Advertisement data updating method and system

Abstract

The present invention relates to the technical field of video advertisement insertion, and specifically discloses a method and system for updating advertisement data, the method comprising obtaining a video to be processed and its playback parameters at different playback ends, determining the exposure of each frame image according to the playback parameters at all playback ends; comparing the advertisement theme and the video theme to determine the advertisement abruptness; receiving the exposure requirement input by the user, selecting the advertisement insertion position according to the exposure requirement and the advertisement abruptness, and inserting the advertisement data into the video to be processed; transmitting the video to be processed containing the advertisement data to the playback end, receiving feedback information responded by the playback end, and adjusting the advertisement insertion process according to the feedback information. The present invention obtains the playback parameters of the video to be processed, analyzes the playback parameters, determines the exposure, and on this basis, for each advertisement, compares the advertisement with the video, determines the abruptness, combines the two parameters of exposure and abruptness, and updates the advertisement insertion position in real time, with extremely high flexibility.

Description

Technical Field

[0001] The present invention relates to the technical field of video advertisement insertion, and in particular to an advertisement data updating method and system. Background Art

[0002] With the development of network technology and the popularization of smart devices, video has gradually become a mainstream information carrier. Most users of smart devices receive a large amount of video information every day. Since video information is highly intuitive, inserting advertisements in video information can achieve better advertising effects.

[0003] However, the existing video advertisement insertion process mostly only inserts the advertisement at the beginning or end of the video, or at most inserts some advertisements at some time in the middle. Once inserted, it will hardly be changed, and the flexibility is very low. How to provide a more flexible advertisement data update solution to improve the advertisement effect is the technical problem that the technical solution of the present invention aims to solve. Summary of the invention

[0004] The object of the present invention is to provide an advertisement data updating method and system to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for updating advertisement data, the method comprising:

[0007] Obtaining the video to be processed and its playback parameters at different playback ends, and determining the exposure of each frame of image according to the playback parameters at all playback ends; the playback parameter is the playback speed of each frame of image;

[0008] Perform content recognition on the video to be processed and extract the video theme;

[0009] When receiving the advertisement data, the advertisement data is subjected to content recognition, the advertisement theme of the advertisement data is extracted, the advertisement theme and the video theme are compared, and the advertisement abruptness is determined; wherein, in the process of performing content recognition on the processed video and advertisement data, an AI module is introduced; the advertisement theme and the video theme are both phrases;

[0010] Receive the exposure requirement input by the user, select the advertisement insertion position according to the exposure requirement and the advertisement abruptness, and insert the advertisement data into the video to be processed;

[0011] The video to be processed containing the advertisement data is transmitted to the playback end, feedback information of the playback end is received, and the advertisement insertion process is adjusted according to the feedback information.

[0012] As a further solution of the present invention: the step of obtaining the video to be processed and its playback parameters at different playback ends, and determining the exposure of each frame image according to the playback parameters at all playback ends includes:

[0013] Transmit the video to be processed to the playback end, and send a permission acquisition request to the playback end;

[0014] After receiving the permission granted by the playback end, the playback end obtains the playback time of each image in the video to be processed, and calculates the playback speed according to the playback time;

[0015] Read images in the video to be processed one by one, query the playback speed of the images in each playback end, and calculate the average playback speed;

[0016] The exposure is determined according to the average playback speed; the exposure is inversely proportional to the average playback speed.

[0017] As a further solution of the present invention: the process of comparing the advertisement theme with the video theme and determining the advertisement obtrusiveness includes:

[0018] Read the advertisement theme and the video theme, input them into the preset word vector conversion model, and obtain the advertisement word vector group and the video word vector group respectively;

[0019] Pair the elements in the advertisement word vector group and the video word vector group and calculate the vector distance;

[0020] Select the longer vector group from the advertisement word vector group and the video word vector group, and read the minimum vector distance corresponding to each word vector in the longer vector group in turn;

[0021] The mean of the minimum vector distances corresponding to each word vector is calculated, and the ad abruptness is determined based on the inverse ratio of the mean.

[0022] As a further solution of the present invention: the step of receiving the exposure requirement input by the user, selecting the advertisement insertion position according to the exposure requirement and the advertisement obtrusiveness, and inserting the advertisement data into the video to be processed comprises:

[0023] Receive the exposure requirement input by the user, and modify the exposure requirement according to the advertisement abruptness; the exposure requirement is a range; the modification of the exposure requirement by the advertisement abruptness is a reverse modification, and the left endpoint and the right endpoint of the exposure requirement are modified at the same time;

[0024] Select an image whose exposure meets the exposure requirement in the video to be processed;

[0025] Query the position of the selected image as the advertisement insertion position;

[0026] The advertisement data is inserted into the video to be processed based on the advertisement insertion position.

[0027] As a further solution of the present invention: the steps of transmitting the to-be-processed video containing the advertisement data to the playback end, receiving the feedback information of the playback end, and adjusting the advertisement insertion process according to the feedback information include:

[0028] The video to be processed containing the advertising data is transmitted to the playback end, and the playback time of each image is obtained based on the authority granted by the playback end as the feedback playback time;

[0029] Query the playback time of each image when the playback end plays the video to be processed without advertising data, as the initial playback time;

[0030] Compare the feedback playback time and the initial playback time of the same image to determine the degree of disgust;

[0031] The obtrusiveness of the advertisement is adjusted according to the offensiveness.

[0032] As a further solution of the present invention: the method further comprises:

[0033] Determine the total interaction time between the player and the platform based on the player's playback record;

[0034] Determining the impact level of the playback terminal according to the total interaction duration;

[0035] Adjust the dislike level based on the impact level.

[0036] The technical solution of the present invention also provides an advertisement data updating system, the system comprising:

[0037] An exposure determination module is used to obtain the video to be processed and its playback parameters at different playback ends, and determine the exposure of each frame of image according to the playback parameters at all playback ends; the playback parameter is the playback speed of each frame of image;

[0038] The video theme extraction module is used to identify the content of the video to be processed and extract the video theme;

[0039] The abruptness calculation module is used to perform content recognition on the advertisement data when receiving the advertisement data, extract the advertisement theme of the advertisement data, compare the advertisement theme with the video theme, and determine the advertisement abruptness; wherein, in the process of performing content recognition on the video to be processed and the advertisement data, an AI module is introduced; the advertisement theme and the video theme are both phrases;

[0040] An advertisement data insertion module is used to receive the exposure requirement input by the user, select an advertisement insertion position according to the exposure requirement and the advertisement abruptness, and insert the advertisement data into the video to be processed;

[0041] The response information receiving module is used to transmit the to-be-processed video containing advertisement data to the playback end, receive the feedback information of the playback end, and adjust the advertisement insertion process according to the feedback information.

[0042] As a further solution of the present invention: the exposure determination module includes:

[0043] The permission acquisition unit is used to transmit the video to be processed to the playback end and send a permission acquisition request to the playback end;

[0044] A speed calculation unit, configured to obtain the playback time of each image in the video to be processed by the playback end after receiving the permission granted by the playback end, and calculate the playback speed according to the playback time;

[0045] An average speed calculation unit, used to read images in the video to be processed in sequence, query the playback speed of the images in each playback end, and calculate the average playback speed;

[0046] The speed application unit is used to determine the exposure according to the average playback speed; the exposure is inversely proportional to the average playback speed.

[0047] As a further solution of the present invention: the abruptness calculation module includes:

[0048] A topic conversion unit, used to read the advertisement topic and the video topic, input them into a preset word vector conversion model, and obtain an advertisement word vector group and a video word vector group respectively;

[0049] An element pairing unit, used to pair elements in the advertisement word vector group and the video word vector group and calculate the vector distance;

[0050] A distance selection unit is used to select a longer vector group from the advertisement word vector group and the video word vector group, and sequentially read the minimum vector distance corresponding to each word vector in the longer vector group;

[0051] The mean calculation application unit is used to calculate the mean of the minimum vector distance corresponding to each word vector, and determine the advertisement abruptness according to the inverse ratio of the mean.

[0052] As a further solution of the present invention: the advertisement data insertion module includes:

[0053] A demand receiving unit, used to receive the exposure demand input by the user, and to modify the exposure demand according to the advertisement abruptness; the exposure demand is a range; the modification of the exposure demand by the advertisement abruptness is a reverse modification, used to modify the left endpoint and the right endpoint of the exposure demand at the same time;

[0054] An image selection unit, used to select an image whose exposure meets the exposure requirement in the video to be processed;

[0055] A query unit, used to query the position of the selected image as an advertisement insertion position;

[0056] The insertion execution unit is used to insert advertisement data into the video to be processed based on the advertisement insertion position.

[0057] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention obtains the playback parameters of the video to be processed, analyzes the playback parameters, determines the exposure, and on this basis, for each advertisement, compares the advertisement with the video to determine the abruptness. Combining the two parameters of exposure and abruptness, the insertion position of the advertisement can be updated very easily and frequently, with extremely high flexibility, thus preventing aesthetic fatigue. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0059] Figure 1 The flowchart of the advertisement data updating method is shown in FIG.

[0060] Figure 2 This is a structural block diagram of the advertising data update system. DETAILED DESCRIPTION

[0061] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0062] Figure 1 The flowchart of the advertisement data updating method is as follows. In an embodiment of the present invention, an advertisement data updating method is provided, and the method includes:

[0063] Step S100: obtaining a video to be processed and its playback parameters at different playback ends, and determining the exposure of each frame of image according to the playback parameters at all playback ends; the playback parameter is the playback speed of each frame of image;

[0064] The video to be processed is a video that needs to be inserted with advertisements. The executor of this method is generally also the provider of the video. When the video is sent to other playback terminals, the playback parameters of the video to be processed at different playback terminals are obtained based on the permissions granted by the playback terminals. Of course, the playback process is carried out at the playback terminal, and the playback terminal has the right to refuse to share the playback parameters. Then the subsequent steps of the present invention cannot be carried out. For this application, it is assumed that authorization has been obtained from each playback terminal. In other words, this application is limited to authorized playback terminals; the playback parameters of all playback terminals are obtained, and then the exposure of each frame of the image is determined; the playback parameters are the playback speed of each frame of the image.

[0065] Step S200: performing content recognition on the video to be processed and extracting the video theme;

[0066] Content recognition is performed on the video to be processed. Content recognition mainly involves image recognition models and text recognition models, which are used to extract keywords from the video to be processed. The obtained keywords are not unique, but are actually keyword phrases, called video themes.

[0067] Step S300: when receiving the advertisement data, performing content recognition on the advertisement data, extracting the advertisement theme of the advertisement data, comparing the advertisement theme with the video theme, and determining the advertisement abruptness; wherein, in the process of performing content recognition on the video to be processed and the advertisement data, an AI module is introduced; the advertisement theme and the video theme are both phrases;

[0068] The advertising data mentioned in the present application is advertising data in video format, which can be understood as a short video, such as 15 seconds or 30 seconds. After receiving the advertising data, the same content recognition scheme as the video to be processed is used to perform content recognition on it, and the advertising theme is extracted. Similar to the video theme, the advertising theme is also a keyword phrase; the advertising theme and the video theme are compared to determine the degree of differentiation between the advertising theme and the video theme, and the advertising abruptness is determined according to the degree of differentiation; wherein, the greater the degree of differentiation, the greater the advertising abruptness.

[0069] Step S400: receiving the exposure requirement input by the user, selecting an advertisement insertion position according to the exposure requirement and the advertisement abruptness, and inserting the advertisement data into the video to be processed;

[0070] The user refers to the advertising provider. For example, if a company wants to insert its advertisement in a video, the company is the user. The execution subject of this method receives the exposure requirements input by the user, indicating the positions where the advertisements need to be inserted. The greater the exposure, the more important the corresponding position is in the entire video to be processed, and the better the effect of the advertisement can be achieved. However, correspondingly, it will also cause a certain degree of disgust, affecting the playback volume of the video to be processed itself. The degree of influence is represented by the parameter of advertisement abruptness. Therefore, after receiving the exposure requirements input by the user, it is also necessary to consider the impact of advertisement abruptness, comprehensively determine the advertisement insertion position, and then insert the advertisement data into the video to be processed. In layman's terms, the exposure requirements represent the needs of the user, and the advertisement abruptness represents the needs of the video provider. The process of selecting the advertisement insertion position based on the exposure requirements and advertisement abruptness is actually the process of reaching an agreement between the two parties.

[0071] Step S500: transmitting the to-be-processed video containing advertisement data to the playback end, receiving feedback information of the playback end, and adjusting the advertisement insertion process according to the feedback information;

[0072] Finally, the video to be processed containing advertising data is transmitted to the playback end, and feedback information from each playback end is received. The feedback information is the user's opinion on the advertising data. Some users skip it directly, some watch it slowly, and some even click directly to show their disinterest. Based on these feedback information, the abruptness of the advertisement is adjusted (essentially, the viewing experience of the viewers on the playback end towards the video to be processed) and the insertion position of the advertisement is changed, thereby achieving a more coordinated insertion plan.

[0073] Regarding step S100, the step of obtaining the video to be processed and its playback parameters at different playback ends, and determining the exposure of each frame image according to the playback parameters at all playback ends includes:

[0074] Transmit the video to be processed to the playback end, and send a permission acquisition request to the playback end;

[0075] After receiving the permission granted by the playback end, the playback end obtains the playback time of each image in the video to be processed, and calculates the playback speed according to the playback time;

[0076] Read images in the video to be processed one by one, query the playback speed of the images in each playback end, and calculate the average playback speed;

[0077] The exposure is determined according to the average playback speed; the exposure is inversely proportional to the average playback speed.

[0078] In an example of the technical solution of the present invention, the calculation process of the exposure is limited, the video to be processed is transmitted to the playback end, and a permission acquisition request is sent to the playback end. After receiving the permission granted by the playback end, the playback time of each image in the video to be processed is obtained, and the playback speed is calculated according to the playback time. The playback speed means how long it takes to play an image. Therefore, the reciprocal of the playback time can be directly used as the playback speed. In addition, the playback time of the present application is specific to the image, which is different from the existing recognition scheme. It means that for some key video clips, viewers will go back and watch them multiple times. At this time, the playback time of each image will become longer, which also means that the longer the playback time of the image, the more popular it is with viewers.

[0079] Furthermore, the same video to be processed will not be sent to only one playback end, it has a certain large-scale transmission process, that is, each video to be processed has different playback parameters on different playback ends; for this, the present application analyzes each image separately, obtains the playback speed of each image in each playback end, and calculates the average playback speed. The slower the average playback speed, the more popular the corresponding image is, and the higher the exposure is. In short, the exposure is inversely proportional to the average playback speed.

[0080] Regarding step S200, the process of comparing the advertisement theme with the video theme and determining the advertisement obtrusiveness includes:

[0081] Read the advertisement theme and the video theme, input them into the preset word vector conversion model, and obtain the advertisement word vector group and the video word vector group respectively;

[0082] Pair the elements in the advertisement word vector group and the video word vector group and calculate the vector distance;

[0083] Select the longer vector group from the advertisement word vector group and the video word vector group, and read the minimum vector distance corresponding to each word vector in the longer vector group in turn;

[0084] The mean of the minimum vector distances corresponding to each word vector is calculated, and the ad abruptness is determined based on the inverse ratio of the mean.

[0085] Both the advertising theme and the video theme are phrases, and the comparison process is actually the comparison process of two word sets. For this, the present application provides a simplified solution, which vectorizes the words in the advertising theme and the video theme to obtain an advertising word vector group and a video word vector group. This process can be completed with the help of an existing word vector conversion model; the existing word vector conversion model itself takes into account the semantics of words, and the vector distance of synonyms is very small; based on this feature, the longer vector group in the advertising word vector group and the video word vector group is selected, and the word vectors are selected in the vector group in turn, and the word vector closest to it is queried in the other vector group. This process is executed in a loop so that each vector in the longer vector group can obtain a minimum distance. The mean of these minimum distances is calculated to represent the degree of differentiation of the two vector groups. The smaller the mean, the smaller the degree of differentiation, and correspondingly, the smaller the abruptness of the advertisement.

[0086] As for the description of the practical significance of the above process, first select the topic with more words in the advertising topic and the video topic, then traverse each word in the topic, search for the most similar word in the other topic, determine the degree of similarity, and finally calculate the average of the similarity of all words as the degree of similarity between the two topics.

[0087] Regarding step S300, the steps of receiving the exposure requirement input by the user, selecting an advertisement insertion position according to the exposure requirement and the advertisement obtrusiveness, and inserting the advertisement data into the video to be processed include:

[0088] Receive the exposure requirement input by the user, and modify the exposure requirement according to the advertisement abruptness; the exposure requirement is a range; the modification of the exposure requirement by the advertisement abruptness is a reverse modification, and the left endpoint and the right endpoint of the exposure requirement are modified at the same time;

[0089] Select an image whose exposure meets the exposure requirement in the video to be processed;

[0090] Query the position of the selected image as the advertisement insertion position;

[0091] The advertisement data is inserted into the video to be processed based on the advertisement insertion position.

[0092] In an example of the technical solution of the present invention, an exposure requirement input by a user is received. The exposure requirement is a range, for example, 10% to 20%. The reason for using a ratio is that the exposure of different videos is different. The ratio can be used without considering the specific exposure value. When necessary, the ratio can be multiplied by the maximum exposure. Then, the exposure requirement is corrected according to the abruptness of the advertisement. The correction method is to reversely correct the two endpoints. The reverse correction means that the greater the abruptness of the advertisement, the smaller the corrected endpoint value. A feasible method is:

[0093] ; In the formula, is the endpoint value before correction, is the corrected endpoint value, For the obtrusiveness of the advertisement.

[0094] After the exposure requirement correction process is completed, an image whose exposure meets the exposure requirement is selected in the video to be processed, and the position of the selected image is queried as the advertisement insertion position. The advertisement data is inserted into the video to be processed based on the advertisement insertion position. In general, the advertisement insertion position can be used as the advertisement playback position. If the matched advertisement insertion position is not unique, the advertisement insertion position with the maximum exposure is generally selected, which is more user-friendly. Of course, the earliest advertisement insertion position or the last advertisement insertion position can also be selected. The effects obtained at different positions are different. Some videos that are relatively boring in the early stage are suitable for the last advertisement insertion position, while videos that are relatively exciting in the early stage are suitable for the last advertisement insertion position.

[0095] It is worth mentioning that the charges corresponding to different exposure requirements are different. Therefore, different users will choose different exposure requirements.

[0096] Regarding step S500, the steps of transmitting the to-be-processed video containing the advertisement data to the playback end, receiving the feedback information of the playback end, and adjusting the advertisement insertion process according to the feedback information include:

[0097] The video to be processed containing the advertising data is transmitted to the playback end, and the playback time of each image is obtained based on the authority granted by the playback end as the feedback playback time;

[0098] Query the playback time of each image when the playback end plays the video to be processed without advertising data, as the initial playback time;

[0099] Compare the feedback playback time and the initial playback time of the same image to determine the degree of disgust;

[0100] The obtrusiveness of the advertisement is adjusted according to the offensiveness.

[0101] In an example of the technical solution of the present invention, a specific feedback adjustment scheme is provided, which connects the adjustment process with the calculation process of the advertisement abruptness. Specifically, the to-be-processed video containing advertisement data is transmitted to the playback end, and the playback time of each image is obtained based on the authority granted by the playback end as the feedback playback time. The playback end is queried for the playback time of each image when playing the to-be-processed video that does not contain advertisement data, as the initial playback time. The feedback playback time and the initial playback time of the same image are compared to determine the degree of disgust. One of the schemes is to calculate the ratio of the feedback playback time to the initial playback time. The smaller the ratio is, the more unwilling the viewer is to watch the video after the advertisement is inserted. At this time, the degree of disgust is greater. The advertisement abruptness is adjusted according to the degree of disgust, and the advertisement abruptness is proportional to the degree of disgust.

[0102] As a preferred embodiment of the technical solution of the present invention, the method further includes:

[0103] Determine the total interaction time between the player and the platform based on the player's playback record;

[0104] Determining the impact level of the playback terminal according to the total interaction duration;

[0105] Adjust the dislike level based on the impact level.

[0106] In an example of the technical solution of the present invention, a user grading scheme is also provided, which determines the total interaction time between the playback end and the platform based on the playback record of the playback end. The higher the total interaction time, the more important the viewer at the corresponding playback end is to the execution subject of the method. At this time, it is necessary to consider the feedback information at the playback end more and consider this demand more. The method of adjusting the disgust degree satisfies the requirement. The influence level of the playback end is determined based on the total interaction time, and its disgust degree is adjusted based on the influence level. The influence level is proportional to the total interaction time. The higher the influence level, the more important the corresponding playback end is, and the greater the disgust degree after adjustment is. The simplest way is to determine the amplification factor based on the influence level and multiply it by the disgust degree. The amplification factor is proportional to the influence level.

[0107] Figure 2 1 is a structural block diagram of an advertisement data updating system. In an embodiment of the present invention, an advertisement data updating system 10 includes:

[0108] The exposure determination module 11 is used to obtain the video to be processed and its playback parameters at different playback ends, and determine the exposure of each frame of the image according to the playback parameters at all playback ends; the playback parameter is the playback speed of each frame of the image;

[0109] The video theme extraction module 12 is used to perform content recognition on the video to be processed and extract the video theme;

[0110] The abruptness calculation module 13 is used to perform content recognition on the advertisement data when receiving the advertisement data, extract the advertisement theme of the advertisement data, compare the advertisement theme with the video theme, and determine the advertisement abruptness; wherein, in the process of performing content recognition on the video to be processed and the advertisement data, an AI module is introduced; the advertisement theme and the video theme are both phrases;

[0111] The advertisement data insertion module 14 is used to receive the exposure requirement input by the user, select the advertisement insertion position according to the exposure requirement and the advertisement abruptness, and insert the advertisement data into the video to be processed;

[0112] The response information receiving module 15 is used to transmit the to-be-processed video containing the advertisement data to the playback end, receive the feedback information of the playback end, and adjust the advertisement insertion process according to the feedback information.

[0113] Furthermore, the exposure determination module 11 includes:

[0114] The permission acquisition unit is used to transmit the video to be processed to the playback end and send a permission acquisition request to the playback end;

[0115] A speed calculation unit, configured to obtain the playback time of each image in the video to be processed by the playback end after receiving the permission granted by the playback end, and calculate the playback speed according to the playback time;

[0116] An average speed calculation unit, used to read images in the video to be processed in sequence, query the playback speed of the images in each playback end, and calculate the average playback speed;

[0117] The speed application unit is used to determine the exposure according to the average playback speed; the exposure is inversely proportional to the average playback speed.

[0118] Specifically, the abruptness calculation module 13 includes:

[0119] A topic conversion unit, used to read the advertisement topic and the video topic, input them into a preset word vector conversion model, and obtain an advertisement word vector group and a video word vector group respectively;

[0120] An element pairing unit, used to pair elements in the advertisement word vector group and the video word vector group and calculate the vector distance;

[0121] A distance selection unit is used to select a longer vector group from the advertisement word vector group and the video word vector group, and sequentially read the minimum vector distance corresponding to each word vector in the longer vector group;

[0122] The mean calculation application unit is used to calculate the mean of the minimum vector distance corresponding to each word vector, and determine the advertisement abruptness according to the inverse ratio of the mean.

[0123] Furthermore, the advertisement data inserting module 14 includes:

[0124] A demand receiving unit, used to receive the exposure demand input by the user, and to modify the exposure demand according to the advertisement abruptness; the exposure demand is a range; the modification of the exposure demand by the advertisement abruptness is a reverse modification, used to modify the left endpoint and the right endpoint of the exposure demand at the same time;

[0125] An image selection unit, used to select an image whose exposure meets the exposure requirement in the video to be processed;

[0126] A query unit, used to query the position of the selected image as an advertisement insertion position;

[0127] The insertion execution unit is used to insert advertisement data into the video to be processed based on the advertisement insertion position.

[0128] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for updating advertisement data, characterized in that: The method comprises: Obtaining the video to be processed and its playback parameters at different playback ends, and determining the exposure of each frame of image according to the playback parameters at all playback ends; the playback parameter is the playback speed of each frame of image; Perform content recognition on the video to be processed and extract the video theme; When receiving the advertisement data, the advertisement data is subjected to content recognition, the advertisement theme of the advertisement data is extracted, the advertisement theme and the video theme are compared, and the advertisement abruptness is determined; wherein, in the process of performing content recognition on the processed video and advertisement data, an AI module is introduced; the advertisement theme and the video theme are both phrases; Receive the exposure requirement input by the user, select the advertisement insertion position according to the exposure requirement and the advertisement abruptness, and insert the advertisement data into the video to be processed; Transmitting the video to be processed containing the advertisement data to the playback end, receiving the feedback information of the playback end, and adjusting the advertisement insertion process according to the feedback information; The process of receiving the exposure requirement input by the user, selecting the advertisement insertion position according to the exposure requirement and the advertisement obtrusiveness, and inserting the advertisement data into the video to be processed includes: Receive the exposure requirement input by the user, select an image whose exposure meets the exposure requirement in the video to be processed, query the position of the selected image as the advertisement insertion position, and insert the advertisement data into the video to be processed based on the advertisement insertion position; There is a correction process for the exposure requirement, which is: receiving the exposure requirement input by the user, and correcting the exposure requirement according to the abruptness of the advertisement, and the correction method is to reversely correct the two endpoints; There is a correction process for the advertisement abruptness, which is: calculate the ratio of the feedback playback time to the initial playback time, calculate the disgust degree according to the inverse ratio of the ratio, and correct the advertisement abruptness according to the disgust degree, wherein the advertisement abruptness is proportional to the disgust degree.

2. The advertisement data updating method according to claim 1, characterized in that: The step of obtaining the video to be processed and its playback parameters at different playback ends, and determining the exposure of each frame image according to the playback parameters at all playback ends includes: Transmit the video to be processed to the playback end, and send a permission acquisition request to the playback end; After receiving the permission granted by the playback end, the playback end obtains the playback time of each image in the video to be processed, and calculates the playback speed according to the playback time; Read images in the video to be processed one by one, query the playback speed of the images in each playback end, and calculate the average playback speed; The exposure is determined according to the average playback speed; the exposure is inversely proportional to the average playback speed.

3. The advertisement data updating method according to claim 1, characterized in that: The process of comparing the advertisement theme with the video theme and determining the advertisement obtrusiveness includes: Read the advertisement theme and the video theme, input them into the preset word vector conversion model, and obtain the advertisement word vector group and the video word vector group respectively; Pair the elements in the advertisement word vector group and the video word vector group and calculate the vector distance; Select the longer vector group from the advertisement word vector group and the video word vector group, and read the minimum vector distance corresponding to each word vector in the longer vector group in turn; The mean of the minimum vector distances corresponding to each word vector is calculated, and the ad abruptness is determined based on the inverse ratio of the mean.

4. The advertisement data updating method according to claim 1, characterized in that: The steps of receiving the exposure requirement input by the user, selecting an advertisement insertion position according to the exposure requirement and the advertisement obtrusiveness, and inserting the advertisement data into the video to be processed include: Receive the exposure requirement input by the user, and modify the exposure requirement according to the advertisement abruptness; the exposure requirement is a range; the modification of the exposure requirement by the advertisement abruptness is a reverse modification, and the left endpoint and the right endpoint of the exposure requirement are modified at the same time; Select an image whose exposure meets the exposure requirement in the video to be processed; Query the position of the selected image as the advertisement insertion position; The advertisement data is inserted into the video to be processed based on the advertisement insertion position.

5. The advertisement data updating method according to claim 2, characterized in that: The steps of transmitting the to-be-processed video containing advertisement data to the playback end, receiving feedback information of the playback end, and adjusting the advertisement insertion process according to the feedback information include: The video to be processed containing the advertising data is transmitted to the playback end, and the playback time of each image is obtained based on the authority granted by the playback end as the feedback playback time; Query the playback time of each image when the playback end plays the video to be processed without advertising data, as the initial playback time; Compare the feedback playback time and the initial playback time of the same image to determine the degree of disgust; The obtrusiveness of the advertisement is adjusted according to the offensiveness.

6. The advertisement data updating method according to claim 5, characterized in that: The method further comprises: Determine the total interaction time between the player and the platform based on the player's playback record; Determining the impact level of the playback terminal according to the total interaction duration; Adjust the dislike level based on the impact level.

7. An advertisement data updating system, characterized in that: The system comprises: An exposure determination module is used to obtain the video to be processed and its playback parameters at different playback ends, and determine the exposure of each frame of image according to the playback parameters at all playback ends; the playback parameter is the playback speed of each frame of image; The video theme extraction module is used to identify the content of the video to be processed and extract the video theme; The abruptness calculation module is used to perform content recognition on the advertisement data when receiving the advertisement data, extract the advertisement theme of the advertisement data, compare the advertisement theme with the video theme, and determine the advertisement abruptness; wherein, in the process of performing content recognition on the video to be processed and the advertisement data, an AI module is introduced; the advertisement theme and the video theme are both phrases; An advertisement data insertion module is used to receive the exposure requirement input by the user, select an advertisement insertion position according to the exposure requirement and the advertisement abruptness, and insert the advertisement data into the video to be processed; A response information receiving module is used to transmit the to-be-processed video containing the advertisement data to the playback end, receive the feedback information of the playback end, and adjust the advertisement insertion process according to the feedback information; The process of receiving the exposure requirement input by the user, selecting the advertisement insertion position according to the exposure requirement and the advertisement obtrusiveness, and inserting the advertisement data into the video to be processed includes: Receive the exposure requirement input by the user, select an image whose exposure meets the exposure requirement in the video to be processed, query the position of the selected image as the advertisement insertion position, and insert the advertisement data into the video to be processed based on the advertisement insertion position; There is a correction process for the exposure requirement, which is: receiving the exposure requirement input by the user, and correcting the exposure requirement according to the abruptness of the advertisement, and the correction method is to reversely correct the two endpoints; There is a correction process for the advertisement abruptness, which is: calculate the ratio of the feedback playback time to the initial playback time, calculate the disgust degree according to the inverse ratio of the ratio, and correct the advertisement abruptness according to the disgust degree, wherein the advertisement abruptness is proportional to the disgust degree.

8. The advertisement data updating system according to claim 7, characterized in that: The exposure determination module comprises: The permission acquisition unit is used to transmit the video to be processed to the playback end and send a permission acquisition request to the playback end; A speed calculation unit, configured to obtain the playback time of each image in the video to be processed by the playback end after receiving the permission granted by the playback end, and calculate the playback speed according to the playback time; An average speed calculation unit, used to read images in the video to be processed in sequence, query the playback speed of the images in each playback end, and calculate the average playback speed; The speed application unit is used to determine the exposure according to the average playback speed; the exposure is inversely proportional to the average playback speed.

9. The advertisement data updating system according to claim 7, characterized in that: The abruptness calculation module comprises: A topic conversion unit, used to read the advertisement topic and the video topic, input them into a preset word vector conversion model, and obtain an advertisement word vector group and a video word vector group respectively; An element pairing unit, used to pair elements in the advertisement word vector group and the video word vector group and calculate the vector distance; A distance selection unit is used to select a longer vector group from the advertisement word vector group and the video word vector group, and sequentially read the minimum vector distance corresponding to each word vector in the longer vector group; The mean calculation application unit is used to calculate the mean of the minimum vector distance corresponding to each word vector, and determine the advertisement abruptness according to the inverse ratio of the mean.

10. The advertisement data updating system according to claim 7, characterized in that: The advertisement data insertion module comprises: A demand receiving unit, used to receive the exposure demand input by the user, and to modify the exposure demand according to the advertisement abruptness; the exposure demand is a range; the modification of the exposure demand by the advertisement abruptness is a reverse modification, used to modify the left endpoint and the right endpoint of the exposure demand at the same time; An image selection unit, used to select an image whose exposure meets the exposure requirement in the video to be processed; A query unit, used to query the position of the selected image as an advertisement insertion position; The insertion execution unit is used to insert advertisement data into the video to be processed based on the advertisement insertion position.

Images