CN108108996A - Advertisement placement method, device, computer equipment and readable medium in video - Google Patents

Abstract

Translated fromChinese

本发明提供一种视频中的广告投放方法、装置、计算机设备及可读介质。其方法包括：接收到视频播放器发送的携带目标视频标识的广告请求之后，根据目标视频标识获取对应的目标视频的视频结构化信息；根据目标视频的视频结构化信息以及预先设置的广告库，生成目标视频对应的广告数组；向视频播放器发送广告数组，以供视频播放器根据广告数组投放各目标广告。本发明的技术方案可以实现智能化地为目标视频配置广告，同时还能够保证投放的广告与目标视频的关联性，从而能够有效地提高广告投放的效率。

The invention provides a method, device, computer equipment and readable medium for advertising in video. The method includes: after receiving the advertisement request carrying the target video identifier sent by the video player, obtaining the video structured information of the corresponding target video according to the target video identifier; according to the video structured information of the target video and the preset advertisement library, Generate an array of advertisements corresponding to the target video; send the array of advertisements to the video player, so that the video player can deliver each target advertisement according to the array of advertisements. The technical scheme of the present invention can intelligently configure advertisements for target videos, and at the same time, can ensure the relevance of the delivered advertisements and target videos, thereby effectively improving the efficiency of advertising delivery.

Description

Translated fromChinese

视频中的广告投放方法、装置、计算机设备及可读介质Advertising method, device, computer equipment and readable medium in video

【技术领域】【Technical field】

本发明涉及计算机应用技术领域，尤其涉及一种视频中的广告投放方法、装置、计算机设备及可读介质。The present invention relates to the field of computer application technology, in particular to a video advertising method, device, computer equipment and readable media.

【背景技术】【Background technique】

随着用户原创内容(User Generated Content；UGC)/专业生产内容(Professional Generated Content；PGC)之类的等短视频的爆发式增长，短视频播放平台承受着巨大的版权、带宽、存储方面的成本压力，视频流量方一般借助广告来变现。With the explosive growth of user-generated content (User Generated Content; UGC)/professional generated content (Professional Generated Content; PGC) and other short videos, short video playback platforms bear huge copyright, bandwidth, and storage costs. Under pressure, video traffic parties generally use advertisements to monetize.

但是，视频播放过程中投放广告不同于常见的网页浏览时的广告投放，网页浏览时的广告投放可直接提取网页内容作为关键字定向，而视频的信息蕴含在视频文件的视频流和音频流中。因此，如何在视频中实现广告的定向投放是一个非常重要的技术问题。现有技术中，通常采用人工标注的方式，提前在视频中埋好想要投放的广告，并指定广告在视频中的出现时间、隐藏时间，并指定具体的广告及其相关信息。这样，在播放该视频的过程中，在广告出现时间到来时，便可以播放该广告。However, the placement of advertisements during video playback is different from the common advertisement placement during web browsing. Ad placement during web browsing can directly extract the content of the webpage as a keyword for targeting, while the information of the video is contained in the video stream and audio stream of the video file. . Therefore, how to realize the targeted delivery of advertisements in videos is a very important technical issue. In the prior art, manual marking is usually used to bury the desired advertisement in the video in advance, and specify the time when the advertisement appears in the video, the hiding time, and specify the specific advertisement and its related information. In this way, during the playback of the video, the advertisement can be played when the time for the advertisement to appear arrives.

现有的视频中的广告投放方法，完全由人工来为视频配置相应的要投放的广告，导致视频中的广告投放效率较低。In the existing method for placing advertisements in videos, the corresponding advertisements to be placed are manually configured for the video, resulting in low efficiency in placing advertisements in videos.

【发明内容】【Content of invention】

本发明提供了一种视频中的广告投放方法、装置、计算机设备及可读介质，用于提高视频中的广告投放效率。The present invention provides a video advertisement delivery method, device, computer equipment and readable medium, which are used to improve the video advertisement delivery efficiency.

本发明提供一种视频中的广告投放方法，所述方法包括：The present invention provides a method for placing advertisements in a video, the method comprising:

接收到视频播放器发送的携带所述目标视频标识的广告请求之后，根据所述目标视频标识获取对应的目标视频的视频结构化信息；After receiving the advertisement request carrying the target video identifier sent by the video player, acquiring the video structured information of the corresponding target video according to the target video identifier;

根据所述目标视频的视频结构化信息以及预先设置的广告库，生成所述目标视频对应的广告数组；Generate an array of advertisements corresponding to the target video according to the video structured information of the target video and a preset advertisement library;

向所述视频播放器发送所述广告数组，以供所述视频播放器根据所述广告数组投放各所述目标广告。Sending the advertisement array to the video player for the video player to deliver each of the target advertisements according to the advertisement array.

进一步可选地，如上所述的方法中，根据所述视频结构化信息以及预先设置的广告库，生成所述目标视频对应的广告数组，具体包括：Further optionally, in the above method, generating an advertisement array corresponding to the target video according to the video structured information and a preset advertisement library, specifically includes:

根据所述视频结构化信息从所述广告库中获取与所述目标视频关联的至少一条目标广告的信息；Obtaining at least one target advertisement information associated with the target video from the advertisement library according to the video structured information;

根据所述视频结构化信息和对应的所述至少一条目标广告的信息，生成广告数组。An advertisement array is generated according to the video structural information and the corresponding information of the at least one target advertisement.

进一步可选地，如上所述的方法中，所述视频结构化信息包括所述目标视频的题材类型以及所述目标视频的长度；根据所述视频结构化信息从所述广告库中获取与所述目标视频关联的至少一条目标广告的信息，具体包括：Further optionally, in the above-mentioned method, the video structured information includes the subject type of the target video and the length of the target video; Information about at least one target advertisement associated with the target video, including:

根据所述题材类型，从所述广告库中获取与所述题材类型对应的多条第一候选广告的信息；According to the subject type, acquiring information of a plurality of first candidate advertisements corresponding to the subject type from the advertisement library;

根据所述目标视频的长度和预设的广告间隔时长，确定所述题材类型对应的第一目标广告的条数m；所述m为大于或者等于1的整数；According to the length of the target video and the preset advertisement interval duration, determine the number m of the first target advertisement corresponding to the theme type; the m is an integer greater than or equal to 1;

根据各所述第一候选广告的标价，从所述多条第一候选广告的信息中获取m条第一目标广告的信息；Obtaining information on m pieces of first target advertisements from the information on the multiple pieces of first candidate advertisements according to the bids of each of the first candidate advertisements;

对应地，根据所述视频结构化信息和对应的所述至少一条目标广告的信息，生成广告数组，具体包括：Correspondingly, according to the video structured information and the corresponding information of the at least one target advertisement, an advertisement array is generated, specifically including:

根据所述m条第一目标广告中的各所述第一目标广告的标价、所述目标视频的长度、所述广告间隔时长和所述第一目标广告的数量，确定各所述第一目标广告的投放方式以及投放时间；According to the price of each of the first target advertisements in the m first target advertisements, the length of the target video, the length of the advertisement interval and the number of the first target advertisements, determine each of the first target advertisements How and when advertisements will be delivered;

将各所述第一目标广告的标识、各所述第一目标广告的投放方式及所述投放时间以及各所述第一目标广告的地址，存入所述广告数组中。Store the identifier of each of the first target advertisements, the delivery method and the delivery time of each of the first target advertisements, and the address of each of the first target advertisements into the advertisement array.

进一步可选地，如上所述的方法中，所述视频化结构信息中还包括所述目标视频中出现频次最高的数个关键词；根据所述题材类型，从所述广告库中获取相关的多条第一候选广告的信息，具体包括：Further optionally, in the above-mentioned method, the video structure information also includes several keywords with the highest frequency of occurrence in the target video; according to the subject type, obtain relevant keywords from the advertisement library. Multiple pieces of first candidate advertisement information, specifically including:

根据所述题材类型，并结合所述数个关键词，从所述广告库中获取与所述题材类型对应、且与所述数个关键词相关的多条所述第一候选广告的信息。According to the theme type and in combination with the several keywords, information about a plurality of first candidate advertisements corresponding to the theme type and related to the several keywords is obtained from the advertisement library.

进一步可选地，如上所述的方法中，所述视频结构化信息包括所述目标视频中的人物标识以及所述人物标识对应的人物的出现时间；Further optionally, in the above-mentioned method, the video structured information includes a character identifier in the target video and an appearance time of a character corresponding to the character identifier;

根据所述视频结构化信息从所述广告库中获取与所述目标视频关联的至少一条目标广告的信息，具体包括：Obtaining at least one target advertisement information associated with the target video from the advertisement library according to the video structured information, specifically including:

根据所述人物标识，从所述广告库中获取与所述人物标识对应人物相关的多条第二候选广告的信息；Acquiring, from the advertisement library, information about a plurality of second candidate advertisements related to the person corresponding to the person identifier according to the person identifier;

根据所述人物的所述出现时间以及预设的广告投放时长，确定所述人物对应的第二目标广告的条数n；所述n为大于或者等于1的整数；Determine the number n of second target advertisements corresponding to the character according to the appearance time of the character and the preset advertisement delivery duration; the n is an integer greater than or equal to 1;

根据各所述第二候选广告的标价，从所述多条第二候选广告中获取n条第二目标广告的信息；Acquiring information about n second target advertisements from the plurality of second candidate advertisements according to the bids of each of the second candidate advertisements;

对应地，根据所述视频结构化信息和对应的所述至少一条目标广告的信息，生成广告数组，具体包括：Correspondingly, according to the video structured information and the corresponding information of the at least one target advertisement, an advertisement array is generated, specifically including:

根据所述n条第二目标广告中各所述第二目标广告的标价、所述人物的出现时间和预设的所述广告投放时长，确定各所述第二目标广告的投放时间；Determining the delivery time of each of the second target advertisements according to the price of each of the n second target advertisements, the appearance time of the characters, and the preset advertisement delivery time;

将各所述第二目标广告的标识、各所述第二目标广告的投放方式及所述投放时间以及各所述第二目标广告的地址，存入所述广告数组中。Store the identifier of each of the second target advertisements, the delivery method and the delivery time of each of the second target advertisements, and the address of each of the second target advertisements into the advertisement array.

进一步可选地，如上所述的方法中，所述视频结构化信息还包括所述目标视频中的所述人物的行为；Further optionally, in the above method, the video structured information also includes the behavior of the character in the target video;

根据所述人物标识，从所述广告库中获取所述人物标识对应人物相关的多条第二候选广告的信息，具体包括：According to the person identification, the information of a plurality of second candidate advertisements related to the person corresponding to the person identification is obtained from the advertisement database, specifically including:

根据所述人物标识，并结合所述人物的行为，从所述广告库中获取与所述人物以及所述人物的行为相关的多条所述第二候选广告的信息。Based on the character identification and in combination with the character's behavior, information about the multiple second candidate advertisements related to the character and the character's behavior is acquired from the advertisement library.

进一步可选地，如上所述的方法中，所述视频结构化信息包括所述目标视频中的物体标识以及所述物体标识对应的物体的出现时间；Further optionally, in the above method, the video structured information includes an object identifier in the target video and an appearance time of an object corresponding to the object identifier;

根据所述视频结构化信息从所述广告库中获取与所述目标视频关联的至少一条目标广告的信息，具体包括：Obtaining at least one target advertisement information associated with the target video from the advertisement library according to the video structured information, specifically including:

根据所述物体标识，从所述广告库中获取所述物体标识对应物体相关的多条第三候选广告的信息；According to the object identifier, acquiring information of multiple third candidate advertisements related to the object corresponding to the object identifier from the advertisement library;

根据所述物体的出现时间以及预设的广告投放时长，确定第三目标广告的条数p；所述p为大于或者等于1的整数；Determine the number p of the third target advertisement according to the appearance time of the object and the preset advertisement delivery duration; the p is an integer greater than or equal to 1;

根据各所述第三候选广告的标价，从所述多条第三候选广告中获取p条第三目标广告的信息；Acquiring information about p pieces of third target advertisements from the plurality of pieces of third candidate advertisements according to the price tag of each of the third candidate advertisements;

对应地，根据所述视频结构化信息和对应的所述至少一条目标广告的信息，生成广告数组，具体包括：Correspondingly, according to the video structured information and the corresponding information of the at least one target advertisement, an advertisement array is generated, specifically including:

根据所述p条第三目标广告中各所述第三目标广告的标价、所述物体的出现时间和预设的所述投放时长，确定各所述第三目标广告的投放时间；Determine the delivery time of each of the third target advertisements according to the price of each of the p third target advertisements, the appearance time of the object, and the preset delivery time;

将各所述第三目标广告的标识、各所述第三目标广告的投放方式及所述投放时间以及各所述第三目标广告的地址，存入所述广告数组中。Store the identifier of each of the third target advertisements, the delivery method and the delivery time of each of the third target advertisements, and the address of each of the third target advertisements into the advertisement array.

进一步可选地，如上所述的方法中，接收到视频播放器发送的携带所述目标视频标识的广告请求之后，根据所述目标视频标识获取对应的目标视频的视频结构化信息之前，所述方法还包括：Further optionally, in the above-mentioned method, after receiving the advertisement request carrying the target video identifier sent by the video player, before obtaining the video structured information of the corresponding target video according to the target video identifier, the Methods also include:

从所述目标视频的视频流中按照时间先后顺序提取多张缩略图；各张所述缩略图按照对应的时间戳来标识；Extracting a plurality of thumbnail images in chronological order from the video stream of the target video; each thumbnail image is identified according to a corresponding timestamp;

对所述多张缩略图进行题材类型识别，得到所述目标视频的题材类型；Carry out theme type identification to described multiple thumbnails, obtain the theme type of described target video;

对各张所述缩略图进行人物识别、行为识别和物体识别，得到各张所述缩略图中包括的人物的标识、所述人物的行为以及物体的标识；Performing character recognition, behavior recognition and object recognition on each of the thumbnail images to obtain the identity of the person included in each of the thumbnail images, the behavior of the person and the identity of the object;

根据各张所述缩略图中包括的人物的标识、所述人物的行为、物体的标识以及各张所述缩略图的时间戳，获取所述目标视频中的所述人物的出现时间、所述人物的行为的出现时间以及各所述物体的出现时间；Acquire the appearance time of the character in the target video, the the timing of the behavior of the characters and the timing of each of the objects described;

对所述多张缩略图进行文字识别，得到第一文本信息；performing text recognition on the plurality of thumbnail images to obtain first text information;

对所述目标视频的音频流进行语音识别，得到第二文本信息；performing speech recognition on the audio stream of the target video to obtain second text information;

从所述第一文本信息和所述第二文本信息中提取出现频次最高的数个关键词。Several keywords with the highest frequency of occurrence are extracted from the first text information and the second text information.

本发明提供一种视频中的广告投放装置，所述装置包括：The present invention provides a device for placing advertisements in videos, the device comprising:

获取模块，用于接收到视频播放器发送的携带所述目标视频标识的广告请求之后，根据所述目标视频标识获取对应的目标视频的视频结构化信息；An acquisition module, configured to acquire the video structured information of the corresponding target video according to the target video ID after receiving the advertisement request sent by the video player and carrying the target video ID;

生成模块，用于根据所述目标视频的视频结构化信息以及预先设置的广告库，生成所述目标视频对应的广告数组；A generating module, configured to generate an array of advertisements corresponding to the target video according to the video structured information of the target video and a preset advertisement library;

发送模块，用于向所述视频播放器发送所述广告数组，以供所述视频播放器根据所述广告数组投放各所述目标广告。A sending module, configured to send the advertisement array to the video player, so that the video player can deliver each of the target advertisements according to the advertisement array.

进一步可选地，如上所述的装置中，所述生成模块，具体包括：Further optionally, in the device as described above, the generating module specifically includes:

广告获取单元，用于根据所述视频结构化信息从所述广告库中获取与所述目标视频关联的至少一条目标广告的信息；an advertisement obtaining unit, configured to obtain information of at least one target advertisement associated with the target video from the advertisement library according to the video structured information;

数组生成单元，用于根据所述视频结构化信息和对应的所述至少一条目标广告的信息，生成广告数组。An array generating unit, configured to generate an array of advertisements according to the video structured information and the corresponding information of the at least one target advertisement.

进一步可选地，如上所述的装置中，所述视频结构化信息包括所述目标视频的题材类型以及所述目标视频的长度；所述广告获取单元，具体用于：Further optionally, in the above-mentioned device, the video structured information includes the subject type of the target video and the length of the target video; the advertisement acquisition unit is specifically configured to:

根据所述题材类型，从所述广告库中获取与所述题材类型对应的多条第一候选广告的信息；According to the subject type, acquiring information of a plurality of first candidate advertisements corresponding to the subject type from the advertisement library;

根据所述目标视频的长度和预设的广告间隔时长，确定所述题材类型对应的第一目标广告的条数m；所述m为大于或者等于1的整数；According to the length of the target video and the preset advertisement interval duration, determine the number m of the first target advertisement corresponding to the theme type; the m is an integer greater than or equal to 1;

根据各所述第一候选广告的标价，从所述多条第一候选广告的信息中获取m条第一目标广告的信息；Obtaining information on m pieces of first target advertisements from the information on the multiple pieces of first candidate advertisements according to the bids of each of the first candidate advertisements;

对应地，所述数组生成单元，具体用于：Correspondingly, the array generating unit is specifically used for:

根据所述m条第一目标广告中的各所述第一目标广告的标价、所述目标视频的长度、所述广告间隔时长和所述第一目标广告的数量，确定各所述第一目标广告的投放方式以及投放时间；According to the price of each of the first target advertisements in the m first target advertisements, the length of the target video, the length of the advertisement interval and the number of the first target advertisements, determine each of the first target advertisements How and when advertisements will be delivered;

将各所述第一目标广告的标识、各所述第一目标广告的投放方式及所述投放时间以及各所述第一目标广告的地址，存入所述广告数组中。Store the identifier of each of the first target advertisements, the delivery method and the delivery time of each of the first target advertisements, and the address of each of the first target advertisements into the advertisement array.

进一步可选地，如上所述的装置中，所述视频化结构信息中还包括所述目标视频中出现频次最高的数个关键词；所述广告获取单元，具体用于：根据所述题材类型，并结合所述数个关键词，从所述广告库中获取与所述题材类型对应、且与所述数个关键词相关的多条所述第一候选广告的信息。Further optionally, in the above-mentioned device, the video structure information also includes several keywords with the highest frequency of occurrence in the target video; the advertisement acquisition unit is specifically configured to: according to the subject type , and combined with the several keywords, acquiring a plurality of pieces of information about the first candidate advertisements corresponding to the theme type and related to the several keywords from the advertisement library.

进一步可选地，如上所述的装置中，所述视频结构化信息包括所述目标视频中的人物标识以及所述人物标识对应的人物的出现时间；Further optionally, in the above-mentioned device, the video structured information includes a character identifier in the target video and an appearance time of a character corresponding to the character identifier;

所述广告获取单元，具体用于：The advertisement acquiring unit is specifically used for:

根据所述人物标识，从所述广告库中获取与所述人物标识对应人物相关的多条第二候选广告的信息；Acquiring, from the advertisement library, information about a plurality of second candidate advertisements related to the person corresponding to the person identifier according to the person identifier;

根据所述人物的所述出现时间以及预设的广告投放时长，确定所述人物对应的第二目标广告的条数n；所述n为大于或者等于1的整数；Determine the number n of second target advertisements corresponding to the character according to the appearance time of the character and the preset advertisement delivery duration; the n is an integer greater than or equal to 1;

根据各所述第二候选广告的标价，从所述多条第二候选广告中获取n条第二目标广告的信息；Acquiring information about n second target advertisements from the plurality of second candidate advertisements according to the bids of each of the second candidate advertisements;

对应地，所述数组生成单元，具体用于：Correspondingly, the array generating unit is specifically used for:

根据所述n条第二目标广告中各所述第二目标广告的标价、所述人物的出现时间和预设的所述广告投放时长，确定各所述第二目标广告的投放时间；Determining the delivery time of each of the second target advertisements according to the price of each of the n second target advertisements, the appearance time of the characters, and the preset advertisement delivery time;

将各所述第二目标广告的标识、各所述第二目标广告的投放方式及所述投放时间以及各所述第二目标广告的地址，存入所述广告数组中。Store the identifier of each of the second target advertisements, the delivery method and the delivery time of each of the second target advertisements, and the address of each of the second target advertisements into the advertisement array.

进一步可选地，如上所述的装置中，所述视频结构化信息还包括所述目标视频中的所述人物的行为；Further optionally, in the above-mentioned device, the video structured information further includes the behavior of the character in the target video;

所述广告获取单元，具体用于：The advertisement acquiring unit is specifically used for:

根据所述人物标识，并结合所述人物的行为，从所述广告库中获取与所述人物以及所述人物的行为相关的多条所述第二候选广告的信息。Based on the character identification and in combination with the character's behavior, information about the multiple second candidate advertisements related to the character and the character's behavior is acquired from the advertisement library.

进一步可选地，如上所述的装置中，所述视频结构化信息包括所述目标视频中的物体标识以及所述物体标识对应的物体的出现时间；Further optionally, in the above-mentioned device, the video structured information includes an object identifier in the target video and an appearance time of an object corresponding to the object identifier;

所述广告获取单元，具体用于：The advertisement acquiring unit is specifically used for:

根据所述物体标识，从所述广告库中获取所述物体标识对应物体相关的多条第三候选广告的信息；According to the object identifier, acquiring information of multiple third candidate advertisements related to the object corresponding to the object identifier from the advertisement library;

根据所述物体的出现时间以及预设的广告投放时长，确定第三目标广告的条数p；所述p为大于或者等于1的整数；Determine the number p of the third target advertisement according to the appearance time of the object and the preset advertisement delivery duration; the p is an integer greater than or equal to 1;

根据各所述第三候选广告的标价，从所述多条第三候选广告中获取p条第三目标广告的信息；Acquiring information about p pieces of third target advertisements from the plurality of pieces of third candidate advertisements according to the price tag of each of the third candidate advertisements;

对应地，所述数组生成单元，具体用于：Correspondingly, the array generating unit is specifically used for:

根据所述p条第三目标广告中各所述第三目标广告的标价、所述物体的出现时间和预设的所述投放时长，确定各所述第三目标广告的投放时间；Determine the delivery time of each of the third target advertisements according to the price of each of the p third target advertisements, the appearance time of the object, and the preset delivery time;

将各所述第三目标广告的标识、各所述第三目标广告的投放方式及所述投放时间以及各所述第三目标广告的地址，存入所述广告数组中。Store the identifier of each of the third target advertisements, the delivery method and the delivery time of each of the third target advertisements, and the address of each of the third target advertisements into the advertisement array.

进一步可选地，如上所述的装置中，所述装置还包括：Further optionally, in the device as described above, the device further includes:

提取模块，用于从所述目标视频的视频流中按照时间先后顺序提取多张缩略图；各张所述缩略图按照对应的时间戳来标识；An extracting module, configured to extract a plurality of thumbnails in chronological order from the video stream of the target video; each thumbnail is identified according to a corresponding timestamp;

识别模块，用于对所述多张缩略图进行题材类型识别，得到所述目标视频的题材类型；The identification module is used to identify the theme type of the plurality of thumbnails to obtain the theme type of the target video;

所述识别模块，还用于对各张所述缩略图进行人物识别、行为识别和物体识别，得到各张所述缩略图中包括的人物的标识、所述人物的行为以及物体的标识；The recognition module is also used to perform character recognition, behavior recognition, and object recognition on each of the thumbnail images to obtain the identity of the person included in each of the thumbnails, the behavior of the character, and the identity of the object;

所述获取模块，还用于根据各张所述缩略图中包括的人物的标识、所述人物的行为、物体的标识以及各张所述缩略图的时间戳，获取所述目标视频中的所述人物的出现时间、所述人物的行为的出现时间以及各所述物体的出现时间；The acquisition module is further configured to acquire all the characters in the target video according to the identification of the characters included in each of the thumbnails, the behavior of the characters, the identification of the object, and the time stamp of each of the thumbnails. the time of appearance of said character, the time of appearance of the behavior of said character, and the time of appearance of each said object;

所述识别模块，还用于对所述多张缩略图进行文字识别，得到第一文本信息；The recognition module is also used to perform character recognition on the plurality of thumbnail images to obtain first text information;

所述识别模块，还用于对所述目标视频的音频流进行语音识别，得到第二文本信息；The recognition module is further configured to perform speech recognition on the audio stream of the target video to obtain second text information;

所述提取模块，还用于从所述第一文本信息和所述第二文本信息中提取出现频次最高的数个关键词。The extraction module is further configured to extract several keywords with the highest frequency of occurrence from the first text information and the second text information.

本发明还提供一种计算机设备，所述设备包括：The present invention also provides a computer device, the device comprising:

一个或多个处理器；one or more processors;

存储器，用于存储一个或多个程序；memory for storing one or more programs;

当所述一个或多个程序被所述一个或多个处理器执行，使得所述一个或多个处理器实现如上所述的视频中的广告投放方法。When the one or more programs are executed by the one or more processors, the one or more processors are made to implement the above-mentioned method for placing advertisements in videos.

本发明还提供一种计算机可读介质，其上存储有计算机程序，该程序被处理器执行时实现如上所述的视频中的广告投放方法。The present invention also provides a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, the above-mentioned method for placing advertisements in a video is realized.

本发明的视频中的广告投放方法、装置、计算机设备及可读介质，通过接收到视频播放器发送的携带目标视频标识的广告请求之后，根据目标视频标识获取对应的目标视频的视频结构化信息；根据目标视频的视频结构化信息以及预先设置的广告库，生成目标视频对应的广告数组；向视频播放器发送广告数组，以供视频播放器根据广告数组投放各目标广告，可以实现智能化地为目标视频配置广告，同时还能够保证投放的广告与目标视频的关联性，从而能够有效地提高广告投放的效率。According to the method, device, computer equipment and readable medium for placing advertisements in videos of the present invention, after receiving the advertisement request carrying the target video identifier sent by the video player, the video structured information of the corresponding target video is obtained according to the target video identifier ; According to the video structural information of the target video and the pre-set advertisement library, generate the advertisement array corresponding to the target video; send the advertisement array to the video player, so that the video player can deliver each target advertisement according to the advertisement array, which can realize intelligent The advertisement is configured for the target video, and at the same time, the relevance of the advertisement and the target video can be guaranteed, so that the efficiency of advertisement delivery can be effectively improved.

【附图说明】【Description of drawings】

图1为本发明的视频中的广告投放方法实施例的流程图。FIG. 1 is a flow chart of an embodiment of a method for placing advertisements in a video according to the present invention.

图2为本发明的视频中的广告投放装置实施例一的结构图。FIG. 2 is a structural diagram of Embodiment 1 of the device for placing advertisements in videos according to the present invention.

图3为本发明的视频中的广告投放装置实施例二的结构图。FIG. 3 is a structural diagram of Embodiment 2 of the device for placing advertisements in videos according to the present invention.

图4为本发明的计算机设备实施例的结构图。FIG. 4 is a structural diagram of an embodiment of a computer device of the present invention.

图5为本发明提供的一种计算机设备的示例图。Fig. 5 is an example diagram of a computer device provided by the present invention.

【具体实施方式】【Detailed ways】

为了使本发明的目的、技术方案和优点更加清楚，下面结合附图和具体实施例对本发明进行详细描述。In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

图1为本发明的视频中的广告投放方法实施例的流程图。如图1所示，本实施例的视频中的广告投放方法，具体可以包括如下步骤：FIG. 1 is a flow chart of an embodiment of a method for placing advertisements in a video according to the present invention. As shown in Figure 1, the advertisement placement method in the video of this embodiment may specifically include the following steps:

100、接收到视频播放器发送的携带目标视频标识的广告请求之后，根据目标视频标识获取对应的目标视频的视频结构化信息；100. After receiving the advertisement request carrying the target video identifier sent by the video player, acquire the video structured information of the corresponding target video according to the target video identifier;

本实施例的视频中的广告投放方法的执行主体可以为视频中的广告投放装置，该视频中的广告投放装置可以为视频提供商的广告投放装置，便于该视频提供商的视频服务器在向视频播放器播放视频的同时投放广告。该广告投放装置可以为独立于视频服务器的一个独立的实体，或者也可以直接为视频服务器，由视频服务器在向视频播放器播放视频的同时投放广告。The execution subject of the advertisement placement method in the video of this embodiment may be the advertisement placement device in the video, and the advertisement placement device in the video may be the advertisement placement device of the video provider, so that the video server of the video provider can send the video to the video The ad is served while the player is playing the video. The advertisement delivery device may be an independent entity independent from the video server, or may be a video server directly, and the video server will deliver advertisements while playing the video to the video player.

具体地，用户可以向视频播放器请求播放某个目标视频，对应地，视频播放器接收到用户请求播放该目标视频时，向视频服务器请求该目标视频的资源，同时还会发起广告请求，以请求在该目标视频中的广告，以便于在目标视频播放过程中播放该目标视频中投放的广告。本实施例中，在广告投放装置或者视频服务器在接收到接收视频播放器发送的携带目标视频标识的广告请求之后，根据目标视频标识获取对应的目标视频的视频结构化信息。Specifically, the user can request the video player to play a certain target video. Correspondingly, when the video player receives the user's request to play the target video, it will request the resource of the target video from the video server and at the same time initiate an advertisement request to Request an advertisement in the target video, so that the advertisement served in the target video can be played during the playback of the target video. In this embodiment, after the advertisement delivery device or the video server receives the advertisement request carrying the target video identifier sent by the receiving video player, it acquires the video structured information of the corresponding target video according to the target video identifier.

本实施例的目标视频的视频结构化信息由视频服务器预先对该目标视频进行结构化分析得到的，该目标视频的视频结构化信息可以包括该目标视频的基本信息如长度、视频来源、视频名称等等信息，还包括结构化分析的信息如目标视频中的频次最高的关键词、目标视频的题材类型、目标视频中包括的人物标识以及对应的人物的出现时间、目标视频中的人物的行为、目标视频中的物体标识以及对应的物体的出现时间等等。例如，目标视频的题材类型可以为动漫题材、军事题材或者其他题材等等。目标视频中的人物标识可以为一些明星的名字或者一些文坛或者体育界等为广大公众能够熟知的公众人物的名字。目标视频中的人物的行为可以为人物在慢跑、遛狗、骑车或者瑜伽等等行为。The video structured information of the target video in this embodiment is obtained by the video server in advance through structural analysis of the target video, and the video structured information of the target video may include the basic information of the target video such as length, video source, video name Information such as structural analysis, such as the most frequent keywords in the target video, the subject type of the target video, the identification of the characters included in the target video, the time of appearance of the corresponding characters, and the behavior of the characters in the target video , the object identification in the target video, the appearance time of the corresponding object, and the like. For example, the theme type of the target video may be anime theme, military theme, or other theme. The character identification in the target video may be the names of some celebrities or the names of some public figures familiar to the general public, such as those in the literary or sports circles. The behavior of the character in the target video may be the character is jogging, walking the dog, cycling, or doing yoga.

若执行主体为独立于视频服务器的广告投放装置，对应地，可以从视频服务器中获取目标视频的视频结构化信息。而若执行主体为视频服务器，对应地，可以直接从预先分析得到的结果中获取目标视频的视频结构化信息。If the execution subject is an advertisement delivery device independent of the video server, correspondingly, the video structured information of the target video may be obtained from the video server. However, if the execution subject is a video server, correspondingly, the video structural information of the target video can be obtained directly from the pre-analyzed results.

101、根据目标视频的视频结构化信息以及预先设置的广告库，生成目标视频对应的广告数组；101. Generate an advertisement array corresponding to the target video according to the video structural information of the target video and the preset advertisement library;

本实施例的预先设置的广告库为各个广告主预先投放的广告，广告主投放的广告的同时，还会记录广告的基础信息，例如广告内容的简要、广告内容的地址、广告的类型，广告的代言人以及广告投放时定向的视频题材类型，还可以包括广告的投放方式的限定等等。本实施例的目标视频的视频结构化信息可以记录目标视频的很多相关的结构信息。因此，根据目标视频的视频结构化信息可以从广告库中获取到相关的广告，从而生成该目标视频的广告数组。广告数组中可以包括投放广告所需要的全部信息，以便于视频播放器根据目标视频的广告数组，实现对该目标视频播放中的所有广告的投放。本实施例中，非人为地选择与目标视频关联的目标广告，而是智能地根据目标视频的视频结构化信息以及预先设置的广告库，生成目标视频对应的广告数组，可以提高广告投放的效率。The pre-set advertisement library in this embodiment is the advertisements delivered by each advertiser in advance, and the advertisements delivered by the advertisers will also record the basic information of the advertisement, such as the brief of the advertisement content, the address of the advertisement content, the type of the advertisement, the The spokesperson and the type of video subject matter targeted when the advertisement is placed, can also include restrictions on the way the advertisement is placed, and so on. The video structural information of the target video in this embodiment can record many relevant structural information of the target video. Therefore, relevant advertisements can be obtained from the advertisement library according to the video structural information of the target video, so as to generate an advertisement array of the target video. The advertisement array may include all the information required for serving advertisements, so that the video player can realize serving of all advertisements in the target video according to the advertisement array of the target video. In this embodiment, the target advertisement associated with the target video is not artificially selected, but the advertisement array corresponding to the target video is generated intelligently according to the video structural information of the target video and the preset advertisement library, which can improve the efficiency of advertisement delivery .

例如，本实施例的广告数组中，可以包括广告的标识、广告的地址、广告的投放时间以及广告的投放方式。其中广告的标识用于唯一标识广告库中的该广告。广告的地址为该广告的投放地址，以便于后续投放该广告时可以从该广告的地址中获取到该广告。广告的投放时间为根据视频结构化信息决定的该广告具体投放的时间。广告的投放方式在本实施例中可以存在如下三种情况：For example, the advertisement array in this embodiment may include an advertisement identifier, an advertisement address, an advertisement delivery time, and an advertisement delivery method. The advertisement identifier is used to uniquely identify the advertisement in the advertisement library. The address of the advertisement is the delivery address of the advertisement, so that the advertisement can be obtained from the address of the advertisement when the advertisement is delivered later. The advertisement delivery time is the specific delivery time of the advertisement determined according to the structured information of the video. Advertisement delivery methods may exist in the following three situations in this embodiment:

(A)、线性投放方式：广告可以作为视频的前贴片、中贴片或者后贴片的形式投放，广告投放时一般占据全屏，该广告出现时，视频暂停播放。广告结束后，继续播放影片。例如题材类型相关的广告可以采用线性广告的投放方式来投放。(A) Linear delivery method: the advertisement can be delivered as a pre-roll, mid-roll or post-roll of the video. When the advertisement is placed, it generally occupies the full screen. When the advertisement appears, the video will pause. After the ad ends, continue playing the video. For example, advertisements related to subject types can be delivered in the form of linear advertisements.

(B)、非线性投放方式：视频播放过程中，浮在视频画面的某个小区域的，浮动一段时间后消失；例如视频中的物体相关的广告、视频中人物相关的广告或者人物的行为相关的广告可以采用非线性投放方式来投放。(B) Non-linear delivery method: During the video playback process, it floats in a small area of the video screen and disappears after floating for a period of time; such as advertisements related to objects in the video, advertisements related to characters in the video, or the behavior of characters Relevant advertisements may be delivered using non-linear delivery.

(C)、伴随性投放方式：出现在视频中的非视频区域，小横幅广告居多；例如，视频中的物体相关的广告、视频中人物相关的广告或者人物的行为相关的广告也可以采用伴随广告的方式来投放。(C) Accompanying delivery method: appearing in the non-video area of the video, mostly small banner advertisements; for example, advertisements related to objects in the video, advertisements related to characters in the video, or advertisements related to the behavior of characters can also use accompanying advertisements way of advertising.

102、向视频播放器发送广告数组，以供视频播放器根据广告数组投放各目标广告。102. Send the advertisement array to the video player, so that the video player can deliver each target advertisement according to the advertisement array.

根据上述方式，获取到广告数组之后，向视频播放器发送广告数组。这样，在视频播放器获取到目标视频之后，播放视频的过程中可以根据广告数组，对广告数组中的广告进行投放。According to the above method, after obtaining the advertisement array, send the advertisement array to the video player. In this way, after the video player obtains the target video, the advertisements in the advertisement array can be delivered according to the advertisement array during the video playing process.

例如，本实施例的步骤101“根据目标视频的视频结构化信息以及预先设置的广告库，生成目标视频对应的广告数组”，具体可以包括如下两个步骤：For example, step 101 of this embodiment "generate an array of advertisements corresponding to the target video according to the video structured information of the target video and the preset advertisement library" may specifically include the following two steps:

(1)、根据视频结构化信息从广告库中获取与目标视频关联的至少一条目标广告的信息；(1) Obtain at least one target advertisement information associated with the target video from the advertisement library according to the video structured information;

(2)根据视频结构化信息和对应的至少一条目标广告的信息，生成广告数组。(2) Generate an array of advertisements according to the structured information of the video and the information of at least one corresponding target advertisement.

具体地，根据视频结构化信息中的一个或者多个结构信息，可以确定目标广告的条数，然后可以从广告库中获取与目标视频关联的目标广告的信息。再接着，根据视频结构化信息中的各结构化信息的出现时间等信息，可以限定相关的广告的投放时间、投放方式等等，从而可以实现根据视频结构化信息和对应的至少一条目标广告的信息，生成广告数组。例如具体可以包括如下几种情况：Specifically, according to one or more pieces of structural information in the video structured information, the number of target advertisements can be determined, and then the information of the target advertisements associated with the target video can be obtained from the advertisement library. Then, according to the information such as the appearance time of each structured information in the video structured information, the time and mode of launching relevant advertisements can be limited, so that the advertising based on the video structured information and at least one corresponding target advertisement can be realized. information to generate an array of advertisements. For example, it may specifically include the following situations:

第一种情况、步骤(1)“根据视频结构化信息从广告库中获取与目标视频关联的至少一条目标广告的信息”，具体可以包括如下步骤：In the first case, step (1) "obtain at least one target advertisement information associated with the target video from the advertisement library according to the structured information of the video", may specifically include the following steps:

(a1)根据题材类型，从广告库中获取与题材类型对应的多条第一候选广告的信息；(a1) According to the subject type, obtain the information of multiple first candidate advertisements corresponding to the subject type from the advertisement database;

(b1)根据目标视频的长度和预设的广告间隔时长，确定题材类型对应的第一目标广告的条数m；m为大于或者等于1的整数；(b1) Determine the number m of the first target advertisement corresponding to the theme type according to the length of the target video and the preset advertisement interval; m is an integer greater than or equal to 1;

(c1)根据各第一候选广告的标价，从多条第一候选广告的信息中获取m条第一目标广告的信息。(c1) According to the bids of the first candidate advertisements, information on m pieces of first target advertisements is obtained from information on multiple pieces of first candidate advertisements.

该种情况中，是根据视频结构化信息中包括的目标视频的题材类型以及目标视频的长度来生成广告数组。例如广告库中的各条广告，广告主在投放广告时可以定向广告面向的视频的题材类型。这样，根据当前目标视频的题材类型，可以从广告库中获取定向到该题材类型的多条第一候选广告，并获取到多条第一候选广告的信息，如各第一候选广告的标识、地址、投放方式等等。需要说明的是，本实施例的题材类型的广告的投放方式为固定的线性投放方式，在选取题材类型对应的多条第一候选广告的时候，可以根据线性投放方式的限定，筛选符合条件的多条第一候选广告的信息。对于广告库中未限定投放方式的广告默认可以适用于任何一种投放方式。此时可以由广告投放装置根据广告位的位置来设置广告投放方式，例如对于题材类型的广告设置的广告位是前贴、中贴或者后贴的线性投放方式的广告位。In this case, the advertisement array is generated according to the subject type of the target video included in the video structural information and the length of the target video. For example, for each advertisement in the advertisement library, the advertiser can target the subject type of the video targeted for the advertisement when placing the advertisement. In this way, according to the theme type of the current target video, multiple first candidate advertisements directed to the theme type can be obtained from the advertisement library, and the information of the multiple first candidate advertisements, such as the identification of each first candidate advertisement, address, delivery method, etc. It should be noted that the advertisement delivery method of the theme type in this embodiment is a fixed linear delivery method. When selecting multiple first candidate advertisements corresponding to the theme type, the qualified ones can be screened according to the limitation of the linear delivery method. Multiple pieces of first candidate advertisement information. Advertisements in the ad library that are not limited to delivery methods can be applied to any delivery method by default. At this time, the advertisement delivery device can set the advertisement delivery method according to the position of the advertisement space, for example, the advertisement space set for the advertisement of the subject type is an advertisement space in the linear delivery mode of front post, middle post or post post.

本实施例中会预先设置有广告间隔时长，例如20分钟，或者30分钟等。目标视频的长度为该目标视频的播放时长，如1小时，这样，根据广告间隔时长，可以预测该目标视频中可以投放第一目标广告的条数m，本实施例中，设置m的数量至少为1。也就是说，即使目标视频的长度小于广告间隔时长，也可以设置1条第一目标广告。因此，本实施例中的m为大于或者等于1的整数。由于本实施例中在视频中投放广告本身就是缓解视频播放平台的版权、带宽以及存储方面的成本压力，所以在获取到多条第一候选广告的信息之后，可以根据各第一候选广告的标价，从多条第一候选广告的信息中获取标价最高的m条第一目标广告的信息。本实施例的各候选广告的标价可以为广告库中记录的广告商的出价，还可以为视频服务器根据该候选广告的各参数预测的标价，例如可以根据该候选广告杯投放后带来的收益来预测，或者也可以根据其他方式来预测，在此不作限定。In this embodiment, the advertising interval is preset, for example, 20 minutes or 30 minutes. The length of the target video is the playback duration of the target video, such as 1 hour. Like this, according to the advertisement interval length, it can be predicted that the number m of the first target advertisements can be placed in the target video. In this embodiment, the number of m is set to be at least is 1. That is to say, even if the length of the target video is shorter than the interval between advertisements, one first target advertisement can be set. Therefore, m in this embodiment is an integer greater than or equal to 1. Since placing advertisements in the video itself in this embodiment is to relieve the copyright, bandwidth and storage cost pressures of the video playback platform, after obtaining the information of a plurality of first candidate advertisements, the price of each first candidate advertisement can be , obtaining the information of the m first target advertisements with the highest bid from the information of the plurality of first candidate advertisements. The bidding price of each candidate advertisement in this embodiment can be the advertiser's bid recorded in the advertisement library, and can also be the bidding price predicted by the video server according to the parameters of the candidate advertisement, for example, it can be based on the revenue brought by the placement of the candidate advertisement cup To predict, or it can also be predicted in other ways, which is not limited here.

此时对应地，步骤(2)“根据视频结构化信息和对应的至少一条目标广告的信息，生成广告数组”，具体可以包括：根据m条第一目标广告中的各第一目标广告的标价、目标视频的长度、广告间隔时长和第一目标广告的数量，确定各第一目标广告的投放方式以及投放时间；例如可以先从m条第一目标广告中获取标价最高的第一目标广告，将其投放在目标视频的开头，这样可以设置该第一目标广告的投放时间，该第一目标广告设置在目标视频的开头，所以其投放方式为线性投放的前贴。然后，目标视频播放广告间隔时长之后，可以再投放下一条广告，此时再从剩下的m-1条第一目标广告中获取标价最高的第一目标广告，再设置对应的投放时间、投放方式。由于该第一目标广告设置在目标视频中间，所以其投放方式为线性投放的中贴。依次类推，可以确定每一个第一目标投放广告的投放时间和投放方式。其中，对于投放在目标视频的结束的第一目标广告，其投放方式设置为对应的线性投放的后贴即可。最后将各第一目标广告的标识、各第一目标广告的投放方式及投放时间以及各第一目标广告的地址，存入广告数组中。广告数组中的各第一目标广告的信息，优选地按照各第一目标广告的投放时间的先后顺序来存储。Correspondingly at this time, the step (2) "generate an array of advertisements according to the video structured information and the information of at least one corresponding target advertisement" may specifically include: according to the price of each first target advertisement in the m first target advertisements , the length of the target video, the length of the advertisement interval and the number of the first target advertisements, determine the delivery method and delivery time of each first target advertisement; for example, the first target advertisement with the highest price can be obtained from the m first target advertisements, It is placed at the beginning of the target video, so that the delivery time of the first target advertisement can be set, and the first target advertisement is set at the beginning of the target video, so its delivery method is a pre-post of linear delivery. Then, after the target video plays the advertisement interval, the next advertisement can be delivered. At this time, the first target advertisement with the highest price is obtained from the remaining m-1 first target advertisements, and then the corresponding delivery time and delivery time are set. Way. Since the first target advertisement is set in the middle of the target video, its delivery method is a mid-post of linear delivery. By analogy, the delivery time and delivery method of each first target delivery advertisement can be determined. Wherein, for the first target advertisement placed at the end of the target video, its delivery mode may be set as post-post of the corresponding linear delivery. Finally, the identifier of each first target advertisement, the delivery method and delivery time of each first target advertisement, and the address of each first target advertisement are stored in the advertisement array. The information of the first target advertisements in the advertisement array is preferably stored in the order of the delivery time of the first target advertisements.

需要说明的是，上述情况中，确定题材类型对应的第一目标广告的条数m仅为一种实现方式，实际应用中，对于每一条目标视频，还可以在视频结构化信息中预先设置题材类型的第一目标广告的数量，如允许在目标视频开始投放一条、或者仅允许在目标视频开始和结束各投放一条、或者允许在目标视频开始、结束以及中间各投放一条。或者还可以采用其他方式限定第一目标广告的条数，在此不再一一举例赘述。It should be noted that, in the above case, determining the number m of the first target advertisements corresponding to the theme type is only one way of implementation. In practical applications, for each target video, the theme can also be preset in the video structured information. The number of first target advertisements of the type, for example, one is allowed to be placed at the beginning of the target video, or only one is allowed to be placed at the beginning and the end of the target video, or one is allowed to be placed at the beginning, end and middle of the target video. Alternatively, other methods may be used to limit the number of first target advertisements, which will not be described here one by one.

进一步可选地，在上述情况中，还可以根据视频化结构信息中包括的目标视频中出现频次最高的数个关键词来获取候选广告。关键词的数量可以根据具体需求来设置，如10个、8个或者5个或者其他数值。例如，步骤(a1)“根据题材类型，从广告库中获取相关的多条第一候选广告的信息”，具体可以包括：根据题材类型，并结合数个关键词，从广告库中获取与题材类型对应、且与数个关键词相关的多条第一候选广告的信息。例如可以先根据题材类型筛选出多个广告之后，再根据广告库中的广告摘要，从筛选得到的多个广告中获取与数个关键词中的一个或者多个关键词相关性最高的多个广告，作为多条第一候选广告，并对应获取多条第一候选广告的信息。Further optionally, in the above case, candidate advertisements may also be acquired according to several keywords with the highest frequency of occurrence in the target video included in the video structure information. The number of keywords can be set according to specific requirements, such as 10, 8 or 5 or other values. For example, the step (a1) of "obtaining information about a plurality of first candidate advertisements from the advertisement library according to the subject type" may specifically include: according to the subject type, combined with several keywords, obtaining information related to the subject matter from the advertisement library The information of multiple first candidate advertisements corresponding to the type and related to several keywords. For example, after filtering out multiple advertisements based on the type of subject matter, and then according to the summary of the advertisements in the advertisement library, from the multiple advertisements obtained through screening, obtain the advertisements with the highest correlation with one or more keywords among the several keywords. Advertisements, as multiple first candidate advertisements, correspondingly acquire information of multiple first candidate advertisements.

第二种情况、步骤(1)“根据视频结构化信息从广告库中获取与目标视频关联的至少一条目标广告的信息”，具体可以包括如下步骤：In the second case, step (1) "obtaining information about at least one target advertisement associated with the target video from the advertisement library according to the structured information of the video", may specifically include the following steps:

(a2)根据人物标识，从广告库中获取与人物标识对应人物相关的多条第二候选广告的信息；(a2) According to the person identification, obtain the information of multiple second candidate advertisements related to the person corresponding to the person identification from the advertisement database;

(b2)根据人物的出现时间以及预设的广告投放时长，确定人物对应的第二目标广告的条数n；n为大于或者等于1的整数；(b2) Determine the number n of second target advertisements corresponding to the character according to the appearance time of the character and the preset advertisement delivery duration; n is an integer greater than or equal to 1;

(c2)根据各第二候选广告的标价，从多条第二候选广告中获取n条第二目标广告；(c2) Obtain n pieces of second target advertisements from multiple second candidate advertisements according to the bids of the second candidate advertisements;

该种情况中，是根据视频结构化信息中包括的目标视频中的人物标识以及人物标识对应的人物的出现时间来生成广告数组。In this case, the advertisement array is generated according to the character identifier in the target video included in the video structured information and the appearance time of the character corresponding to the character identifier.

本实施例的人物可以为一些公众人物，例如这些人物可以代言广告，从而实现人物与广告的关联。例如广告库中的各条广告，广告主在投放广告时可以标识广告的代言人。这样，根据人物标识，可以从广告库中获取与人物标识对应人物相关的多条第二候选广告的信息；如各第二候选广告的标识、地址、投放方式等等。需要说明的是，本实施例的人物标识的广告的投放方式为非线性投放方式或者伴随性投放方式，在人物标识对应的多条第二候选广告的时候，可以根据非线性投放方式或者伴随性投放方式的限定，筛选符合条件的多条第二候选广告的信息。对于广告库中未限定投放方式的广告默认可以适用于任何一种投放方式。此时可以由广告投放装置根据广告位的位置来设置广告投放方式，例如对于人物标识对应的广告设置的广告位是非线性投放方式的广告位或者伴随性投放方式的广告位。The characters in this embodiment can be some public figures, for example, these characters can endorse advertisements, so as to realize the association between characters and advertisements. For example, for each advertisement in the advertisement library, the advertiser can identify the spokesperson of the advertisement when placing the advertisement. In this way, according to the person identification, the information of a plurality of second candidate advertisements related to the person corresponding to the person identification can be obtained from the advertisement library; such as the identification, address, delivery method, etc. of each second candidate advertisement. It should be noted that the advertising delivery method of the person identification in this embodiment is a non-linear delivery method or a concomitant delivery method. When there are multiple second candidate advertisements corresponding to the person identification, it can The delivery method is limited, and the information of multiple second candidate advertisements that meet the conditions is screened. Advertisements in the ad library that are not limited to delivery methods can be applied to any delivery method by default. At this time, the advertisement delivery device can set the advertisement delivery method according to the location of the advertisement space, for example, the advertisement space set for the advertisement corresponding to the character identification is an advertisement space of a non-linear delivery method or an advertisement space of an accompanying delivery method.

本实施例中的人物出现次数可以为一次或者多次，对应的人物出现时间可以为一段或者多段。本实施例中，可以预设广告投放时长，例如5s、8s或者10s等其他数值。这样可以根据人物的每次出现时间，确定人物的出现时间是否长到足够投放一条广告，如果可以，则确定这段出现时间可以投放一条第二目标广告。当然，如果出现时间足够长，一次的出现也可以投放两条或者多条第二目标广告。通过对该人物每次的出现时间进行分析，可以获取到目标视频中能够投放第二目标广告的条数n，即为最大条数，本实施例中以n大于或者等于1的正整数为例来介绍第二目标广告的获取。然后根据各第二候选广告的标价，从多条第二候选广告中获取n条第二目标广告；详细参考上述步骤(c1)的实施，在此不再赘述。实际应用中n也可以为0，即表示未获取到该人物相关的广告。The number of appearances of the characters in this embodiment may be one or more times, and the corresponding appearance time of the characters may be one or more periods. In this embodiment, the advertising delivery time can be preset, such as 5s, 8s or 10s and other values. In this way, according to each appearance time of the character, it can be determined whether the appearance time of the character is long enough to place an advertisement, and if yes, it is determined that a second target advertisement can be placed during this appearance time. Of course, if the appearance time is long enough, two or more second target advertisements can also be placed in one appearance. By analyzing the appearance time of the character each time, the number n of the second target advertisements in the target video can be obtained, which is the maximum number. In this embodiment, a positive integer greater than or equal to 1 is taken as an example Let's introduce the acquisition of the second target advertisement. Then, according to the bids of the second candidate advertisements, n pieces of second target advertisements are obtained from multiple second candidate advertisements; refer to the implementation of the above-mentioned step (c1) for details, which will not be repeated here. In practical applications, n may also be 0, which means that no advertisement related to the character has been obtained.

另外，本实施例中，广告投放装置可以设置各种广告投放策略，例如一目标视频中，投放的同一人物相关的第二目标广告的条数可以为1条、2条、3条或者其他条数。此时可以根据该人物的出现时间将标价高的第二目标广告投放在靠前的位置，而标价低的第二目标广告投放在靠后的位置。如果该人物多次出现，每次出现时间足够长，但是第二目标广告的条数足够少，如仅2条，可以设置在人物出现过程中，2条广告轮流播，或者也可以设置每次出现每条第二目标广告仅播一遍。本实施例中，广告投放装置可以设置的广告投放策略还可以为每次人物出现，都采用伴随性投放方式播放一条第二目标广告，此时也可以不限定第二目标广告的投放时长，可以略大于人物出现时间长度。In addition, in this embodiment, the advertisement delivery device can set various advertisement delivery strategies. For example, in a target video, the number of second target advertisements related to the same person can be 1, 2, 3 or other items. number. At this time, according to the appearance time of the character, the second target advertisement with a high price can be placed in the front position, while the second target advertisement with a low price can be placed in the rear position. If the character appears multiple times, the time for each appearance is long enough, but the number of advertisements for the second target is small enough, such as only 2, it can be set that the 2 advertisements will be broadcast in turn during the appearance of the character, or it can be set every time Each secondary target advertisement appears only once. In this embodiment, the advertisement delivery strategy that can be set by the advertisement delivery device can also be that every time a character appears, a second target advertisement will be played in an accompanying delivery mode. At this time, the delivery time of the second target advertisement may not be limited, and may Slightly longer than the length of time the character appears.

此时对应地，步骤(2)“根据视频结构化信息和对应的至少一条目标广告的信息，生成广告数组”，具体可以包括：根据n条第二目标广告中各第二目标广告的标价、人物的出现时间和预设的广告投放时长，确定各第二目标广告的投放时间；将各第二目标广告的标识、各第二目标广告的投放方式及投放时间以及各第二目标广告的地址，存入广告数组中。该步骤实施时与第一种情况中的投放方式不同，该种情况的投放方式为非线性投放或者伴随性投放，其他实现过程与上述第一种情况相同，详细参考上述相关实施的记载，在此不再赘述。Correspondingly at this time, the step (2) "generate an array of advertisements according to the video structured information and the information of at least one corresponding target advertisement" may specifically include: according to the price of each second target advertisement in n pieces of second target advertisements, The appearance time of the characters and the preset advertisement delivery time are used to determine the delivery time of each second target advertisement; the identification of each second target advertisement, the delivery method and delivery time of each second target advertisement, and the address of each second target advertisement , stored in the ads array. The implementation of this step is different from the delivery method in the first case. The delivery method in this case is non-linear delivery or accompanying delivery. The other implementation process is the same as the first case above. For details, refer to the relevant implementation records above. This will not be repeated here.

进一步可选地，在上述情况中，还可以根据视频化结构信息中包括的目标视频中人物的行为来获取候选广告。例如，步骤(a2)“根据人物标识，从广告库中获取人物标识对应人物相关的多条第二候选广告的信息”，具体可以包括：根据人物标识，并结合人物的行为，从广告库中获取与人物以及人物的行为相关的多条第二候选广告的信息。例如可以先根据人物标识筛选出多个广告之后，再根据人物的行为从多个广告中获取与人物行为相关的多条第二候选广告，并获取对应的多条第二候选广告的信息。例如目标视频中某个公众人物在行走，可以获取该公众人物代言或者推荐的饮料的广告作为第二候选广告，因为行走过程中累了，通过喝饮料可以补充能量。公众人物在遛狗过程中可以获取该公众人物代言或推荐的各种与狗相关的宠物医院或者宠物食品或者其他宠物用品等等。公众人物在做瑜伽时，可以获取该公众人物代言的瑜伽服装、或者瑜伽垫或者还可以获取其他相关的运动服装或者装备的广告最为第二候选广告。对应的，在从多个第二候选广告中获取第二目标广告，并生成广告数组时，也可以参考人物的行为，优选地，在该人物的行为发生时，投放该人物的行为相关的第二目标广告。Further optionally, in the above case, the candidate advertisements may also be acquired according to the behavior of the characters in the target video included in the video structure information. For example, the step (a2) of "obtaining from the advertisement database the information of multiple second candidate advertisements related to the person whose identity corresponds to the person according to the person's identity" may specifically include: according to the person's identity and combined with the behavior of the person, obtaining Information about a plurality of second candidate advertisements related to the person and the behavior of the person is obtained. For example, a plurality of advertisements may be screened out according to the identity of the person, and then a plurality of second candidate advertisements related to the behavior of the person may be obtained from the plurality of advertisements according to the behavior of the person, and information of the corresponding plurality of second candidate advertisements may be obtained. For example, in the target video, if a public figure is walking, an advertisement of a drink endorsed or recommended by the public figure can be obtained as the second candidate advertisement. Because you are tired during walking, you can replenish energy by drinking a drink. During the dog walk process, the public figure can obtain various dog-related pet hospitals, pet food or other pet products, etc. endorsed or recommended by the public figure. When a public figure is doing yoga, he can obtain the advertisement of the yoga clothing endorsed by the public figure, or the yoga mat, or other related sports clothing or equipment as the second candidate advertisement. Correspondingly, when the second target advertisement is obtained from a plurality of second candidate advertisements and an advertisement array is generated, the behavior of the character may also be referred to. Preferably, when the behavior of the character occurs, the first advertisement related to the behavior of the character is placed Two-target advertising.

第三种情况、步骤(1)“根据视频结构化信息从广告库中获取与目标视频关联的至少一条目标广告的信息”，具体可以包括如下步骤：The third case, step (1) "obtain at least one target advertisement information associated with the target video from the advertisement library according to the video structured information", may specifically include the following steps:

(a3)根据物体标识，从广告库中获取物体标识对应物体相关的多条第三候选广告的信息；(a3) According to the object identification, obtain the information of multiple third candidate advertisements related to the object corresponding to the object identification from the advertisement library;

(b3)根据物体的出现时间以及预设的广告投放时长，确定第三目标广告的条数p；p为大于或者等于1的整数；(b3) Determine the number p of the third target advertisement according to the appearance time of the object and the preset advertisement delivery duration; p is an integer greater than or equal to 1;

(c3)根据各第三候选广告的标价，从多条第三候选广告中获取p条第三目标广告；(c3) Obtaining p third target advertisements from multiple third candidate advertisements according to the price tag of each third candidate advertisement;

该种情况中，是根据视频结构化信息中包括的目标视频中的物体标识以及物体标识对应的物体的出现时间来生成广告数组。In this case, the advertisement array is generated according to the object identifier in the target video included in the video structured information and the appearance time of the object corresponding to the object identifier.

本实施例的物体可以为动物、植物或者各种生活用品、体育用品或者其他任何一实物。步骤(a3)-(c3)的实施详细可以参考上述步骤(a2)-(c2)的实施，在此不再赘述。The objects in this embodiment can be animals, plants, various daily necessities, sporting goods or any other real objects. For details on the implementation of steps (a3)-(c3), reference may be made to the implementation of the above-mentioned steps (a2)-(c2), which will not be repeated here.

同理，对应地，步骤(2)“根据视频结构化信息和对应的至少一条目标广告的信息，生成广告数组”，具体可以包括：根据p条第三目标广告中各第三目标广告的标价、物体的出现时间和预设的投放时长，确定各第三目标广告的投放时间；将各第三目标广告的标识、各第三目标广告的投放方式及投放时间以及各第三目标广告的地址，存入广告数组中。实物对应的第三目标广告的投放方式与上述第二种情况中的人物对应的广告的投放方式相同，详细可以参考上述相关实施的记载，在此不再赘述。Similarly, correspondingly, the step (2) "generate an array of advertisements based on the structured information of the video and the information of at least one corresponding target advertisement" may specifically include: , the appearance time of the object and the preset delivery time, determine the delivery time of each third target advertisement; the identification of each third target advertisement, the delivery method and delivery time of each third target advertisement, and the address of each third target advertisement , stored in the ads array. The delivery method of the third target advertisement corresponding to the physical object is the same as that of the advertisement corresponding to the person in the second case above, and for details, please refer to the relevant implementation records above, and will not repeat them here.

进一步可选地，若本实施例的广告投放装置为视频服务器时，此时对应地，在步骤100之前，还可以包括视频服务器获取目标视频的视频结构化信息的过程，例如，具体可以包括如下步骤：Further optionally, if the advertisement delivery device in this embodiment is a video server, correspondingly, before step 100, it may also include a process for the video server to obtain the video structured information of the target video, for example, it may specifically include the following step:

(a4)从目标视频的视频流中按照时间先后顺序提取多张缩略图；各张缩略图按照对应的时间戳来标识；(a4) Extract multiple thumbnails in chronological order from the video stream of the target video; each thumbnail is identified according to the corresponding timestamp;

本实施例在获取目标视频的视频结构化信息时，需要先对目标视频拆分为视频流和音频流。首先从视频流中提取多张缩略图。例如本实施例的缩略图可以每秒选取一张，具体可以采用ffmpeg的方式来从目标视频的视频流中每秒提取一张图片，作为该秒的缩略图。并采用该秒的时间戳作为该张缩略图的标识，这样视频长度多少秒，便对应多少张缩略图。(b4)对多张缩略图进行题材类型识别，得到目标视频的题材类型；In this embodiment, when acquiring the video structured information of the target video, it is necessary to first split the target video into a video stream and an audio stream. First extract multiple thumbnails from the video stream. For example, one thumbnail in this embodiment can be selected every second, specifically, ffmpeg can be used to extract one picture per second from the video stream of the target video as the thumbnail for that second. And use the timestamp of the second as the identification of the thumbnail, so how many seconds the video length corresponds to how many thumbnails. (b4) Carry out theme type recognition to a plurality of thumbnails, obtain the theme type of target video;

本实施例中可以通过预先训练题材类型识别模型，来识别目标视频的题材类型。该题材类型识别模型仅可以对预设的多种题材类型进行预测。使用时，可以从多张缩略图中选取预设张数的缩略图，然后输入至题材类型识别模型，这样题材类型识别模型可以预测这些图片对应的目标视频属于每种预设题材类型的概率，可以选取概率最大的题材类型作为预测的题材类型。In this embodiment, the subject type of the target video can be identified by pre-training the subject type identification model. The theme type identification model can only predict multiple preset theme types. When in use, you can select a preset number of thumbnails from multiple thumbnails, and then input them into the theme type recognition model, so that the theme type recognition model can predict the probability that the target video corresponding to these pictures belongs to each preset theme type, The theme type with the highest probability can be selected as the predicted theme type.

另外，可选地，实际应用中，视频的开头部分的内容更能代表视频的题材类型。本实施例中，还可以通过在目标视频的前10分钟中内每秒选取一张缩略图，得到600张缩略图；然后利用这600张缩略图来使用题材类型识别模型识别目标视频的题材类型。相对于使用目标视频的全部缩略图来识别目标视频的题材类型，实现方式更简单。In addition, optionally, in practical applications, the content at the beginning of the video can better represent the genre of the video. In this embodiment, it is also possible to obtain 600 thumbnails by selecting one thumbnail per second in the first 10 minutes of the target video; then use these 600 thumbnails to identify the subject type of the target video using the subject type identification model . Compared with using all the thumbnails of the target video to identify the subject type of the target video, the implementation method is simpler.

(c4)对各张缩略图进行人物识别、行为识别和物体识别，得到各张缩略图中包括的人物、人物的行为以及物体；(c4) Perform character recognition, behavior recognition and object recognition on each thumbnail, and obtain the characters, behaviors and objects included in each thumbnail;

(d4)根据各张缩略图中包括的人物、人物的行为、物体以及各张缩略图的时间戳，获取目标视频中的人物的出现时间、人物的行为出现时间以及各物体的出现时间；(d4) Obtain the appearance time of the characters in the target video, the behavior appearance time of the characters, and the appearance time of each object according to the characters, characters' behaviors, objects and the timestamps of each thumbnail included in each thumbnail;

本实施例中还可以预先存储有待识别的人物的面部特征，这样识别每张缩略图时可以截取缩略图中的人物面部，然后和预先粗处的人物的面部特征对比，判断是不是该人物，若是，记录该张缩略图中包括的人物标识。对每一张缩略图都进行分析，可以知道该人物在哪些缩略图中出现过。然后根据每张缩略图对应的时间戳，可以分析出该人物的出现时间。另外，还可以采用预先训练的人物识别模型来识别缩略图中的人物。例如想要识别某公众人物M，预先采集M的多张图片以及少量的非M的图片来训练该人物识别模型。然后使用时，将每张缩略图输入至该人物识别模型，可以识别出该缩略图中是否包括该人物。In this embodiment, the facial features of the person to be recognized can also be stored in advance, so that when each thumbnail is recognized, the face of the person in the thumbnail can be intercepted, and then compared with the facial features of the person at the rough place in advance to determine whether it is the person. If yes, record the person ID included in the thumbnail. By analyzing each thumbnail, it is possible to know which thumbnails the character has appeared in. Then, according to the timestamp corresponding to each thumbnail, the appearance time of the character can be analyzed. In addition, a pre-trained person recognition model can also be used to recognize people in the thumbnail. For example, if you want to identify a public figure M, you need to collect multiple pictures of M and a small number of non-M pictures in advance to train the character recognition model. Then, when using, each thumbnail is input to the person recognition model, and it can be recognized whether the person is included in the thumbnail.

另外，本实施例中还可以预先训练人物行为识别模型，例如可以预先采集人物的多种已知行为中每种行为连续的多张图片进行训练，这样，在使用时，可以根据连续多张缩略图中人物的行为可以预测该行为是什么行为。识别人物行为时，可以将该人物连续出现的多张缩略图作为一组照片输入至人物行为识别模型，预测该人物的行为各已知行为的概率，并选择概率最大的已知行为作为预测的该人物的行为。In addition, in this embodiment, the character behavior recognition model can also be pre-trained. For example, multiple continuous pictures of each behavior in various known behaviors of the character can be collected in advance for training. The behavior of the characters in the thumbnail can predict what behavior that behavior is. When identifying a person's behavior, multiple thumbnails of the person's continuous appearance can be input into the person's behavior recognition model as a group of photos to predict the probability of each known behavior of the person's behavior, and select the known behavior with the highest probability as the predicted one. the character's behavior.

同理，物体的识别类似，可以预先训练物体识别模型，例如，可以采集多种已知物体对物体识别模型进行训练。在使用时，根据包含物体的缩略图可以预测该物体为各已知物体的概率，并选择概率最大的已知物体的标识作为预测的该物体的标识。同理，对所有缩略图进行分析，可以统计出每个物体的出现时间。Similarly, object recognition is similar, and the object recognition model can be trained in advance, for example, a variety of known objects can be collected to train the object recognition model. In use, the probability that the object is each known object can be predicted according to the thumbnail image containing the object, and the identifier of the known object with the highest probability is selected as the predicted identifier of the object. Similarly, by analyzing all the thumbnails, the appearance time of each object can be counted.

(e4)对多张缩略图进行文字识别，得到第一文本信息；(e4) performing text recognition on multiple thumbnail images to obtain the first text information;

例如，可以采用光学字符识别(Optical Character Recognition；OCR)对每张缩略图进行识别，获取其中的文本信息。对得到的多张缩略图进行文字识别，将得到的文本信息拼接，可以得到第一文本信息。For example, optical character recognition (Optical Character Recognition; OCR) may be used to recognize each thumbnail and obtain text information therein. The first text information can be obtained by performing character recognition on the obtained plurality of thumbnail images, and splicing the obtained text information.

(f4)对目标视频的音频流进行语音识别，得到第二文本信息；(f4) carry out speech recognition to the audio stream of target video, obtain the second text information;

例如，可以采用自动语音识别(Automatic Speech Recognition；ASR)对音频流进行语音识别，得到第二文本信息。For example, automatic speech recognition (Automatic Speech Recognition; ASR) may be used to perform speech recognition on the audio stream to obtain the second text information.

(g4)从第一文本信息和第二文本信息中提取出现频次最高的数个关键词。(g4) Extract several keywords with the highest frequency of occurrence from the first text information and the second text information.

具体地，可以先对第一文本信息和第二文本信息进行切词，得到多个词语；然后统计分析每个词语在第一文本信息和第二文本信息中出现的频率，并从多个词语中提取频次最高的数个关键词。Specifically, the first text information and the second text information can be word-segmented to obtain a plurality of words; then the frequency of occurrence of each word in the first text information and the second text information can be statistically analyzed, and from the plurality of words Several keywords with the highest frequency of extraction.

经上述步骤(a4)-(g4)的处理，可以将获取的所有信息作为该目标视频的视频结构化信息。对每一个目标视频，都可以采用类似的处理，获取对应的视频结构化信息。After the above steps (a4)-(g4), all the acquired information can be used as the video structured information of the target video. For each target video, similar processing can be adopted to obtain the corresponding video structured information.

本实施例的视频中的广告投放方法，通过接收到视频播放器发送的携带目标视频标识的广告请求之后，根据目标视频标识获取对应的目标视频的视频结构化信息；根据目标视频的视频结构化信息以及预先设置的广告库，生成目标视频对应的广告数组；向视频播放器发送广告数组，以供视频播放器根据广告数组投放各目标广告，可以实现智能化地为目标视频配置广告，同时还能够保证投放的广告与目标视频的关联性，从而能够有效地提高广告投放的效率。In the video advertisement placement method of this embodiment, after receiving the advertisement request carrying the target video identifier sent by the video player, the video structured information of the corresponding target video is obtained according to the target video identifier; according to the video structured information of the target video Information and pre-set ad library to generate an ad array corresponding to the target video; send the ad array to the video player for the video player to deliver each target ad based on the ad array, intelligently configure the ad for the target video, and at the same time It can ensure the relevance of the delivered advertisement and the target video, thereby effectively improving the efficiency of advertisement delivery.

图2为本发明的视频中的广告投放装置实施例一的结构图。如图2所示，本实施例的视频中的广告投放装置，具体可以包括：FIG. 2 is a structural diagram of Embodiment 1 of the device for placing advertisements in videos according to the present invention. As shown in Figure 2, the advertisement placement device in the video of this embodiment may specifically include:

获取模块10用于接收到视频播放器发送的携带目标视频标识的广告请求之后，根据目标视频标识获取对应的目标视频的视频结构化信息；After the acquisition module 10 is used to receive the advertisement request carrying the target video identifier sent by the video player, acquire the video structured information of the corresponding target video according to the target video identifier;

生成模块11用于根据获取模块10获取的目标视频的视频结构化信息以及预先设置的广告库，生成目标视频对应的广告数组；The generating module 11 is used for generating the corresponding advertisement array of the target video according to the video structured information of the target video acquired by the acquiring module 10 and the preset advertisement storehouse;

发送模块12用于向视频播放器发送生成模块11生成的广告数组，以供视频播放器根据广告数组投放各目标广告。The sending module 12 is used to send the advertisement array generated by the generation module 11 to the video player, so that the video player can deliver each target advertisement according to the advertisement array.

本实施例的视频中的广告投放装置，通过采用上述模块实现视频中的广告投放的实现原理以及技术效果与上述相关方法实施例的实现相同，详细可以参考上述相关方法实施例的记载，在此不再赘述。The advertisement placement device in the video of this embodiment, by using the above modules to realize the realization principle and technical effect of the advertisement placement in the video, is the same as the implementation of the above-mentioned related method embodiments. For details, please refer to the records of the above-mentioned related method embodiments, here No longer.

图3为本发明的视频中的广告投放装置实施例二的结构图。如图3所示，本实施例的视频中的广告投放装置，在上述图2所示实施例的技术方案的基础上，进一步更加详细地介绍本发明的技术方案。FIG. 3 is a structural diagram of Embodiment 2 of the device for placing advertisements in videos according to the present invention. As shown in FIG. 3 , the advertisement placement device in the video of this embodiment further introduces the technical solution of the present invention in more detail on the basis of the technical solution of the above-mentioned embodiment shown in FIG. 2 .

如图3所示，本实施例的视频中的广告投放装置中，生成模块11具体包括：As shown in FIG. 3 , in the advertisement placement device in the video of this embodiment, the generating module 11 specifically includes:

广告获取单元111用于根据获取模块10获取的视频结构化信息从广告库中获取与目标视频关联的至少一条目标广告的信息；The advertisement obtaining unit 111 is used for obtaining the information of at least one target advertisement associated with the target video from the advertisement library according to the video structured information obtained by the obtaining module 10;

数组生成单元112用于根据获取模块10获取的视频结构化信息和广告获取单元111获取的对应的至少一条目标广告的信息，生成广告数组。The array generation unit 112 is configured to generate an advertisement array according to the video structured information acquired by the acquisition module 10 and the corresponding at least one target advertisement information acquired by the advertisement acquisition unit 111 .

进一步可选地，若视频结构化信息包括目标视频的题材类型以及目标视频的长度；广告获取单元111具体用于：Further optionally, if the video structured information includes the subject type of the target video and the length of the target video; the advertisement obtaining unit 111 is specifically used for:

根据题材类型，从广告库中获取与题材类型对应的多条第一候选广告的信息；According to the subject type, obtain the information of a plurality of first candidate advertisements corresponding to the subject type from the advertisement library;

根据目标视频的长度和预设的广告间隔时长，确定题材类型对应的第一目标广告的条数m；m为大于或者等于1的整数；According to the length of the target video and the preset advertisement interval duration, determine the number m of the first target advertisement corresponding to the theme type; m is an integer greater than or equal to 1;

根据各第一候选广告的标价，从多条第一候选广告的信息中获取m条第一目标广告的信息；According to the price tag of each first candidate advertisement, the information of m first target advertisements is obtained from the information of multiple first candidate advertisements;

对应地，数组生成单元112具体用于：Correspondingly, the array generating unit 112 is specifically used for:

根据m条第一目标广告中的各第一目标广告的标价、目标视频的长度、广告间隔时长和第一目标广告的数量，确定各第一目标广告的投放方式以及投放时间；According to the price of each first target advertisement in the m first target advertisements, the length of the target video, the length of the advertisement interval and the quantity of the first target advertisements, determine the delivery method and delivery time of each first target advertisement;

将各第一目标广告的标识、各第一目标广告的投放方式及投放时间以及各第一目标广告的地址，存入广告数组中。Store the identifier of each first target advertisement, the delivery method and delivery time of each first target advertisement, and the address of each first target advertisement into the advertisement array.

进一步可选地，若视频化结构信息中还包括目标视频中出现频次最高的数个关键词；广告获取单元111具体用于：根据题材类型，并结合数个关键词，从广告库中获取与题材类型对应、且与数个关键词相关的多条第一候选广告的信息。Further optionally, if the video structure information also includes several keywords with the highest frequency of occurrence in the target video; the advertisement acquisition unit 111 is specifically configured to: according to the type of subject matter and in combination with several keywords, acquire the relevant keywords from the advertisement library. Multiple pieces of first candidate advertisement information corresponding to the theme type and related to several keywords.

进一步可选地，若视频结构化信息包括目标视频中的人物标识以及人物标识对应的人物的出现时间；Further optionally, if the video structured information includes the character identification in the target video and the appearance time of the character corresponding to the character identification;

广告获取单元111具体用于：The advertisement acquiring unit 111 is specifically used for:

根据人物标识，从广告库中获取与人物标识对应人物相关的多条第二候选广告的信息；According to the person identification, information about a plurality of second candidate advertisements related to the person corresponding to the person identification is obtained from the advertisement database;

根据人物的出现时间以及预设的广告投放时长，确定人物对应的第二目标广告的条数n；n为大于或者等于1的整数；Determine the number n of second target advertisements corresponding to the character according to the appearance time of the character and the preset advertisement delivery time; n is an integer greater than or equal to 1;

根据各第二候选广告的标价，从多条第二候选广告中获取n条第二目标广告的信息；According to the price tag of each second candidate advertisement, information of n second target advertisements is obtained from multiple second candidate advertisements;

对应地，数组生成单元112具体用于：Correspondingly, the array generating unit 112 is specifically used for:

根据n条第二目标广告中各第二目标广告的标价、人物的出现时间和预设的广告投放时长，确定各第二目标广告的投放时间；According to the price of each second target advertisement in the n second target advertisements, the appearance time of the characters and the preset advertisement delivery time, determine the delivery time of each second target advertisement;

将各第二目标广告的标识、各第二目标广告的投放方式及投放时间以及各第二目标广告的地址，存入广告数组中。The identifier of each second target advertisement, the delivery method and delivery time of each second target advertisement, and the address of each second target advertisement are stored in the advertisement array.

进一步可选地，若视频结构化信息还包括目标视频中的人物的行为；Further optionally, if the video structured information also includes the behavior of the characters in the target video;

广告获取单元111具体用于：The advertisement acquiring unit 111 is specifically used for:

根据人物标识，并结合人物的行为，从广告库中获取与人物以及人物的行为相关的多条第二候选广告的信息。According to the character identification and in combination with the behavior of the character, the information of multiple second candidate advertisements related to the character and the behavior of the character is obtained from the advertisement library.

进一步可选地，若视频结构化信息包括目标视频中的物体标识以及物体标识对应的物体的出现时间；Further optionally, if the video structured information includes the object identifier in the target video and the appearance time of the object corresponding to the object identifier;

广告获取单元111具体用于：The advertisement acquiring unit 111 is specifically used for:

根据物体标识，从广告库中获取物体标识对应物体相关的多条第三候选广告的信息；According to the object identification, the information of multiple third candidate advertisements related to the object identification corresponding to the object is obtained from the advertisement library;

根据物体的出现时间以及预设的广告投放时长，确定第三目标广告的条数p；p为大于或者等于1的整数；Determine the number p of the third target advertisement according to the appearance time of the object and the preset advertisement delivery time; p is an integer greater than or equal to 1;

根据各第三候选广告的标价，从多条第三候选广告中获取p条第三目标广告的信息；According to the bid price of each third candidate advertisement, obtain the information of p third target advertisements from multiple third candidate advertisements;

对应地，数组生成单元112具体用于：Correspondingly, the array generating unit 112 is specifically used for:

根据p条第三目标广告中各第三目标广告的标价、物体的出现时间和预设的投放时长，确定各第三目标广告的投放时间；Determine the delivery time of each third-target advertisement according to the price of each third-target advertisement in the p third-target advertisements, the appearance time of the object, and the preset delivery time;

将各第三目标广告的标识、各第三目标广告的投放方式及投放时间以及各第三目标广告的地址，存入广告数组中。Store the identifier of each third target advertisement, the delivery method and delivery time of each third target advertisement, and the address of each third target advertisement into the advertisement array.

进一步可选地，如图3所示，本实施例的视频中的广告投放装置中还包括：Further optionally, as shown in FIG. 3 , the advertisement delivery device in the video of this embodiment also includes:

提取模块13用于从目标视频的视频流中按照时间先后顺序提取多张缩略图；各张缩略图按照对应的时间戳来标识；Extracting module 13 is used for extracting a plurality of thumbnails in chronological order from the video stream of target video; Each thumbnail is identified according to the corresponding timestamp;

识别模块14用于对提取模块13提取的多张缩略图进行题材类型识别，得到目标视频的题材类型；Recognition module 14 is used for carrying out theme type identification to the plurality of thumbnail images that extraction module 13 extracts, obtains the theme type of target video;

识别模块14还用于对提取模块13提取的各张缩略图进行人物识别、行为识别和物体识别，得到各张缩略图中包括的人物的标识、人物的行为以及物体的标识；The recognition module 14 is also used to carry out character recognition, behavior recognition and object recognition to each thumbnail image extracted by the extraction module 13, so as to obtain the identification of the person, the behavior of the person and the identification of the object included in each thumbnail;

获取模块10还用于根据提取模块13提取的各张缩略图中包括的人物的标识、人物的行为、物体的标识以及各张缩略图的时间戳，获取目标视频中的人物的出现时间、人物的行为的出现时间以及各物体的出现时间；The acquisition module 10 is also used to obtain the appearance time, character number and time stamp of the characters in the target video according to the identification of the characters included in each thumbnail image extracted by the extraction module 13, the behavior of the characters, the identification of the object, and the timestamp of each thumbnail. The time of appearance of the behavior and the appearance time of each object;

识别模块14还用于对提取模块13提取的多张缩略图进行文字识别，得到第一文本信息；The recognition module 14 is also used to perform character recognition on the plurality of thumbnails extracted by the extraction module 13 to obtain the first text information;

识别模块14还用于对目标视频的音频流进行语音识别，得到第二文本信息；需要说明的是，本实施例的识别模块14根据其功能也可以拆分为多个功能单元，在此不再逐一划分。The recognition module 14 is also used for performing speech recognition on the audio stream of the target video to obtain the second text information; Then divide one by one.

提取模块13还用于从第一文本信息和第二文本信息中提取出现频次最高的数个关键词。The extraction module 13 is further configured to extract several keywords with the highest frequency of occurrence from the first text information and the second text information.

本实施例的视频中的广告投放装置，通过采用上述模块实现视频中的广告投放的实现原理以及技术效果与上述相关方法实施例的实现相同，详细可以参考上述相关方法实施例的记载，在此不再赘述。The advertisement placement device in the video of this embodiment, by using the above modules to realize the realization principle and technical effect of the advertisement placement in the video, is the same as the implementation of the above-mentioned related method embodiments. For details, please refer to the records of the above-mentioned related method embodiments, here No longer.

图4为本发明的计算机设备实施例的结构图。如图4所示，本实施例的计算机设备，包括：一个或多个处理器30，以及存储器40，存储器40用于存储一个或多个程序，当存储器40中存储的一个或多个程序被一个或多个处理器30执行，使得一个或多个处理器30实现如上图1所示实施例的视频中的广告投放方法。图4所示实施例中以包括多个处理器30为例。FIG. 4 is a structural diagram of an embodiment of a computer device of the present invention. As shown in FIG. 4, the computer device of this embodiment includes: one or more processors 30, and a memory 40, and the memory 40 is used to store one or more programs. When the one or more programs stored in the memory 40 are One or more processors 30 are executed, so that one or more processors 30 implement the method for placing advertisements in a video in the embodiment shown in FIG. 1 above. In the embodiment shown in FIG. 4 , a plurality of processors 30 are taken as an example.

例如，图5为本发明提供的一种计算机设备的示例图。图5示出了适于用来实现本发明实施方式的示例性计算机设备12a的框图。图5显示的计算机设备12a仅仅是一个示例，不应对本发明实施例的功能和使用范围带来任何限制。For example, FIG. 5 is an example diagram of a computer device provided by the present invention. Figure 5 shows a block diagram of an exemplary computer device 12a suitable for use in implementing embodiments of the present invention. The computer device 12a shown in FIG. 5 is only an example, and should not limit the functions and scope of use of this embodiment of the present invention.

如图5所示，计算机设备12a以通用计算设备的形式表现。计算机设备12a的组件可以包括但不限于：一个或者多个处理器16a，系统存储器28a，连接不同系统组件(包括系统存储器28a和处理器16a)的总线18a。As shown in FIG. 5, computer device 12a takes the form of a general-purpose computing device. Components of computer device 12a may include, but are not limited to, one or more processors 16a, system memory 28a, bus 18a connecting various system components including system memory 28a and processor 16a.

总线18a表示几类总线结构中的一种或多种，包括存储器总线或者存储器控制器，外围总线，图形加速端口，处理器或者使用多种总线结构中的任意总线结构的局域总线。举例来说，这些体系结构包括但不限于工业标准体系结构(ISA)总线，微通道体系结构(MAC)总线，增强型ISA总线、视频电子标准协会(VESA)局域总线以及外围组件互连(PCI)总线。Bus 18a represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus structures. These architectures include, by way of example, but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect ( PCI) bus.

计算机设备12a典型地包括多种计算机系统可读介质。这些介质可以是任何能够被计算机设备12a访问的可用介质，包括易失性和非易失性介质，可移动的和不可移动的介质。Computer device 12a typically includes a variety of computer system readable media. Such media can be any available media that can be accessed by computer device 12a and includes both volatile and nonvolatile media, removable and non-removable media.

系统存储器28a可以包括易失性存储器形式的计算机系统可读介质，例如随机存取存储器(RAM)30a和/或高速缓存存储器32a。计算机设备12a可以进一步包括其它可移动/不可移动的、易失性/非易失性计算机系统存储介质。仅作为举例，存储系统34a可以用于读写不可移动的、非易失性磁介质(图5未显示，通常称为“硬盘驱动器”)。尽管图5中未示出，可以提供用于对可移动非易失性磁盘(例如“软盘”)读写的磁盘驱动器，以及对可移动非易失性光盘(例如CD-ROM,DVD-ROM或者其它光介质)读写的光盘驱动器。在这些情况下，每个驱动器可以通过一个或者多个数据介质接口与总线18a相连。系统存储器28a可以包括至少一个程序产品，该程序产品具有一组(例如至少一个)程序模块，这些程序模块被配置以执行本发明上述图1-图3各实施例的功能。System memory 28a may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30a and/or cache memory 32a. The computer device 12a may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 34a may be used to read and write to non-removable, non-volatile magnetic media (not shown in FIG. 5, commonly referred to as a "hard drive"). Although not shown in FIG. 5, a disk drive for reading and writing to removable nonvolatile disks (e.g., "floppy disks") may be provided, as well as for removable nonvolatile optical disks (e.g., CD-ROM, DVD-ROM or other optical media) CD-ROM drive. In these cases, each drive may be connected to bus 18a via one or more data media interfaces. The system memory 28a may include at least one program product having a set (eg, at least one) of program modules configured to perform the functions of the above-described embodiments of the present invention in FIGS. 1-3 .

具有一组(至少一个)程序模块42a的程序/实用工具40a，可以存储在例如系统存储器28a中，这样的程序模块42a包括——但不限于——操作系统、一个或者多个应用程序、其它程序模块以及程序数据，这些示例中的每一个或某种组合中可能包括网络环境的实现。程序模块42a通常执行本发明所描述的上述图1-图3各实施例中的功能和/或方法。A program/utility 40a having a set (at least one) of program modules 42a may be stored, for example, in system memory 28a, such program modules 42a including - but not limited to - an operating system, one or more application programs, other Program modules, as well as program data, each or some combination of these examples may include implementations of network environments. The program module 42a generally executes the functions and/or methods described in the present invention in the above-mentioned embodiments of FIG. 1-FIG. 3 .

计算机设备12a也可以与一个或多个外部设备14a(例如键盘、指向设备、显示器24a等)通信，还可与一个或者多个使得用户能与该计算机设备12a交互的设备通信，和/或与使得该计算机设备12a能与一个或多个其它计算设备进行通信的任何设备(例如网卡，调制解调器等等)通信。这种通信可以通过输入/输出(I/O)接口22a进行。并且，计算机设备12a还可以通过网络适配器20a与一个或者多个网络(例如局域网(LAN)，广域网(WAN)和/或公共网络，例如因特网)通信。如图所示，网络适配器20a通过总线18a与计算机设备12a的其它模块通信。应当明白，尽管图中未示出，可以结合计算机设备12a使用其它硬件和/或软件模块，包括但不限于：微代码、设备驱动器、冗余处理器、外部磁盘驱动阵列、RAID系统、磁带驱动器以及数据备份存储系统等。The computer device 12a may also communicate with one or more external devices 14a (e.g., a keyboard, pointing device, display 24a, etc.), may also communicate with one or more devices that enable a user to interact with the computer device 12a, and/or communicate with Any device (eg, network card, modem, etc.) that enables the computing device 12a to communicate with one or more other computing devices. Such communication may occur through input/output (I/O) interface 22a. Also, computer device 12a may communicate with one or more networks (eg, local area network (LAN), wide area network (WAN) and/or a public network, such as the Internet) via network adapter 20a. As shown, network adapter 20a communicates with other modules of computer device 12a via bus 18a. It should be appreciated that although not shown, other hardware and/or software modules may be used in conjunction with computer device 12a, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives And data backup storage system, etc.

处理器16a通过运行存储在系统存储器28a中的程序，从而执行各种功能应用以及数据处理，例如实现上述实施例所示的视频中的广告投放方法。The processor 16a executes various functional applications and data processing by running the programs stored in the system memory 28a, such as implementing the method for placing advertisements in videos as shown in the above-mentioned embodiments.

本发明还提供一种计算机可读介质，其上存储有计算机程序，该程序被处理器执行时实现如上述实施例所示的视频中的广告投放方法。The present invention also provides a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, the method for placing advertisements in a video as shown in the above-mentioned embodiments is realized.

本实施例的计算机可读介质可以包括上述图5所示实施例中的系统存储器28a中的RAM30a、和/或高速缓存存储器32a、和/或存储系统34a。The computer-readable medium in this embodiment may include the RAM 30a in the system memory 28a in the embodiment shown in FIG. 5, and/or the cache memory 32a, and/or the storage system 34a.

随着科技的发展，计算机程序的传播途径不再受限于有形介质，还可以直接从网络下载，或者采用其他方式获取。因此，本实施例中的计算机可读介质不仅可以包括有形的介质，还可以包括无形的介质。With the development of science and technology, the transmission channels of computer programs are no longer limited to tangible media, and can also be directly downloaded from the Internet or obtained in other ways. Therefore, the computer-readable medium in this embodiment may include not only tangible media, but also intangible media.

本实施例的计算机可读介质可以采用一个或多个计算机可读的介质的任意组合。计算机可读介质可以是计算机可读信号介质或者计算机可读存储介质。计算机可读存储介质例如可以是——但不限于——电、磁、光、电磁、红外线、或半导体的系统、装置或器件，或者任意以上的组合。计算机可读存储介质的更具体的例子(非穷举的列表)包括：具有一个或多个导线的电连接、便携式计算机磁盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、光纤、便携式紧凑磁盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。在本文件中，计算机可读存储介质可以是任何包含或存储程序的有形介质，该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。The computer-readable medium of this embodiment may use any combination of one or more computer-readable mediums. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (non-exhaustive list) of computer readable storage media include: electrical connections with one or more leads, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), Erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

计算机可读的信号介质可以包括在基带中或者作为载波一部分传播的数据信号，其中承载了计算机可读的程序代码。这种传播的数据信号可以采用多种形式，包括——但不限于——电磁信号、光信号或上述的任意合适的组合。计算机可读的信号介质还可以是计算机可读存储介质以外的任何计算机可读介质，该计算机可读介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。A computer readable signal medium may include a data signal carrying computer readable program code in baseband or as part of a carrier wave. Such propagated data signals may take many forms, including - but not limited to - electromagnetic signals, optical signals, or any suitable combination of the foregoing. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. .

计算机可读介质上包含的程序代码可以用任何适当的介质传输，包括——但不限于——无线、电线、光缆、RF等等，或者上述的任意合适的组合。Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including - but not limited to - wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

可以以一种或多种程序设计语言或其组合来编写用于执行本发明操作的计算机程序代码，所述程序设计语言包括面向对象的程序设计语言—诸如Java、Smalltalk、C++，还包括常规的过程式程序设计语言—诸如”C”语言或类似的程序设计语言。程序代码可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络——包括局域网(LAN)或广域网(WAN)—连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。Computer program code for carrying out the operations of the present invention may be written in one or more programming languages, or combinations thereof, including object-oriented programming languages—such as Java, Smalltalk, C++, and conventional Procedural programming language—such as "C" or a similar programming language. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (such as through an Internet service provider). Internet connection).

在本发明所提供的几个实施例中，应该理解到，所揭露的系统，装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式。In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation.

所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用硬件加软件功能单元的形式实现。In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware, or in the form of hardware plus software functional units.

上述以软件功能单元的形式实现的集成的单元，可以存储在一个计算机可读取存储介质中。上述软件功能单元存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)或处理器(processor)执行本发明各个实施例所述方法的部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。The above-mentioned integrated units implemented in the form of software functional units may be stored in a computer-readable storage medium. The above-mentioned software functional units are stored in a storage medium, and include several instructions to make a computer device (which may be a personal computer, server, or network device, etc.) or a processor (processor) execute the methods described in various embodiments of the present invention. partial steps. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other various media that can store program codes. .

以上所述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，所做的任何修改、等同替换、改进等，均应包含在本发明保护的范围之内。The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the present invention. within the scope of protection.

Claims (18)

Translated fromChinese

1.一种视频中的广告投放方法，其特征在于，所述方法包括：1. A method for advertising in a video, characterized in that the method comprises:接收到视频播放器发送的携带所述目标视频标识的广告请求之后，根据所述目标视频标识获取对应的目标视频的视频结构化信息；After receiving the advertisement request carrying the target video identifier sent by the video player, acquiring the video structured information of the corresponding target video according to the target video identifier;根据所述目标视频的视频结构化信息以及预先设置的广告库，生成所述目标视频对应的广告数组；Generate an array of advertisements corresponding to the target video according to the video structured information of the target video and a preset advertisement library;向所述视频播放器发送所述广告数组，以供所述视频播放器根据所述广告数组投放各所述目标广告。Sending the advertisement array to the video player for the video player to deliver each of the target advertisements according to the advertisement array.2.根据权利要求1所述的方法，其特征在于，根据所述视频结构化信息以及预先设置的广告库，生成所述目标视频对应的广告数组，具体包括：2. The method according to claim 1, characterized in that, according to the structured information of the video and the preset advertisement storehouse, an array of advertisements corresponding to the target video is generated, specifically comprising:根据所述视频结构化信息从所述广告库中获取与所述目标视频关联的至少一条目标广告的信息；Obtaining at least one target advertisement information associated with the target video from the advertisement library according to the video structured information;根据所述视频结构化信息和对应的所述至少一条目标广告的信息，生成广告数组。An advertisement array is generated according to the video structural information and the corresponding information of the at least one target advertisement.3.根据权利要求2所述的方法，其特征在于，所述视频结构化信息包括所述目标视频的题材类型以及所述目标视频的长度；根据所述视频结构化信息从所述广告库中获取与所述目标视频关联的至少一条目标广告的信息，具体包括：3. The method according to claim 2, wherein the video structured information includes the subject type of the target video and the length of the target video; Obtain information about at least one target advertisement associated with the target video, specifically including:根据所述题材类型，从所述广告库中获取与所述题材类型对应的多条第一候选广告的信息；According to the subject type, acquiring information of a plurality of first candidate advertisements corresponding to the subject type from the advertisement library;根据所述目标视频的长度和预设的广告间隔时长，确定所述题材类型对应的第一目标广告的条数m；所述m为大于或者等于1的整数；According to the length of the target video and the preset advertisement interval duration, determine the number m of the first target advertisement corresponding to the theme type; the m is an integer greater than or equal to 1;根据各所述第一候选广告的标价，从所述多条第一候选广告的信息中获取m条第一目标广告的信息；Obtaining information on m pieces of first target advertisements from the information on the multiple pieces of first candidate advertisements according to the bids of each of the first candidate advertisements;对应地，根据所述视频结构化信息和对应的所述至少一条目标广告的信息，生成广告数组，具体包括：Correspondingly, according to the video structured information and the corresponding information of the at least one target advertisement, an advertisement array is generated, specifically including:根据所述m条第一目标广告中的各所述第一目标广告的标价、所述目标视频的长度、所述广告间隔时长和所述第一目标广告的数量，确定各所述第一目标广告的投放方式以及投放时间；According to the price of each of the first target advertisements in the m first target advertisements, the length of the target video, the length of the advertisement interval and the number of the first target advertisements, determine each of the first target advertisements How and when advertisements will be delivered;将各所述第一目标广告的标识、各所述第一目标广告的投放方式及所述投放时间以及各所述第一目标广告的地址，存入所述广告数组中。Store the identifier of each of the first target advertisements, the delivery method and the delivery time of each of the first target advertisements, and the address of each of the first target advertisements into the advertisement array.4.根据权利要求3所述的方法，其特征在于，所述视频化结构信息中还包括所述目标视频中出现频次最高的数个关键词；根据所述题材类型，从所述广告库中获取相关的多条第一候选广告的信息，具体包括：4. The method according to claim 3, wherein the video structure information also includes several keywords with the highest frequency of occurrence in the target video; according to the subject matter type, from the advertisement library Obtain information about multiple first candidate advertisements, including:根据所述题材类型，并结合所述数个关键词，从所述广告库中获取与所述题材类型对应、且与所述数个关键词相关的多条所述第一候选广告的信息。According to the theme type and in combination with the several keywords, information about a plurality of first candidate advertisements corresponding to the theme type and related to the several keywords is obtained from the advertisement library.5.根据权利要求2所述的方法，其特征在于，所述视频结构化信息包括所述目标视频中的人物标识以及所述人物标识对应的人物的出现时间；5. The method according to claim 2, wherein the structured video information includes a person identification in the target video and an appearance time of a person corresponding to the person identification;根据所述视频结构化信息从所述广告库中获取与所述目标视频关联的至少一条目标广告的信息，具体包括：Obtaining at least one target advertisement information associated with the target video from the advertisement library according to the video structured information, specifically including:根据所述人物标识，从所述广告库中获取与所述人物标识对应人物相关的多条第二候选广告的信息；Acquiring, from the advertisement library, information about a plurality of second candidate advertisements related to the person corresponding to the person identifier according to the person identifier;根据所述人物的所述出现时间以及预设的广告投放时长，确定所述人物对应的第二目标广告的条数n；所述n为大于或者等于1的整数；Determine the number n of second target advertisements corresponding to the character according to the appearance time of the character and the preset advertisement delivery duration; the n is an integer greater than or equal to 1;根据各所述第二候选广告的标价，从所述多条第二候选广告中获取n条第二目标广告的信息；Acquiring information about n second target advertisements from the plurality of second candidate advertisements according to the bids of each of the second candidate advertisements;对应地，根据所述视频结构化信息和对应的所述至少一条目标广告的信息，生成广告数组，具体包括：Correspondingly, according to the video structured information and the corresponding information of the at least one target advertisement, an advertisement array is generated, specifically including:根据所述n条第二目标广告中各所述第二目标广告的标价、所述人物的出现时间和预设的所述广告投放时长，确定各所述第二目标广告的投放时间；Determining the delivery time of each of the second target advertisements according to the price of each of the n second target advertisements, the appearance time of the characters, and the preset advertisement delivery time;将各所述第二目标广告的标识、各所述第二目标广告的投放方式及所述投放时间以及各所述第二目标广告的地址，存入所述广告数组中。Store the identifier of each of the second target advertisements, the delivery method and the delivery time of each of the second target advertisements, and the address of each of the second target advertisements into the advertisement array.6.根据权利要求5所述的方法，其特征在于，所述视频结构化信息还包括所述目标视频中的所述人物的行为；6. The method according to claim 5, wherein the video structured information also includes the behavior of the characters in the target video;根据所述人物标识，从所述广告库中获取所述人物标识对应人物相关的多条第二候选广告的信息，具体包括：According to the person identification, the information of a plurality of second candidate advertisements related to the person corresponding to the person identification is obtained from the advertisement database, specifically including:根据所述人物标识，并结合所述人物的行为，从所述广告库中获取与所述人物以及所述人物的行为相关的多条所述第二候选广告的信息。Based on the character identification and in combination with the character's behavior, information about the multiple second candidate advertisements related to the character and the character's behavior is acquired from the advertisement library.7.根据权利要求2所述的方法，其特征在于，所述视频结构化信息包括所述目标视频中的物体标识以及所述物体标识对应的物体的出现时间；7. The method according to claim 2, wherein the structured video information includes an object identifier in the target video and an appearance time of an object corresponding to the object identifier;根据所述视频结构化信息从所述广告库中获取与所述目标视频关联的至少一条目标广告的信息，具体包括：Obtaining at least one target advertisement information associated with the target video from the advertisement library according to the video structured information, specifically including:根据所述物体标识，从所述广告库中获取所述物体标识对应物体相关的多条第三候选广告的信息；According to the object identifier, acquiring information of multiple third candidate advertisements related to the object corresponding to the object identifier from the advertisement library;根据所述物体的出现时间以及预设的广告投放时长，确定第三目标广告的条数p；所述p为大于或者等于1的整数；Determine the number p of the third target advertisement according to the appearance time of the object and the preset advertisement delivery duration; the p is an integer greater than or equal to 1;根据各所述第三候选广告的标价，从所述多条第三候选广告中获取p条第三目标广告的信息；Acquiring information about p pieces of third target advertisements from the plurality of pieces of third candidate advertisements according to the price tag of each of the third candidate advertisements;对应地，根据所述视频结构化信息和对应的所述至少一条目标广告的信息，生成广告数组，具体包括：Correspondingly, according to the video structured information and the corresponding information of the at least one target advertisement, an advertisement array is generated, specifically including:根据所述p条第三目标广告中各所述第三目标广告的标价、所述物体的出现时间和预设的所述投放时长，确定各所述第三目标广告的投放时间；Determine the delivery time of each of the third target advertisements according to the price of each of the p third target advertisements, the appearance time of the object, and the preset delivery time;将各所述第三目标广告的标识、各所述第三目标广告的投放方式及所述投放时间以及各所述第三目标广告的地址，存入所述广告数组中。Store the identifier of each of the third target advertisements, the delivery method and the delivery time of each of the third target advertisements, and the address of each of the third target advertisements into the advertisement array.8.根据权利要求1或2所述的方法，其特征在于，接收到视频播放器发送的携带所述目标视频标识的广告请求之后，根据所述目标视频标识获取对应的目标视频的视频结构化信息之前，所述方法还包括：8. The method according to claim 1 or 2, characterized in that, after receiving the advertisement request carrying the target video identifier sent by the video player, the video structure of the corresponding target video is obtained according to the target video identifier Before the information, the method also includes:从所述目标视频的视频流中按照时间先后顺序提取多张缩略图；各张所述缩略图按照对应的时间戳来标识；Extracting a plurality of thumbnail images in chronological order from the video stream of the target video; each thumbnail image is identified according to a corresponding timestamp;对所述多张缩略图进行题材类型识别，得到所述目标视频的题材类型；Carry out theme type identification to described multiple thumbnails, obtain the theme type of described target video;对各张所述缩略图进行人物识别、行为识别和物体识别，得到各张所述缩略图中包括的人物的标识、所述人物的行为以及物体的标识；Performing character recognition, behavior recognition and object recognition on each of the thumbnail images to obtain the identity of the person included in each of the thumbnail images, the behavior of the person and the identity of the object;根据各张所述缩略图中包括的人物的标识、所述人物的行为、物体的标识以及各张所述缩略图的时间戳，获取所述目标视频中的所述人物的出现时间、所述人物的行为的出现时间以及各所述物体的出现时间；Acquire the appearance time of the character in the target video, the the timing of the behavior of the characters and the timing of each of the objects described;对所述多张缩略图进行文字识别，得到第一文本信息；performing text recognition on the plurality of thumbnail images to obtain first text information;对所述目标视频的音频流进行语音识别，得到第二文本信息；performing speech recognition on the audio stream of the target video to obtain second text information;从所述第一文本信息和所述第二文本信息中提取出现频次最高的数个关键词。Several keywords with the highest frequency of occurrence are extracted from the first text information and the second text information.9.一种视频中的广告投放装置，其特征在于，所述装置包括：9. A device for advertising in a video, characterized in that the device comprises:获取模块，用于接收到视频播放器发送的携带所述目标视频标识的广告请求之后，根据所述目标视频标识获取对应的目标视频的视频结构化信息；An acquisition module, configured to acquire the video structured information of the corresponding target video according to the target video ID after receiving the advertisement request sent by the video player and carrying the target video ID;生成模块，用于根据所述目标视频的视频结构化信息以及预先设置的广告库，生成所述目标视频对应的广告数组；A generating module, configured to generate an array of advertisements corresponding to the target video according to the video structured information of the target video and a preset advertisement library;发送模块，用于向所述视频播放器发送所述广告数组，以供所述视频播放器根据所述广告数组投放各所述目标广告。A sending module, configured to send the advertisement array to the video player, so that the video player can deliver each of the target advertisements according to the advertisement array.10.根据权利要求9所述的装置，其特征在于，所述生成模块，具体包括：10. The device according to claim 9, wherein the generating module specifically comprises:广告获取单元，用于根据所述视频结构化信息从所述广告库中获取与所述目标视频关联的至少一条目标广告的信息；an advertisement obtaining unit, configured to obtain information of at least one target advertisement associated with the target video from the advertisement library according to the video structured information;数组生成单元，用于根据所述视频结构化信息和对应的所述至少一条目标广告的信息，生成广告数组。An array generating unit, configured to generate an array of advertisements according to the video structured information and the corresponding information of the at least one target advertisement.11.根据权利要求10所述的装置，其特征在于，所述视频结构化信息包括所述目标视频的题材类型以及所述目标视频的长度；所述广告获取单元，具体用于：11. The device according to claim 10, wherein the structured video information includes the subject type of the target video and the length of the target video; the advertisement obtaining unit is specifically used for:根据所述题材类型，从所述广告库中获取与所述题材类型对应的多条第一候选广告的信息；According to the subject type, acquiring information of a plurality of first candidate advertisements corresponding to the subject type from the advertisement library;根据所述目标视频的长度和预设的广告间隔时长，确定所述题材类型对应的第一目标广告的条数m；所述m为大于或者等于1的整数；According to the length of the target video and the preset advertisement interval duration, determine the number m of the first target advertisement corresponding to the theme type; the m is an integer greater than or equal to 1;根据各所述第一候选广告的标价，从所述多条第一候选广告的信息中获取m条第一目标广告的信息；Obtaining information on m pieces of first target advertisements from the information on the multiple pieces of first candidate advertisements according to the bids of each of the first candidate advertisements;对应地，所述数组生成单元，具体用于：Correspondingly, the array generating unit is specifically used for:根据所述m条第一目标广告中的各所述第一目标广告的标价、所述目标视频的长度、所述广告间隔时长和所述第一目标广告的数量，确定各所述第一目标广告的投放方式以及投放时间；According to the price of each of the first target advertisements in the m first target advertisements, the length of the target video, the length of the advertisement interval and the number of the first target advertisements, determine each of the first target advertisements How and when advertisements will be delivered;将各所述第一目标广告的标识、各所述第一目标广告的投放方式及所述投放时间以及各所述第一目标广告的地址，存入所述广告数组中。Store the identifier of each of the first target advertisements, the delivery method and the delivery time of each of the first target advertisements, and the address of each of the first target advertisements into the advertisement array.12.根据权利要求11所述的装置，其特征在于，所述视频化结构信息中还包括所述目标视频中出现频次最高的数个关键词；所述广告获取单元，具体用于：根据所述题材类型，并结合所述数个关键词，从所述广告库中获取与所述题材类型对应、且与所述数个关键词相关的多条所述第一候选广告的信息。12. The device according to claim 11, wherein the video structure information also includes several keywords with the highest frequency of occurrence in the target video; the advertisement acquisition unit is specifically configured to: The subject type, combined with the several keywords, is used to obtain a plurality of pieces of information of the first candidate advertisements corresponding to the subject type and related to the several keywords from the advertisement library.13.根据权利要求10所述的装置，其特征在于，所述视频结构化信息包括所述目标视频中的人物标识以及所述人物标识对应的人物的出现时间；13. The device according to claim 10, wherein the structured video information includes a character identification in the target video and an appearance time of a character corresponding to the character identification;所述广告获取单元，具体用于：The advertisement acquiring unit is specifically used for:根据所述人物标识，从所述广告库中获取与所述人物标识对应人物相关的多条第二候选广告的信息；Acquiring, from the advertisement library, information about a plurality of second candidate advertisements related to the person corresponding to the person identifier according to the person identifier;根据所述人物的所述出现时间以及预设的广告投放时长，确定所述人物对应的第二目标广告的条数n；所述n为大于或者等于1的整数；Determine the number n of second target advertisements corresponding to the character according to the appearance time of the character and the preset advertisement delivery duration; the n is an integer greater than or equal to 1;根据各所述第二候选广告的标价，从所述多条第二候选广告中获取n条第二目标广告的信息；Acquiring information about n second target advertisements from the plurality of second candidate advertisements according to the bids of each of the second candidate advertisements;对应地，所述数组生成单元，具体用于：Correspondingly, the array generating unit is specifically used for:根据所述n条第二目标广告中各所述第二目标广告的标价、所述人物的出现时间和预设的所述广告投放时长，确定各所述第二目标广告的投放时间；Determining the delivery time of each of the second target advertisements according to the price of each of the n second target advertisements, the appearance time of the characters, and the preset advertisement delivery time;将各所述第二目标广告的标识、各所述第二目标广告的投放方式及所述投放时间以及各所述第二目标广告的地址，存入所述广告数组中。Store the identifier of each of the second target advertisements, the delivery method and the delivery time of each of the second target advertisements, and the address of each of the second target advertisements into the advertisement array.14.根据权利要求13所述的装置，其特征在于，所述视频结构化信息还包括所述目标视频中的所述人物的行为；14. The device according to claim 13, wherein the structured video information further includes behaviors of the characters in the target video;所述广告获取单元，具体用于：The advertisement acquiring unit is specifically used for:根据所述人物标识，并结合所述人物的行为，从所述广告库中获取与所述人物以及所述人物的行为相关的多条所述第二候选广告的信息。Based on the character identification and in combination with the character's behavior, information about the multiple second candidate advertisements related to the character and the character's behavior is acquired from the advertisement library.15.根据权利要求10所述的装置，其特征在于，所述视频结构化信息包括所述目标视频中的物体标识以及所述物体标识对应的物体的出现时间；15. The device according to claim 10, wherein the structured video information includes an object identifier in the target video and an appearance time of an object corresponding to the object identifier;所述广告获取单元，具体用于：The advertisement acquiring unit is specifically used for:根据所述物体标识，从所述广告库中获取所述物体标识对应物体相关的多条第三候选广告的信息；According to the object identifier, acquiring information of multiple third candidate advertisements related to the object corresponding to the object identifier from the advertisement library;根据所述物体的出现时间以及预设的广告投放时长，确定第三目标广告的条数p；所述p为大于或者等于1的整数；Determine the number p of the third target advertisement according to the appearance time of the object and the preset advertisement delivery duration; the p is an integer greater than or equal to 1;根据各所述第三候选广告的标价，从所述多条第三候选广告中获取p条第三目标广告的信息；Acquiring information about p pieces of third target advertisements from the plurality of pieces of third candidate advertisements according to the price tag of each of the third candidate advertisements;对应地，所述数组生成单元，具体用于：Correspondingly, the array generating unit is specifically used for:根据所述p条第三目标广告中各所述第三目标广告的标价、所述物体的出现时间和预设的所述投放时长，确定各所述第三目标广告的投放时间；Determine the delivery time of each of the third target advertisements according to the price of each of the p third target advertisements, the appearance time of the object, and the preset delivery time;将各所述第三目标广告的标识、各所述第三目标广告的投放方式及所述投放时间以及各所述第三目标广告的地址，存入所述广告数组中。Store the identifier of each of the third target advertisements, the delivery method and the delivery time of each of the third target advertisements, and the address of each of the third target advertisements into the advertisement array.16.根据权利要求9或10所述的装置，其特征在于，所述装置还包括：16. The device according to claim 9 or 10, further comprising:提取模块，用于从所述目标视频的视频流中按照时间先后顺序提取多张缩略图；各张所述缩略图按照对应的时间戳来标识；An extracting module, configured to extract a plurality of thumbnails in chronological order from the video stream of the target video; each thumbnail is identified according to a corresponding timestamp;识别模块，用于对所述多张缩略图进行题材类型识别，得到所述目标视频的题材类型；The identification module is used to identify the theme type of the plurality of thumbnails to obtain the theme type of the target video;所述识别模块，还用于对各张所述缩略图进行人物识别、行为识别和物体识别，得到各张所述缩略图中包括的人物的标识、所述人物的行为以及物体的标识；The recognition module is also used to perform character recognition, behavior recognition, and object recognition on each of the thumbnail images to obtain the identity of the person included in each of the thumbnails, the behavior of the character, and the identity of the object;所述获取模块，还用于根据各张所述缩略图中包括的人物的标识、所述人物的行为、物体的标识以及各张所述缩略图的时间戳，获取所述目标视频中的所述人物的出现时间、所述人物的行为的出现时间以及各所述物体的出现时间；The acquisition module is further configured to acquire all the characters in the target video according to the identification of the characters included in each of the thumbnails, the behavior of the characters, the identification of the object, and the time stamp of each of the thumbnails. the time of appearance of said character, the time of appearance of the behavior of said character, and the time of appearance of each said object;所述识别模块，还用于对所述多张缩略图进行文字识别，得到第一文本信息；The recognition module is also used to perform character recognition on the plurality of thumbnail images to obtain first text information;所述识别模块，还用于对所述目标视频的音频流进行语音识别，得到第二文本信息；The recognition module is further configured to perform speech recognition on the audio stream of the target video to obtain second text information;所述提取模块，还用于从所述第一文本信息和所述第二文本信息中提取出现频次最高的数个关键词。The extraction module is further configured to extract several keywords with the highest frequency of occurrence from the first text information and the second text information.17.一种计算机设备，其特征在于，所述设备包括：17. A computer device, characterized in that said device comprises:一个或多个处理器；one or more processors;存储器，用于存储一个或多个程序；memory for storing one or more programs;当所述一个或多个程序被所述一个或多个处理器执行，使得所述一个或多个处理器实现如权利要求1-8中任一所述的方法。When the one or more programs are executed by the one or more processors, the one or more processors are made to implement the method according to any one of claims 1-8.18.一种计算机可读介质，其上存储有计算机程序，其特征在于，该程序被处理器执行时实现如权利要求1-8中任一所述的方法。18. A computer-readable medium, on which a computer program is stored, wherein, when the program is executed by a processor, the method according to any one of claims 1-8 is realized.