WO2017024507A1 - 根据用户面部特征推送广告的方法和广告推送系统 - Google Patents

根据用户面部特征推送广告的方法和广告推送系统 Download PDF

Info

Publication number
WO2017024507A1
WO2017024507A1 PCT/CN2015/086579 CN2015086579W WO2017024507A1 WO 2017024507 A1 WO2017024507 A1 WO 2017024507A1 CN 2015086579 W CN2015086579 W CN 2015086579W WO 2017024507 A1 WO2017024507 A1 WO 2017024507A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
facial feature
movie
advertisement
label
Prior art date
Application number
PCT/CN2015/086579
Other languages
English (en)
French (fr)
Inventor
常平
Original Assignee
常平
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 常平 filed Critical 常平
Priority to PCT/CN2015/086579 priority Critical patent/WO2017024507A1/zh
Publication of WO2017024507A1 publication Critical patent/WO2017024507A1/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising

Definitions

  • the present invention belongs to the field of the Internet, and more particularly to a method and an advertisement pushing system for pushing advertisements according to facial features of a user.
  • the key success factors of the advertising delivery mechanism are mainly two points:
  • Embodiments of the present invention provide a method for pushing an advertisement according to a facial feature of a user, and the purpose is to Aiming at the defects of current system identification of user interest, a solution is proposed to facilitate the system to more accurately advertise on the basis of more active and intelligent user interest recognition.
  • a method of pushing an advertisement based on a facial feature of a user comprising the steps of:
  • the specific tag is set as the user's interest corresponding to the facial feature. label
  • the user-on-demand movie corresponding to the facial feature is detected, the user corresponding to the facial feature is pushed with the same advertisement as the interest tag.
  • An embodiment of the present invention further provides an advertisement pushing system, including:
  • a recording unit configured to record a label corresponding to the movie and a label corresponding to the advertisement
  • a statistical unit configured to count a label corresponding to a user-on-demand movie corresponding to the specific facial feature
  • An interest identification unit whose input ends are respectively connected to the output end of the recording unit and the output end of the statistical unit, and is used for the number of times the movie corresponding to the specific facial feature is clicked on the specific tag is larger than the user's on-demand tape
  • the specific tag is set as the user's interest tag corresponding to the facial feature
  • An advertisement pushing unit whose input end is connected to the output end of the interest recognition unit, and is configured to push the user corresponding to the facial feature to have the same as the interest tag when detecting the user-on-demand movie corresponding to the facial feature ad.
  • the invention recognizes the user's interest corresponding to the facial feature by monitoring the user's facial feature and the movie corresponding to the facial feature, thereby realizing automatic push-related advertisement and enhancing the accuracy of the advertisement delivery.
  • FIG. 1 is a schematic flowchart of a method for pushing an advertisement according to a facial feature of a user according to an embodiment of the present invention
  • FIG. 2 is a schematic structural diagram of an advertisement pushing system according to an embodiment of the present invention.
  • Figure 1 is a diagram of an embodiment of the invention. A schematic flowchart of a method of pushing an advertisement according to a user's facial features, for convenience of explanation, only parts related to the embodiment of the present invention are shown.
  • step S101 the label corresponding to the movie and the label corresponding to the advertisement are recorded.
  • the 'Children' can be preset as its corresponding label in the movie "Happy Ocean Wolf”
  • the "Old Man” is preset as its corresponding label in the movie "Beijing Opera: Zhiwei Weihushan”.
  • the face recognition is applied rapidly, and the face recognition technology is now mature, and its application will become increasingly popular.
  • Face recognition especially computer technology that uses analysis and comparison. Face recognition is a popular field of computer technology research, face tracking detection, automatic adjustment of image magnification, nighttime infrared detection, automatic adjustment of exposure intensity; it belongs to biometrics technology, is a specific organism ) its own biological characteristics to distinguish between individual organisms.
  • the face recognition payment system was first established in July 2013 by the Finnish startup Uniqul. Launched the application for the first time in the world and submitted related patent applications.
  • the face recognition payment system developed by the Intelligent Multimedia Technology Research Center of the Chinese Academy of Sciences, users only need to use the relevant equipment to take a photo to achieve online payment, and the relevant experts said that the system is in the United States Carnegie Mellon International Standards Face recognition test library reached The recognition rate is 99.8%, while the best result of other international algorithms is 97.6%.
  • Apple has implemented the technology of 'low threshold facial recognition' on the mobile side.
  • Apple's patent captures images based on the front camera and then analyzes the images. Unlike traditional face recognition technology, Apple does not analyze all the details of the image, but only recognizes areas with large facial information, such as eyes, nose and mouth. This is also the low threshold for Apple's technology. 'Foundation.
  • step S102 the tag corresponding to the movie requested by the user corresponding to the specific facial feature is counted.
  • the statistical method can be used by the user to play related video resources, or the number of times the user clicks on the relevant video resources.
  • the prior art has been implemented, according to the facial features to unlock, corresponding to the relevant channel ordering and the like, so in the process of using the user, technically, the facial features and the requested content can be recorded in the system to form a computer identifiable.
  • the data has been implemented, according to the facial features to unlock, corresponding to the relevant channel ordering and the like, so in the process of using the user, technically, the facial features and the requested content can be recorded in the system to form a computer identifiable.
  • the data has been implemented, according to the facial features to unlock, corresponding to the relevant channel ordering and the like, so in the process of using the user, technically, the facial features and the requested content can be recorded in the system to form a computer identifiable.
  • step S103 The specific label is set to correspond to the facial feature when the number of times the movie corresponding to the specific facial feature is clicked on the movie with the specific tag is greater than the number of times the movie of the user has other tags is over a certain threshold.
  • User's interest tag
  • step S104 When the user-on-demand movie corresponding to the facial feature is detected, the user corresponding to the facial feature is pushed with the same advertisement as the interest tag.
  • the advertisement content and the movie content are further formed, and the advertisement delivery scheme further matches the user interest point, and the delivery scheme can be targeted for the individual user.
  • the invention recognizes the user's interest corresponding to the facial feature by monitoring the user's facial feature and the movie corresponding to the facial feature, thereby realizing automatic push-related advertisement and enhancing the accuracy of the advertisement delivery.
  • FIG. 2 is a schematic structural diagram of an advertisement pushing system according to an embodiment of the present invention, where the advertisement pushing system includes:
  • a recording unit 21 configured to record a label corresponding to the movie and a label corresponding to the advertisement
  • the statistic unit 22 is configured to count a label corresponding to a movie requested by the user corresponding to the specific facial feature
  • the interest recognition unit 23 has its input end and the output end of the recording unit 21 and the statistical unit 22, respectively.
  • the output connection is configured to set the specific label as the facial feature when the number of times the user clicks on the specific label-related movie corresponding to the specific facial feature is greater than the number of times the user clicks on the other labeled movie exceeds a certain threshold.
  • Advertising push unit 24 The input end is connected to the output end of the interest recognition unit, and is configured to push an advertisement corresponding to the interest tag to a user corresponding to the facial feature when detecting a user-on-demand movie corresponding to the facial feature.
  • the working principle is: the recording unit 21 records the label corresponding to the movie and the label corresponding to the advertisement; the statistical unit 22 Counting tags corresponding to a user-on-demand movie corresponding to a specific facial feature; interest recognition unit 23 When the number of times the user clicks on the specific tag-related movie corresponding to the specific facial feature is greater than the number of times the user clicks on the other tag's movie exceeds a certain threshold, the specific tag is set as the user's interest tag corresponding to the facial feature. , advertising push unit 24 When a user-on-demand movie corresponding to the facial feature is detected, the user corresponding to the facial feature is pushed with the same advertisement as the interest tag.
  • the invention recognizes the user's interest corresponding to the facial feature by monitoring the user's facial feature and the movie corresponding to the facial feature, thereby realizing automatic push-related advertisement and enhancing the accuracy of the advertisement delivery.

Abstract

一种根据用户面部特征推送广告的方法以及广告推送系统，属于互联网领域，所述方法包括：记录电影所对应的标签以及广告所对应的标签；统计与特定的面部特征对应的用户点播的电影所对应的标签；当用户点播带特定标签的电影的次数大于用户点播带其他标签的电影的次数超过一定阈值时，将所述特定标签设定为该面部特征锁对应的用户的兴趣标签；当检测到面部特征对应的用户点播电影时，向该面部特征对应的用户推送具有与所述兴趣标签相同的广告。该方法通过监控用户面部特征和该面部特征对应的电影，从而识别出该面部特征对应的用户的兴趣，从而实现自动推送相关广告，增强了广告投放的精确度。

Description

根据用户面部特征推送广告的方法和广告推送系统 技术领域
本发明属于互联网领域，尤其是涉及一种 根据用户面部特征推送广告的方法和广告推送系统 。
背景技术
广告投放机制在于的关键成功要素主要在于两点：
1 、广告的投放效果，即广告给用户留下的印象；
2 、广告的投放精度，即广告发送的受众。
对于广告投放效果而言，较多的属于艺术的创意；而对于广告的投放精度而言，较多属于广告的投放策略，是可以通过技术把控的。
现在在youku等影视网站点播影视中的用户很多，但在所点播的影视之前插入的广告给人的主观印象非常差，大多数广告都令人感到反感，其主要原因也是投放精度把握不够，将用户不感兴趣的内容强行塞给用户。
综上，针对于当前点播影视之前插入广告投放不精确、用户体验不佳的缺陷，需要提供一种结合用户兴趣投放广告的方法。
 技术问题
本发明实施例提供了 一种 根据用户面部特征推送广告的方法 ， 目的在于 针对于当前系统对用户兴趣识别的缺陷，特提出一个解决方案，以便于系统在用户兴趣识别方面更加主动和智能化的基础上能够更精确的投放广告。
技术解决方案
本发明是这样实现的： 一种 根据用户面部特征推送广告的方法，包括以下步骤：
记录电影所对应的标签以及广告所对应的标签；
统计与特定的面部特征对应的用户点播的电影所对应的标签；
当与所述特定面部特征对应的用户点播带特定标签的电影的次数大于用户点播带其他标签的电影的次数超过一定阈值时，将所述特定标签设定为该面部特征所对应的用户的兴趣标签；
当检测到面部特征对应的用户点播电影时，向该面部特征对应的用户推送具有与所述兴趣标签相同的广告。
本发明实施例还提供了一种 广告推送系统 ，包括：
记录单元，用于记录电影所对应的标签以及广告所对应的标签；
统计单元，用于统计与特定的面部特征对应的用户点播的电影所对应的标签；
兴趣识别单元，其输入端分别与所述记录单元的输出端以及所述统计单元的输出端连接，用于在与所述特定面部特征对应的用户点播带特定标签的电影的次数大于用户点播带其他标签的电影的次数超过一定阈值时，将所述特定标签设定为该面部特征所对应的用户的兴趣标签；
广告推送单元，其输入端与所述兴趣识别单元的输出端连接，用于在检测到所述面部特征对应的用户点播电影时，向该面部特征对应的用户推送具有与所述兴趣标签相同的广告。
 有益效果
该发明通过监控用户面部特征和该面部特征对应的电影，从而识别出该面部特征对应的用户的兴趣，从而实现自动推送相关广告，增强了广告投放的精确度。
附图说明
图 1 是本发明实施例提供的 一种 根据用户面部特征推送广告的方法 的流程示意图 ；
图 2 是本发明实施例提供的广告推送系统的结构示意图。
 本发明的实施方式
为了使本发明的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本发明进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。
图 1 是发明实施例提供的 一种 根据用户面部特征推送广告的方法的流程示意图，为了便于说明，只示出了与本发明实施例相关的部分。
在步骤 S101 中 记录电影所对应的标签以及广告所对应的标签。
在本实施例中，可在电影《喜洋洋灰太狼》上，将'儿童'预设为其对应标签，在电影《京剧：智取威虎山》上，将'老人'预设为其对应标签。
在本发明的实施例中， 应用了人脸识别急速，人脸识别技术现在已经较为成熟，其应用也将日益普及。
人脸识别，特指利用分析比较的计算机技术。人脸识别是一项热门的计算机技术研究领域，人脸追踪侦测，自动调整影像放大，夜间红外侦测，自动调整曝光强度；它属于生物特征识别技术，是对生物体（一般特指人）本身的生物特征来区分生物体个体。
人脸识别支付系统最早于 2013 年 7 月由芬兰创业公司 Uniqul 在全球首次推出应用，并提交了相关的专利申请。在国内，中国科学院智能多媒体技术研究中心开发的人脸识别支付系统，用户只需使用相关设备拍张照片即可实现在线支付，并且相关专家表示'该系统在美国卡内基梅隆国际标准人脸识别测试库上达到了 99.8% 的识别率，而国际上其他算法的最好结果为 97.6% 。
此外，据大智慧阿思达克通讯社报道，欧菲光与中国银行联合开发了一套安装于 ATM 的'人脸识别' 系统，目前这一系统已开发完毕并送样检测 , 已送样检测的产品为'包括摄像头、人脸识别模组与人脸识别软件'的一整套系统。此次合作是中行在互联网金融领域的全新尝试。
同时，根据 2015 年 7 月 8 日多家门户网站上转载的新闻看，苹果已经实现了在移动端'低门槛面部识别'的相关技术，苹果的这项专利基于前置摄像头捕捉图像，然后对图像进行分析处理。与传统的人脸识别技术不同的是，苹果不会对图像所有细节进行分析，而只会识别脸部信息较大的部位，如眼睛、鼻子与嘴巴等，这也是苹果这项技术'低门槛'的基础。
据此，该技术同样可以适用于本发明的相关领域。
在步骤 S102 中，统计与特定的面部特征对应的用户点播的电影所对应的标签。
该统计方式可以用户播放相关影视资源的时间，或用户点播相关影视资源的次数。
此外，现有技术已经实现，根据面部特征来解锁、对应相关频道排序等方案，所以用户在使用的过程中，技术上可以实现将面部特征与所点播的内容对应记录在系统中形成计算机可以识别的数据。
在步骤 S103 中，当在与所述特定面部特征对应的用户点播带特定标签的电影的次数大于用户点播带其他标签的电影的次数超过一定阈值时，将所述特定标签设定为该面部特征所对应的用户的兴趣标签；
这样就使得系统自动识别了用户点播影视资源的关注点和兴趣点。
在步骤 S104 中，当检测到面部特征对应的用户点播电影时，向该面部特征对应的用户推送具有与所述兴趣标签相同的广告。
这样就形成了广告内容和电影内容，进一步和用户兴趣点匹配的广告投送方案，且该投送方案能针对于用户个体进行投放。
该发明通过监控用户面部特征和该面部特征对应的电影，从而识别出该面部特征对应的用户的兴趣，从而实现自动推送相关广告，增强了广告投放的精确度。
图 2 是本发明实施例提供的一种 广告推送系统 的结构示意图，该广告推送系统包括 ：
记录单元 21 ，用于记录电影所对应的标签以及广告所对应的标签；
统计单元 22 ，用于统计与特定的面部特征对应的用户点播的电影所对应的标签；
兴趣识别单元 23 ，其输入端分别与所述记录单元 21 的输出端以及所述统计单元 22 的输出端连接，用于与所述特定面部特征对应的用户点播带特定标签的电影的次数大于用户点播带其他标签的电影的次数超过一定阈值时，将所述特定标签设定为该面部特征所对应的用户的兴趣标签；
广告推送单元 24 ，其输入端与所述兴趣识别单元的输出端连接，用于在检测到所述面部特征对应的用户点播电影时，向该面部特征对应的用户推送具有与所述兴趣标签相同的广告。
其工作原理是：记录单元 21 记录电影所对应的标签以及广告所对应的标签；统计单元 22 统计与特定的面部特征对应的用户点播的电影所对应的标签；兴趣识别单元 23 与所述特定面部特征对应的用户点播带特定标签的电影的次数大于用户点播带其他标签的电影的次数超过一定阈值时，将所述特定标签设定为该面部特征所对应的用户的兴趣标签，广告推送单元 24 在检测到面部特征对应的用户点播电影时，向该面部特征对应的用户推送具有与所述兴趣标签相同的广告。
该发明通过监控用户面部特征和该面部特征对应的电影，从而识别出该面部特征对应的用户的兴趣，从而实现自动推送相关广告，增强了广告投放的精确度。
以上仅为本发明的较佳实施例而已，并不用以限制本发明，凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明的保护范围之内。

Claims (2)

  1. 一种根据用户面部特征推送广告的方法 ，其特征在于，所述方法包括如下步骤：
    记录电影所对应的标签以及广告所对应的标签；
    统计与特定的面部特征对应的用户点播的电影所对应的标签；
    当与所述特定面部特征对应的用户点播带特定标签的电影的次数大于用户点播带其他标签的电影的次数超过一定阈值时，将所述特定标签设定为该面部特征所对应的用户的兴趣标签；
    当检测到面部特征对应的用户点播电影时，向该面部特征对应的用户推送具有与所述兴趣标签相同的广告。
  2. 一种 广告推送系统 ，其特征在于，所述 广告推送系统 包括：
    记录单元，用于记录电影所对应的标签以及广告所对应的标签；
    统计单元，用于统计与特定的面部特征对应的用户点播的电影所对应的标签；
    兴趣识别单元，其输入端分别与所述记录单元的输出端以及所述统计单元的输出端连接，用于在与所述特定面部特征对应的用户点播带特定标签的电影的次数大于用户点播带其他标签的电影的次数超过一定阈值时，将所述特定标签设定为该面部特征所对应的用户的兴趣标签；
    广告推送单元，其输入端与所述兴趣识别单元的输出端连接，用于在检测到所述面部特征对应的用户点播电影时，向该面部特征对应的用户推送具有与所述兴趣标签相同的广告。
PCT/CN2015/086579 2015-08-11 2015-08-11 根据用户面部特征推送广告的方法和广告推送系统 WO2017024507A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2015/086579 WO2017024507A1 (zh) 2015-08-11 2015-08-11 根据用户面部特征推送广告的方法和广告推送系统

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2015/086579 WO2017024507A1 (zh) 2015-08-11 2015-08-11 根据用户面部特征推送广告的方法和广告推送系统

Publications (1)

Publication Number Publication Date
WO2017024507A1 true WO2017024507A1 (zh) 2017-02-16

Family

ID=57984601

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/086579 WO2017024507A1 (zh) 2015-08-11 2015-08-11 根据用户面部特征推送广告的方法和广告推送系统

Country Status (1)

Country Link
WO (1) WO2017024507A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107563813A (zh) * 2017-09-05 2018-01-09 北京奇艺世纪科技有限公司 一种pdb广告推送频次的控制方法及系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090010499A1 (en) * 2007-02-21 2009-01-08 Vaelsys Formacion Y Desarrollo S.L. Advertising impact measuring system and method
CN102346898A (zh) * 2010-09-20 2012-02-08 微软公司 自动定制广告生成系统
CN103279878A (zh) * 2013-05-22 2013-09-04 广州威信信息科技发展有限公司 一种互动数字电视广告精准投放系统及方法
CN104486680A (zh) * 2014-12-19 2015-04-01 珠海全志科技股份有限公司 基于视频的广告推送方法及系统
CN104573619A (zh) * 2014-07-25 2015-04-29 北京智膜科技有限公司 基于人脸识别的智能广告大数据分析方法及系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090010499A1 (en) * 2007-02-21 2009-01-08 Vaelsys Formacion Y Desarrollo S.L. Advertising impact measuring system and method
CN102346898A (zh) * 2010-09-20 2012-02-08 微软公司 自动定制广告生成系统
CN103279878A (zh) * 2013-05-22 2013-09-04 广州威信信息科技发展有限公司 一种互动数字电视广告精准投放系统及方法
CN104573619A (zh) * 2014-07-25 2015-04-29 北京智膜科技有限公司 基于人脸识别的智能广告大数据分析方法及系统
CN104486680A (zh) * 2014-12-19 2015-04-01 珠海全志科技股份有限公司 基于视频的广告推送方法及系统

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107563813A (zh) * 2017-09-05 2018-01-09 北京奇艺世纪科技有限公司 一种pdb广告推送频次的控制方法及系统
CN107563813B (zh) * 2017-09-05 2021-01-22 北京奇艺世纪科技有限公司 一种pdb广告推送频次的控制方法及系统

Similar Documents

Publication Publication Date Title
WO2017028093A1 (zh) 广告投放方法和广告投放系统
WO2017024507A1 (zh) 根据用户面部特征推送广告的方法和广告推送系统
WO2017024506A1 (zh) 播放影视前插入广告时的信息提示方法和广告推送系统
WO2017024509A1 (zh) 广告推送方法和广告推送系统
WO2018023665A1 (zh) 视频推荐技术的数据采集方法和视频推荐系统
WO2023019927A1 (zh) 一种人脸识别方法、装置、存储介质及电子设备
WO2017128057A1 (zh) 视频推送技术的技术使用数据发送方法以及推送系统
WO2017024505A1 (zh) 基于虹膜的广告推送方法和广告推送系统
WO2018023753A1 (zh) 面部识别技术匹配电脑解锁时的数据采集方法和识别系统
WO2017128425A1 (zh) 根据浏览记录推送广告时的信息提示方法以及推送系统
WO2017128406A1 (zh) 广告推送技术的数据采集方法以及推送系统
WO2017128424A1 (zh) 推送广告技术的数据采集方法以及推送系统
WO2017024508A1 (zh) 播放影视前插入广告时的信息提示方法和广告推送系统
WO2018023754A1 (zh) 防止误操作电脑时的权属信息展示方法和识别系统
CN205665711U (zh) 人脸情绪识别系统
WO2018023453A1 (zh) 一种自动鉴黄时的专利信息推送方法和识别系统
WO2017028190A1 (zh) 根据面部特征匹配风扇状态时的信息推送方法和风扇
WO2018023450A1 (zh) 推送服装信息时的权属信息推送方法和推送系统
WO2018027774A1 (zh) 广告推送技术的权属信息推送方法和推送系统
CN209373681U (zh) 一种具有指纹识别功能的红外图像人脸门禁机
WO2018018190A1 (zh) 一种根据地域推荐节目的数据采集方法以及节目播放系统
WO2018023755A1 (zh) 根据面部识别防止小孩误操作电脑的方法和识别系统
WO2017024492A1 (zh) 面部特征匹配出水温度时信息提醒的方法以及饮水机
WO2017124264A1 (zh) 推荐书籍时的信息采集方法以及图书推荐系统
WO2018023763A1 (zh) 一种手机拍照的技术数据采集方法以及手机

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15900705

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15900705

Country of ref document: EP

Kind code of ref document: A1