 , A63F13/825 , A63F13/837 , A63F13/87 , G06K9/00362
摘要： Disclosed is an image recognition method and device for a game, which belongs to the field of image processing. The method includes: when a picture to be matched is taken, acquiring (101) first body characteristic information corresponding to the picture to be matched, determining (102) a first user identifier corresponding to the first body characteristic information according to a pre-established relationship between a user identifier and body characteristic information, and executing (103) shooting processing operation on the first user identifier.
摘要翻译：公开了一种用于游戏的图像识别方法和装置，属于图像处理领域。该方法包括：在拍摄待匹配的图片时，获取（101）待匹配图片对应的第一身体特征信息，根据预先建立的第一身体特征信息确定（102）第一用户标识用户标识符和身体特征信息之间的关系，以及对第一用户标识符执行（103）拍摄处理操作。

7. 发明公开

EP3163509A1 METHOD FOR REGION EXTRACTION, METHOD FOR MODEL TRAINING, AND DEVICES THEREOF 审中-公开
标题翻译：区域提取方法，模型训练方法及其装置
公开(公告)号：EP3163509A1
公开(公告)日：2017-05-03
申请号：EP16195235.3
申请日：2016-10-24
申请人： Xiaomi Inc.
发明人： LONG, Fei , ZHANG, Tao , CHEN, Zhijun
IPC分类号： G06K9/34
CPC分类号： G06K9/4604 , G06K9/00456 , G06K9/344 , G06T7/0081
摘要： Method for region extraction, method for model training, and devices thereof are provided in the disclosure, which belongs to image processing field. The method for region extraction may include: acquiring (201) a recognition model, the recognition model being formed by training predefined sample images with a classification algorithm, the predefined sample images including positive sample images and negative sample images, each of the positive sample images having at least one numeral character, each of the negative sample images having no numeral character or incomplete numeral characters only; performing (202) recognition on an image to be recognized, in accordance with the recognition model, to obtain at least one numeral region; and performing (203) region segmentation on the numeral region to obtain at least one single-numeral region.
摘要翻译：本发明提供了一种区域提取方法，模型训练方法及其装置，属于图像处理领域。该区域提取方法可以包括：获取（201）识别模型，该识别模型通过训练预定义样本图像和分类算法形成，预定义样本图像包括正样本图像和负样本图像，每个正样本图像具有至少一个数字字符，每个负样本图像仅具有数字字符或不完整的数字字符; 根据识别模型对要识别的图像执行（202）识别以获得至少一个数字区域; 以及对所述数字区域执行（203）区域分割以获得至少一个单数字区域。

8. 发明公开

EP3133527A1 HUMAN FACE RECOGNITION METHOD, APPARATUS AND TERMINAL 审中-公开
标题翻译：男子歌手GESICHTSERKENNUNGSVERFAHREN，VORRICHTUNG UNDENDGERÄT
公开(公告)号：EP3133527A1
公开(公告)日：2017-02-22
申请号：EP16174462.8
申请日：2016-06-14
申请人： Xiaomi Inc.
发明人： CHEN, Zhijun , WANG, Pingze , QIN, Qiuping
IPC分类号： G06K9/00 , G06K9/32
CPC分类号： G06K9/00288 , G06K9/00234 , G06K9/00248 , G06K9/32 , G06K9/6262 , G06K9/627
摘要： A human face recognition method, a human face recognition apparatus and a terminal are provided. The method includes: adding a specified number of pixels to an edge area of the original image to acquire an image to be processed; performing human face recognition on the image to be processed; and determining a human face in the original image according to a result of the human face recognition.
摘要翻译：提供人脸识别方法，人脸识别装置和终端。该方法包括：将特定数量的像素添加到原始图像的边缘区域以获取要处理的图像; 对要处理的图像执行人脸识别; 以及根据人脸识别的结果来确定原始图像中的人脸。

9. 发明公开

EP3125156A1 METHOD, APPARATUS AND SERVER FOR IMAGE SCENE DETERMINATION 审中-公开
标题翻译： VERFAHREN，VORRICHTUNG UND SERVER ZUR BILDSZENENBESTIMMUNG
公开(公告)号：EP3125156A1
公开(公告)日：2017-02-01
申请号：EP16180960.3
申请日：2016-07-25
申请人： Xiaomi Inc.
发明人： ZHANG, Tao , CHEN, Zhijun , LONG, Fei
IPC分类号： G06K9/00 , G06K9/62
CPC分类号： G06K9/00718 , G06K9/00684 , G06K9/42 , G06K9/6262 , G06K9/66 , G06N3/04 , G06N3/08
摘要： The present disclosure refers to method, apparatus and server for image scene determination. Wherein the method comprises: obtaining (101) a gallery of a user terminal, the gallery comprising at least one image to be processed; identifying (102) the at least one the image to be processed using an image scene identification model respectively, to determine a scene to which each of the at least one the image to be processed corresponds; and marking (103) each one of the at least one the image to be processed with the scene to which it the image to be processed corresponds. It facilitates in classifying images to be processed in a gallery according to scenes they corresponds and providing the images to a user when the user is viewing the images, so as to improve users experience on usage of the gallery.
摘要翻译：本公开涉及用于图像场景确定的方法，装置和服务器。其中所述方法包括：获取（101）用户终端的画廊，所述画廊包括至少一个要处理的图像; 使用图像场景识别模型分别识别（102）要处理的图像中的至少一个，以确定要处理的至少一个图像中的每一个对应的场景; 以及将要处理的图像中的每一个与要处理的图像对应的场景标记（103）。它有助于根据他们对应的场景对图库中的图像进行分类处理，并在用户正在观看图像时向用户提供图像，从而提高用户使用图库的体验。

10. 发明公开

EP3113051A1 METHOD AND APPARATUS FOR ACQUIRING SEARCH RESULTS, COMPUTER PROGRAM AND RECORDING MEDIUM 审中-公开
标题翻译：方法和设备检测搜索结果，计算机程序和记录介质
公开(公告)号：EP3113051A1
公开(公告)日：2017-01-04
申请号：EP16157908.1
申请日：2016-02-29
申请人： Xiaomi Inc.
发明人： CHEN, Zhijun , ZHANG, Tao , LONG, Fei
IPC分类号： G06F17/30 , G06F21/32
CPC分类号： G06F17/30867 , G06F21/32 , H04L63/0861
摘要： This disclosure discloses a method and apparatus for acquiring search results, a computer program and a recording medium, pertaining to the technical field of Internet technology. The method comprises: acquiring face information of a user by face recognition when detecting that a terminal has triggered a search event; sending a search request to a search server according to the face information and the information to be searched in the search event; receiving user-related search results determined by the search server according to the face information and the information to be searched. When a terminal triggers a search event, it sends the face information and the information to be searched to the search server for login and search, and finally receives the user-related search results determined by the search server according to the face information and the information to be searched, whereby the efficiency of login during the search process, the accuracy of the search results and the human-machine interactivity are improved.
摘要翻译：本公开盘松动，用于获取搜索结果，计算机程序和记录介质中，与互联网技术的技术领域的方法和装置。该方法包括：通过面部识别获得的用户的面部信息当检测做了终端已触发的搜索事件; 发送搜索请求到一个搜索服务器gemäß至面部信息并将该信息在搜索事件被搜索; 接收由搜索服务器gemäß至面部信息并将该信息挖掘用户相关的搜索结果的确定性被搜索。当终端触发搜索事件时，它发送的面部信息和该信息来进行搜索以用于登录和搜索的搜索服务器，最后接收由搜索服务器开采的用户相关的搜索确定的结果雅丁到面部信息并将该信息要被搜索，从而登录的在搜索过程中的效率，搜索结果的准确度和所述人机交互得到改善。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式