WO2018035830A1 - 一种广告视频文件分级查找方法 - Google Patents

一种广告视频文件分级查找方法 Download PDF

Info

Publication number
WO2018035830A1
WO2018035830A1 PCT/CN2016/096803 CN2016096803W WO2018035830A1 WO 2018035830 A1 WO2018035830 A1 WO 2018035830A1 CN 2016096803 W CN2016096803 W CN 2016096803W WO 2018035830 A1 WO2018035830 A1 WO 2018035830A1
Authority
WO
WIPO (PCT)
Prior art keywords
video file
advertisement video
current scene
description data
camera
Prior art date
Application number
PCT/CN2016/096803
Other languages
English (en)
French (fr)
Inventor
武克易
Original Assignee
武克易
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 武克易 filed Critical 武克易
Priority to PCT/CN2016/096803 priority Critical patent/WO2018035830A1/zh
Publication of WO2018035830A1 publication Critical patent/WO2018035830A1/zh

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/25Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
    • H04N21/266Channel or content management, e.g. generation and management of keys and entitlement messages in a conditional access system, merging a VOD unicast channel into a multicast channel
    • H04N21/2668Creating a channel for a dedicated end-user group, e.g. insertion of targeted commercials based on end-user profiles
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor

Definitions

  • the present invention relates to the field of software technologies, and in particular, to a method for hierarchically searching for advertisement video files.
  • the existing advertisement playing mode it is generally limited to playing or looping some advertisement videos, and does not adaptively change according to changes in the surrounding environment, resulting in a waste of many advertisement playing resources, and does not play a good advertising effect, failing to reach Expected economic benefits.
  • the technical problem to be solved by the present invention is to provide a method for hierarchically searching for an advertisement video file for the above-mentioned drawbacks of the prior art.
  • the method for hierarchically searching for an advertisement video file wherein the step of: acquiring one or more feature description data of the target user, specifically includes:
  • the sensor automatically senses whether there is a target user before the advertising machine
  • the camera When there is a target user, the camera is started to work, and the current scene image including the target user is acquired.
  • the method for categorizing the advertisement video file according to the present invention wherein the feature description data specifically includes: user height, age, gender, weight, and skin color.
  • the advertisement video file hierarchical search method of the present invention wherein the advertisement video in the advertisement video file database is stored according to different feature description data.
  • the advertisement video file hierarchical searching method of the present invention wherein the advertisement video files corresponding to the same feature description data are stored in a folder.
  • the method for hierarchically searching for an advertisement video file wherein the method further comprises the steps of:
  • the feature description data is arranged in order of priority, and suitable advertisement videos are sequentially searched from the advertisement video library according to the priority order.
  • the method for hierarchically searching for an advertisement video file wherein, in the priority order, user gender and age are juxtaposed as feature description data of the first priority.
  • the method for hierarchically searching for an advertisement video file wherein the method further comprises the steps of:
  • the current scene image is merged with the 3D model and the advertisement video file to output and display.
  • the advertisement video file hierarchical searching method of the present invention wherein the step: obtaining the camera
  • the current scene image is taken for 3D rendering to obtain a 3D model, which specifically includes:
  • a rectangular planar 3D model capable of completely covering the target image in the current scene image is drawn.
  • the advertisement video file hierarchical searching method of the present invention wherein the steps are as follows:
  • a homography matrix which specifically includes:
  • the initial homography matrix and the current scene image captured by the camera in real time the optical flow tracking is performed on the target image, and the homography matrix is obtained by the tracking algorithm.
  • the invention has the beneficial effects that: by acquiring and analyzing the current scene image, the specific information of the target user that needs to be placed in the advertisement is determined in real time, so as to achieve accurate advertisement delivery.
  • FIG. 1 is a flow chart of a method for hierarchically searching for an advertisement video file according to a preferred embodiment of the present invention.
  • the flow of the method for locating the advertisement video file according to the preferred embodiment of the present invention is as shown in FIG. 1 and includes the following steps:
  • Step S10 Obtain one or more feature description data of the target user.
  • Step S20 searching for an advertisement video file database according to one of the feature description data, to obtain a first sub-database
  • Step S30 searching for an advertisement video file in the first sub-database according to the second feature description data, to obtain a second sub-database
  • Step S40 Playing an advertisement video file in the second sub-database.
  • the feature description data in each step includes: user height, age, gender, weight, and skin color.
  • each feature description data is a feature of the target user.
  • multiple target users multiple target users are simultaneously detected, and respective feature description data are obtained, and according to respective target users.
  • Different types of feature description data generate charts, and take the features with the same type of feature description data.
  • the above advertisement video file hierarchical search method further includes the steps of:
  • the current scene image is merged with the 3D model and the advertisement video file to output and display.
  • the above steps can be used to play 3D advertising video and place the target user in the advertising video.
  • it gives you an immersive feeling and a better experience of advertising.
  • the advertisement video in the advertisement video library is stored according to different feature description data. Specifically, the storage is performed according to different types of feature description data, for example, the advertisement video files for the same gender are stored in one folder, the advertisement video files for the same age are stored in one folder, and the like.
  • the above method further comprises the steps of: arranging the feature description data in order of priority, and sequentially searching for an appropriate advertisement video from the advertisement video library according to the priority order. For example, first searching for a folder corresponding to the feature data of the first priority, and obtaining a first group of advertisement videos, and searching for the feature data of the second priority in the current first group of advertisement video files, and obtaining the first
  • the two sets of advertisement videos, in the current second group of advertisement video files, and so on can find a plurality of feature description data thereof, and obtain a list of advertisement video files that can be finally played, and play them in sequence.
  • the user gender and the age are listed as the feature data of the first priority.
  • a rectangular planar 3D model capable of completely covering the target image in the current scene image is drawn.
  • a homography matrix which specifically includes:
  • the initial homography matrix and the current scene image captured by the camera in real time the optical flow tracking is performed on the target image, and the homography matrix is obtained by the tracking algorithm.
  • a rectangular planar 3D model capable of completely covering the target image in the current scene image is drawn.

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种广告视频文件分级查找方法,包括以下步骤:获取目标用户的一个或多个特征描述数据(S10);按照其中一个特征描述数据查找广告视频文件数据库,得到第一子数据库(S20);按照第二个特征描述数据,在所述第一子数据库中查找广告视频文件,得到第二子数据库(S30);播放所述第二子数据库中的广告视频文件(S40)。该方法通过对当前场景图像的获取及分析,实时判断当前需要投放广告的目标用户具体信息,以实现广告精准投放。

Description

一种广告视频文件分级查找方法 技术领域
本发明涉及软件技术领域,更具体地说,涉及一种广告视频文件分级查找方法。
背景技术
现有的广告播放方式中,一般仅限于固定或循环播放一些广告视频,而不会根据周围环境变化自适应调换,导致很多广告播放资源浪费,没有起到很好的广告投放效果,达不到预期的经济收益。
发明内容
本发明要解决的技术问题在于,针对现有技术的上述缺陷,提供一种广告视频文件分级查找方法。
本发明解决其技术问题所采用的技术方案是:
构造一种广告视频文件分级查找方法,其中,包括以下步骤:
获取目标用户的一个或多个特征描述数据;
按照其中一个特征描述数据查找广告视频文件数据库,得到第一子数据库;
按照第二个特征描述数据,在所述第一子数据库中查找广告视频文件,得到第二子数据库;
播放所述第二子数据库中的广告视频文件。
本发明所述的广告视频文件分级查找方法,其中,所述步骤:获取目标用户的一个或多个特征描述数据,具体包括:
传感器自动感应广告机前是否有目标用户;
在有目标用户时,启动所述摄像机工作,获取包含目标用户在内的当前场景图像。
本发明所述的广告视频文件分级查找方法,其中,所述特征描述数据具体包括:用户身高、年龄、性别、体重、肤色。
本发明所述的广告视频文件分级查找方法,其中,所述广告视频文件数据库中的广告视频按不同特征描述数据存储。
本发明所述的广告视频文件分级查找方法,其中,同一特征描述数据对应的广告视频文件存储在一个文件夹中。
本发明所述的广告视频文件分级查找方法,其中,所述方法进一步包括步骤:
将所述特征描述数据按优先级顺序排列,根据所述优先级顺序依次从所述广告视频库中查找合适的广告视频。
本发明所述的广告视频文件分级查找方法,其中,所述优先级顺序中,用户性别和年龄并列为第一优先级的特征描述数据。
本发明所述的广告视频文件分级查找方法,其中,所述方法进一步包括步骤:
获取包含目标用户在内的当前场景图像;
对摄像机获取的当前场景图像进行3D渲染,得到3D模型;
将当前场景图像与3D模型、广告视频文件融合输出显示。
本发明所述的广告视频文件分级查找方法,其中,所述步骤:对摄像机获 取的当前场景图像进行3D渲染,得到3D模型,具体包括:
对摄像机捕获的当前场景图像和样本图像进行配准,计算摄像机的姿态,得到单应性矩阵;
根据单应性矩阵,绘制出能将当前场景图像中的目标图片完全覆盖的矩形平面3D模型。
本发明所述的广告视频文件分级查找方法,其中,所述步骤:
对摄像机捕获的当前场景图像和样本图像进行配准,计算摄像机的姿态,得到单应性矩阵,具体包括:
对摄像机捕获的当前场景图像和样本图像进行配准,得到初始单应性矩阵;
根据图像配准成功的图像特征点、初始单应性矩阵及摄像机实时捕获的当前场景图像,对目标图片进行光流跟踪,由跟踪算法得到单应性矩阵。
本发明的有益效果在于:通过对当前场景图像的获取及分析,实时判断当前需要投放广告的目标用户具体信息,以实现广告精准投放。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将结合附图及实施例对本发明作进一步说明,下面描述中的附图仅仅是本发明的部分实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他附图:
图1是本发明较佳实施例的广告视频文件分级查找方法流程。
具体实施方式
为了使本发明实施例的目的、技术方案和优点更加清楚,下面将结合本发明实施例中的技术方案进行清楚、完整的描述,显然,所描述的实施例是本发明的部分实施例,而不是全部实施例。基于本发明的实施例,本领域普通技术人员在没有付出创造性劳动的前提下所获得的所有其他实施例,都属于本发明的保护范围。
本发明较佳实施例的广告视频文件分级查找方法流程如图1所示,包括以下步骤:
步骤S10、获取目标用户的一个或多个特征描述数据;
步骤S20、按照其中一个特征描述数据查找广告视频文件数据库,得到第一子数据库;
步骤S30、按照第二个特征描述数据,在第一子数据库中查找广告视频文件,得到第二子数据库;
步骤S40、播放第二子数据库中的广告视频文件。
具体地,上述各步骤中特征描述数据具体包括:用户身高、年龄、性别、体重、肤色。当只有一个目标用户时,上述各特征描述数据为该目标用户的特征,当有多个目标用户时,同时检测多个目标用户,并得到各自的特征描述数据,并根据多个目标用户各自的不同类型特征描述数据生成图表,取其中同类型特征描述数据最多的特征。
上述广告视频文件分级查找方法进一步包括步骤:
获取包含目标用户在内的当前场景图像;
对摄像机获取的当前场景图像进行3D渲染,得到3D模型;
将当前场景图像与3D模型、广告视频文件融合输出显示。
采用上述步骤可以实现3D广告视频播放,并将目标用户置于该广告视频 文件中,使其有身临其境的感觉,更好的体验广告效果。
上述广告视频文件分级查找方法中,广告视频库中的广告视频按不同特征描述数据进行存储。具体地,是按照不同类型的特征描述数据进行存储,例如,将针对同一性别的广告视频文件存储在一个文件夹中,将针对同一年龄的广告视频文件存储在一个文件夹中,等等。
上述方法进一步包括步骤:将特征描述数据按优先级顺序排列,根据优先级顺序依次从广告视频库中查找合适的广告视频。例如,先查找第一优先级的特征描述数据对应的文件夹,得到第一组广告视频,在当前的第一组广告视频文件中,再按第二优先级的特征描述数据进行查找,得到第二组广告视频,在当前的第二组广告视频文件中,依此类推,可以查找其中多个特征描述数据,得到最终可以播放的广告视频文件列表,依次进行播放。
优选地,上述优先级顺序中,用户性别和年龄并列为第一优先级的特征描述数据。
上述步骤:对摄像机获取的当前场景图像进行3D渲染,得到3D模型,具体包括:
对摄像机捕获的当前场景图像和样本图像进行配准,计算摄像机的姿态,得到单应性矩阵;
根据单应性矩阵,绘制出能将当前场景图像中的目标图片完全覆盖的矩形平面3D模型。
上述广告视频文件分级查找方法中,对摄像机捕获的当前场景图像和样本图像进行配准,计算摄像机的姿态,得到单应性矩阵,具体包括:
对摄像机捕获的当前场景图像和样本图像进行配准,得到初始单应性矩阵;
根据图像配准成功的图像特征点、初始单应性矩阵及摄像机实时捕获的当前场景图像,对目标图片进行光流跟踪,由跟踪算法得到单应性矩阵。
上述步骤:对摄像机获取的当前场景图像进行3D渲染,得到3D模型,具体包括:
对摄像机捕获的当前场景图像和样本图像进行配准,计算摄像机的姿态,得到单应性矩阵;
根据单应性矩阵,绘制出能将当前场景图像中的目标图片完全覆盖的矩形平面3D模型。
应当理解的是,对本领域普通技术人员来说,可以根据上述说明加以改进或变换,而所有这些改进和变换都应属于本发明所附权利要求的保护范围。

Claims (10)

  1. 一种广告视频文件分级查找方法,其特征在于,包括以下步骤:
    获取目标用户的一个或多个特征描述数据;
    按照其中一个特征描述数据查找广告视频文件数据库,得到第一子数据库;
    按照第二个特征描述数据,在所述第一子数据库中查找广告视频文件,得到第二子数据库;
    播放所述第二子数据库中的广告视频文件。
  2. 根据权利要求1所述的广告视频文件分级查找方法,其特征在于,所述步骤:获取目标用户的一个或多个特征描述数据,具体包括:
    传感器自动感应广告机前是否有目标用户;
    在有目标用户时,启动所述摄像机工作,获取包含目标用户在内的当前场景图像。
  3. 根据权利要求1所述的广告视频文件分级查找方法,其特征在于,所述特征描述数据具体包括:用户身高、年龄、性别、体重、肤色。
  4. 根据权利要求1所述的广告视频文件分级查找方法,其特征在于,所述广告视频文件数据库中的广告视频按不同特征描述数据存储。
  5. 根据权利要求4所述的广告视频文件分级查找方法,其特征在于,同一特征描述数据对应的广告视频文件存储在一个文件夹中。
  6. 根据权利要求3所述的广告视频文件分级查找方法,其特征在于,所述方法进一步包括步骤:
    将所述特征描述数据按优先级顺序排列,根据所述优先级顺序依次从所述 广告视频库中查找合适的广告视频。
  7. 根据权利要求6所述的广告视频文件分级查找方法,其特征在于,所述优先级顺序中,用户性别和年龄并列为第一优先级的特征描述数据。
  8. 根据权利要求4所述的广告视频文件分级查找方法,其特征在于,所述方法进一步包括步骤:
    获取包含目标用户在内的当前场景图像;
    对摄像机获取的当前场景图像进行3D渲染,得到3D模型;
    将当前场景图像与3D模型、广告视频文件融合输出显示。
  9. 根据权利要求8所述的广告视频文件分级查找方法,其特征在于,所述步骤:对摄像机获取的当前场景图像进行3D渲染,得到3D模型,具体包括:
    对摄像机捕获的当前场景图像和样本图像进行配准,计算摄像机的姿态,得到单应性矩阵;
    根据单应性矩阵,绘制出能将当前场景图像中的目标图片完全覆盖的矩形平面3D模型。
  10. 根据权利要求8所述的广告视频文件分级查找方法,其特征在于,所述步骤:
    对摄像机捕获的当前场景图像和样本图像进行配准,计算摄像机的姿态,得到单应性矩阵,具体包括:
    对摄像机捕获的当前场景图像和样本图像进行配准,得到初始单应性矩阵;
    根据图像配准成功的图像特征点、初始单应性矩阵及摄像机实时捕获的当前场景图像,对目标图片进行光流跟踪,由跟踪算法得到单应性矩阵。
PCT/CN2016/096803 2016-08-25 2016-08-25 一种广告视频文件分级查找方法 WO2018035830A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2016/096803 WO2018035830A1 (zh) 2016-08-25 2016-08-25 一种广告视频文件分级查找方法

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2016/096803 WO2018035830A1 (zh) 2016-08-25 2016-08-25 一种广告视频文件分级查找方法

Publications (1)

Publication Number Publication Date
WO2018035830A1 true WO2018035830A1 (zh) 2018-03-01

Family

ID=61245909

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/096803 WO2018035830A1 (zh) 2016-08-25 2016-08-25 一种广告视频文件分级查找方法

Country Status (1)

Country Link
WO (1) WO2018035830A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114612473A (zh) * 2022-05-11 2022-06-10 深圳联和智慧科技有限公司 一种基于无人机的桥梁巡检处理方法、系统及云平台

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102821323A (zh) * 2012-08-01 2012-12-12 成都理想境界科技有限公司 基于增强现实技术的视频播放方法、系统及移动终端
CN103514242A (zh) * 2012-12-19 2014-01-15 Tcl集团股份有限公司 一种电子广告牌的智能交互方法及系统
CN204119403U (zh) * 2014-10-09 2015-01-21 禾联硕股份有限公司 智能型的广告播放系统
CN104965893A (zh) * 2015-06-18 2015-10-07 山东师范大学 一种大数据广告投放方法
CN106303622A (zh) * 2016-08-25 2017-01-04 武克易 一种广告视频文件分级查找方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102821323A (zh) * 2012-08-01 2012-12-12 成都理想境界科技有限公司 基于增强现实技术的视频播放方法、系统及移动终端
CN103514242A (zh) * 2012-12-19 2014-01-15 Tcl集团股份有限公司 一种电子广告牌的智能交互方法及系统
CN204119403U (zh) * 2014-10-09 2015-01-21 禾联硕股份有限公司 智能型的广告播放系统
CN104965893A (zh) * 2015-06-18 2015-10-07 山东师范大学 一种大数据广告投放方法
CN106303622A (zh) * 2016-08-25 2017-01-04 武克易 一种广告视频文件分级查找方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114612473A (zh) * 2022-05-11 2022-06-10 深圳联和智慧科技有限公司 一种基于无人机的桥梁巡检处理方法、系统及云平台
CN114612473B (zh) * 2022-05-11 2022-08-16 深圳联和智慧科技有限公司 一种基于无人机的桥梁巡检处理方法、系统及云平台

Similar Documents

Publication Publication Date Title
US11657084B2 (en) Correlating image annotations with foreground features
Calli et al. Yale-CMU-Berkeley dataset for robotic manipulation research
CN103426003B (zh) 增强现实交互的实现方法和系统
WO2021012837A1 (zh) 推荐信息植入位置的确定方法、装置、设备及存储介质
US11551424B2 (en) Interaction analysis systems and methods
US20170065888A1 (en) Identifying And Extracting Video Game Highlights
CN110516096A (zh) 合成感知数字图像搜索
CN103686344A (zh) 增强视频系统及方法
Wong et al. Smartannotator an interactive tool for annotating indoor rgbd images
TW202141475A (zh) 物品名稱確定方法、裝置、電腦設備及儲存媒體
CN108230168B (zh) 通过定制多媒体方法共享个性化路线规划的方法和系统
CN105612511A (zh) 标识并结构化相关数据
JP6031210B1 (ja) 販売予測装置、販売予測方法、およびプログラム
US11817129B2 (en) 3D media elements in 2D video
Broadwell et al. Comparative K-Pop Choreography Analysis through Deep-Learning Pose Estimation across a Large Video Corpus.
US11062349B2 (en) Dynamic marketing asset generation based on user attributes and asset features
Styliaras et al. The MuseLearn platform: Personalized content for museum visitors assisted by vision-based recognition and 3D pose estimation of exhibits
WO2018035830A1 (zh) 一种广告视频文件分级查找方法
JP2012194691A (ja) 識別器の再学習方法、再学習のためのプログラム、及び画像認識装置
US20130100296A1 (en) Media content distribution
JP4995770B2 (ja) 画像辞書生成装置,画像辞書生成方法,および画像辞書生成プログラム
Wang et al. Action snapshot with single pose and viewpoint
CN109658172A (zh) 一种商圈推荐方法、计算设备、装置及存储介质
WO2018035832A1 (zh) 一种视频广告播放装置
WO2018035829A1 (zh) 一种广告播放装置

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: 16913852

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 23/07/2019)

122 Ep: pct application non-entry in european phase

Ref document number: 16913852

Country of ref document: EP

Kind code of ref document: A1