CN110046263B

CN110046263B - Multimedia recommendation method, device, server and storage medium

Info

Publication number: CN110046263B
Application number: CN201910332578.XA
Authority: CN
Inventors: 董治; 李深远
Original assignee: Tencent Music Entertainment Technology Shenzhen Co Ltd
Current assignee: Tencent Music Entertainment Technology Shenzhen Co Ltd
Priority date: 2019-04-23
Filing date: 2019-04-23
Publication date: 2023-08-18
Anticipated expiration: 2039-04-23
Also published as: CN110046263A

Abstract

The embodiment of the invention discloses a multimedia recommendation method, a device, a server and a storage medium, wherein the method comprises the following steps: acquiring a multimedia edit and a candidate cover set of the multimedia edit, wherein the multimedia edit comprises at least two multimedia fragments, and the candidate cover set comprises covers corresponding to the multimedia fragments in the multimedia edit; acquiring a user portrait, and determining a target cover of the multimedia editing from the candidate cover set according to the user portrait; outputting the multimedia edit and the target cover so that the target cover is displayed on a user interface when the multimedia edit is recommended; the multimedia editing can be recommended better, and the click rate and the play quantity of the multimedia editing are improved.

Description

Multimedia recommendation method, device, server and storage medium

Technical Field

The present invention relates to the field of computer technologies, and in particular, to the field of multimedia processing technologies, and in particular, to a multimedia recommendation method, a multimedia recommendation device, a server, and a computer storage medium.

Background

The multimedia editing means: the multimedia is a combination of multiple media, and generally includes multiple media forms such as text, sound and image. When the user plays the multimedia using the terminal, the terminal can recommend the multimedia edit to the user at the user interface. If the terminal detects the clicking operation of the user on the multimedia editing in the user interface, each multimedia segment in the multimedia editing can be played in sequence.

In the process of implementing the present invention, the inventor finds that the existing multimedia recommendation method is generally: optionally selecting a preset cover as a cover of the multimedia edit, and displaying the cover in a user interface to recommend the multimedia edit; such a multimedia recommendation method cannot better attract users to click on the multimedia edit, resulting in lower click rate and play amount of the multimedia edit.

Disclosure of Invention

The embodiment of the invention provides a multimedia recommendation method, a device, a server and a storage medium, which can better recommend multimedia editing and improve the click rate and play quantity of the multimedia editing.

In one aspect, an embodiment of the present invention provides a multimedia recommendation method, including:

acquiring a multimedia edit and a candidate cover set of the multimedia edit, wherein the multimedia edit comprises at least two multimedia fragments, and the candidate cover set comprises covers corresponding to the multimedia fragments in the multimedia edit;

acquiring a user portrait, and determining a target cover of the multimedia editing from the candidate cover set according to the user portrait, wherein the user portrait is used for reflecting the multimedia preference of a user indicated by the user portrait;

and outputting the multimedia edit and the target cover so that the target cover is displayed on a user interface when the multimedia edit is recommended.

On the other hand, the multimedia recommendation method provided by the embodiment of the invention comprises the following steps:

acquiring a multimedia edit and a candidate cover set of the multimedia edit, wherein the multimedia edit comprises at least two multimedia fragments of different multimedia, and the candidate cover set comprises covers corresponding to each multimedia fragment in the multimedia edit;

acquiring a user portrait, and determining a target element according to the user portrait, wherein the target element is used for reflecting the multimedia preference of a user indicated by the user portrait;

Selecting a target multimedia fragment matched with the target element from the multimedia editing, and selecting a cover corresponding to the target multimedia fragment from the candidate cover set as a target cover of the multimedia editing;

In still another aspect, an embodiment of the present invention provides a multimedia recommendation apparatus, including:

the system comprises an acquisition unit, a storage unit and a display unit, wherein the acquisition unit is used for acquiring a multimedia edit and a candidate cover set of the multimedia edit, the multimedia edit comprises at least two multimedia fragments, and the candidate cover set comprises covers corresponding to all the multimedia fragments in the multimedia edit;

the acquisition unit is used for acquiring a user portrait, and determining a target cover of the multimedia editing from the candidate cover set according to the user portrait, wherein the user portrait is used for reflecting the multimedia preference of a user indicated by the user portrait;

and the output unit is used for outputting the multimedia editing and the target cover so that the target cover is displayed on a user interface when the multimedia editing is recommended.

the system comprises an acquisition unit, a storage unit and a display unit, wherein the acquisition unit is used for acquiring a multimedia edit and a candidate cover set of the multimedia edit, the multimedia edit comprises at least two multimedia fragments of different multimedia, and the candidate cover set comprises covers corresponding to all the multimedia fragments in the multimedia edit;

the acquisition unit is used for acquiring the user portrait and determining a target element according to the user portrait, wherein the target element is used for reflecting the multimedia preference of the user indicated by the user portrait;

a selecting unit, configured to select a target multimedia segment matched with the target element from the multimedia composition, and select a cover corresponding to the target multimedia segment from the candidate cover set as a target cover of the multimedia composition;

In yet another aspect, an embodiment of the present invention provides a server, including a processor, a transceiver, and a memory, where the processor, the transceiver, and the memory are connected to each other, and where the memory is configured to store a computer program, the computer program including first program instructions, and the processor is configured to invoke the first program instructions, and perform the steps of:

Alternatively, the computer program comprises second program instructions, the processor being configured to invoke the second program instructions to perform the steps of:

In yet another aspect, embodiments of the present invention provide a computer storage medium storing first computer program instructions adapted to be loaded by a processor and to perform the steps of:

Alternatively, the computer storage medium stores second computer program instructions adapted to be loaded by a processor and to perform the steps of:

When recommending the multimedia editing, the embodiment of the invention can acquire the user portrait, and then determine the target cover of the multimedia editing to be recommended from the candidate cover set according to the user portrait; because the user portrait is used for reflecting the multimedia preference of the user, the target cover determined according to the user portrait can meet the preference of the user to a certain extent, so that the user can be attracted to click to play the multimedia editing when the target cover is displayed on the user interface, and the click rate and play quantity of the multimedia editing can be improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1a is a schematic diagram of an interactive system according to an embodiment of the present invention;

FIG. 1b is a schematic diagram of a multimedia composition and a candidate cover set according to an embodiment of the present invention;

FIG. 1c is a schematic illustration of a user interface provided by an embodiment of the present invention;

fig. 2 is a flow chart of a multimedia recommendation method according to an embodiment of the present invention;

FIG. 3 is a flowchart illustrating a multimedia recommendation method according to another embodiment of the present invention;

FIG. 4 is a schematic diagram of a method for inserting transition animation according to an embodiment of the present invention;

FIG. 5a is a schematic diagram of a method for determining a target cover according to an embodiment of the present invention;

FIG. 5b is a schematic diagram of another method for determining a target cover according to an embodiment of the present invention;

FIG. 6 is a flowchart illustrating a multimedia recommendation method according to another embodiment of the present invention;

Fig. 7 is a schematic structural diagram of a multimedia recommendation device according to an embodiment of the present invention;

FIG. 8 is a schematic diagram of a multimedia recommendation device according to another embodiment of the present invention;

fig. 9 is a schematic structural diagram of a server according to an embodiment of the present invention.

Detailed Description

Covers refer to images that may represent the entire content of multimedia; the cover of the multimedia can be used for showing the whole content of the multimedia when the multimedia is not played so as to attract users to click to play the multimedia. Therefore, when the multimedia edit is recommended to the user, the cover of the multimedia edit can be acquired and displayed to the user at the user interface, so that the user is attracted to click to play the multimedia edit. Based on the above, the embodiment of the invention provides a multimedia recommendation scheme aiming at the multimedia editing, and the target cover which accords with the preference of the user can be determined by adopting the multimedia recommendation scheme, so that the click rate and the play quantity of the multimedia editing are improved. The multimedia recommendation scheme may be performed by a server or a terminal, where the server may be a service device for performing multimedia data processing, which may be a data processing server, a web server, etc.; the server can be an independent service device or a cluster device formed by a plurality of service devices together; the terminal here may be a smart phone, a laptop computer, a tablet computer or the like, a portable device, a desktop computer or the like.

Taking the server executing the multimedia recommendation scheme as an example: the user may send a multimedia request to the server during the process of playing the multimedia using the terminal, as shown in fig. 1 a. The server may acquire the multimedia composition and the candidate cover set of the multimedia composition after receiving the multimedia request from the terminal. The multimedia composition comprises at least two multimedia fragments of different multimedia, each multimedia fragment is orderly arranged in the multimedia composition, thereby forming the multimedia composition; the candidate cover set comprises covers corresponding to all the multimedia fragments in the multimedia editing, and the covers of the multimedia fragments are used for representing the whole content of the multimedia fragments. As shown in fig. 1b, the multimedia composition includes three multimedia clips, which are: multimedia segment a, multimedia segment b, and multimedia segment c; correspondingly, covers of the three multimedia fragments are included in the candidate cover set, and are respectively: cover 1, cover 2 and cover 3.

The server can also acquire a user portrait of the terminal, wherein the user portrait refers to a portrait of a user using the terminal, and the user portrait is used for reflecting the multimedia preference of the user indicated by the user portrait; for example: the user portrait may include historical multimedia playing information, and analysis of the historical multimedia playing information can know that the user has played the multimedia a, and that the playing frequency of the multimedia a is high, so that the user can be predicted to like the multimedia a, that is, the multimedia preference reflected by the user portrait in this case is "multimedia a". After the user portrait is acquired, a target cover meeting the user preference can be determined from the candidate cover set according to the user portrait, and the target cover and the multimedia editing are sent to the terminal. The terminal, after receiving the multimedia edit and the target cover, may store the multimedia edit in a memory and display the target cover in a user interface to recommend the multimedia edit. As shown in FIG. 1c, if it is determined from the candidate cover set that the target cover that meets the user's preference is cover 2 based on the user representation, then the cover 2 may be displayed on the user interface to attract the user to click to play the multimedia composition.

It should be noted that, in other embodiments, the terminal may also execute the multimedia recommendation scheme. Specifically, the terminal may send a multimedia request to the server to request the server to issue a multimedia edit. After the multimedia composition transmitted by the server is obtained, a candidate cover set for the multimedia composition may be determined. The terminal can also acquire the user portrait, and determine a target cover of the multimedia editing from the candidate cover set according to the user portrait; the multimedia compilation and the target cover are then output at the user interface.

Therefore, when recommending the multimedia editing, the embodiment of the invention can acquire the user portrait first, and then determine the target cover of the multimedia editing to be recommended from the candidate cover set according to the user portrait; because the user portrait can be used for reflecting the multimedia preference of the user, the target cover determined according to the user portrait can meet the preference of the user to a certain extent, so that the user can be attracted to click to play the multimedia edit when the target cover is displayed on the user interface, and the click rate and play quantity of the multimedia edit can be improved.

Based on the above description, an embodiment of the present invention proposes a schematic flowchart of a multimedia recommendation method in fig. 2, which may be implemented by the above-mentioned server or terminal. For convenience of description, the embodiment of the present invention takes the server to execute the multimedia recommendation method as an example. The multimedia recommendation method may include the following steps S201 to S203:

S201, acquiring a candidate cover set of the multimedia editing.

The multimedia composition herein may comprise at least two multimedia segments of different multimedia, and the candidate cover set comprises covers corresponding to each multimedia segment in the multimedia composition. Wherein, the multimedia fragments in the multimedia editing are: the segments obtained from the multimedia meeting the preset conditions are intercepted, and the multimedia meeting the preset conditions can include, but is not limited to: music, video, audio-video, text, etc.; wherein, audio and video can include: MV (Music Video); for convenience of description, the MV is taken as an example for the multimedia to be mentioned later in the embodiments of the present invention.

The preset conditions may include any one of the following: the play quantity is larger than a preset play quantity threshold, the effective play times is larger than a preset time threshold, and the click rate is larger than a preset click rate threshold. Wherein, the preset play quantity threshold value, the preset times threshold value and the preset click rate threshold value can be set according to experience values; the effective playing times are as follows: the number of times of playing the multimedia completely, wherein the time length of playing the multimedia completely is equal to the total time length of the multimedia. For example, if the total duration of the multimedia a is 300 seconds and the playing duration of the multimedia a is 200 seconds, the multimedia a is considered to be not completely played at this time, and then the playing cannot be recorded as effective playing; if the playing duration of the multimedia a is 300 seconds, the multimedia a is considered to be completely played, and the playing can be recorded as an effective playing. In other embodiments, the full play may also refer to: the difference between the playing time length and the total multimedia time length is smaller than a preset difference, and the preset difference can be set according to an empirical value or actual requirements, for example, the preset difference is set to be 10 seconds. In one embodiment, the multimedia clip may include: segments, such as climax segments, highlight segments, which are intercepted from the multimedia satisfying the preset condition and can be used to represent the subject idea of the multimedia satisfying the preset condition; or any one of the segments of the multimedia satisfying the preset condition.

The cover of the multimedia clip may include: static covers or dynamic covers. The static cover refers to a cover in a static image format, and can be composed of one frame of image; the moving cover is a cover in a moving image format, and may be composed of a plurality of frames of images. In one embodiment, the static cover is a cover made up of one frame of image taken from the multimedia segment, and the dynamic cover is a cover made up of multiple frames of image taken from the multimedia segment. It should be noted that, in other embodiments, the static cover may be a cover made up of a frame of image associated with the multimedia clip: for example, the static cover may be a cover formed by capturing a frame of image from multimedia corresponding to a multimedia clip; alternatively, the static cover may be a cover made of one frame of image made according to the theme concept of the multimedia clip. Similarly, in other embodiments, the dynamic cover may be a cover composed of multiple frames of images associated with a multimedia clip.

The server may obtain the multimedia edit and the candidate cover collection of the multimedia edit after detecting the triggering event recommended by the multimedia edit. Wherein the triggering event may include, but is not limited to: receiving an event of a multimedia request of a terminal, detecting an event of updating a multimedia database in a server, and an event of reaching a recommendation period of a multimedia edit; wherein, the recommended period of the multimedia arrives as follows: the time interval between the current time and the last time the multimedia edit was recommended is equal to an event of a preset time interval, which is set according to the recommendation period, for example, the recommendation period may be set to 7 days, and then the preset time interval is 7 days.

S202, acquiring a user portrait, and determining a target cover of the multimedia editing from the candidate cover set according to the user portrait, wherein the user portrait is used for reflecting the multimedia preference of a user indicated by the user portrait.

The server can acquire the user identification of the target terminal, and acquire the associated user portrait according to the user identification; the target terminal herein may include: any terminal in the terminal set connected with the server, or a terminal for sending a multimedia request to the server; the user representation may include at least one of the following information: historical multimedia play information, historical multimedia search information, and historical multimedia collection information.

Because the user portraits can reflect the multimedia preferences of the user, the target covers which accord with the user preferences can be determined from the candidate cover sets according to the user portraits. Specifically, the server may determine, according to the user representation, a target multimedia segment that meets the user preference, and then determine, according to the target multimedia segment, a target cover that meets the user preference from the candidate cover set.

S203, outputting the multimedia editing and the target cover so that the target cover is displayed on the user interface when the multimedia editing is recommended.

After determining the target cover of the multimedia edit, the server may output the multimedia edit and the target cover to the target terminal. After receiving the multimedia edit and the target cover, the target terminal may display the target cover on the user interface to attract the user to click to play the multimedia edit. And if the terminal detects the clicking operation on the target cover of the multimedia editing, playing the multimedia editing. As described above, the target terminal may be any terminal in the terminal set connected to the server, or may be a terminal that transmits a multimedia request to the server. Therefore, the server can actively transmit the multimedia edit and the target cover to any terminal, so that any terminal can display the target cover to recommend the multimedia edit; the multimedia edit and the target cover may also be transmitted to a terminal that transmits the multimedia request to the server after receiving the multimedia request, so that the terminal may display the target cover to recommend the multimedia edit.

In one embodiment, when the server outputs the multimedia edit and the target cover to the target terminal, an edit identifier can be created for the multimedia edit, and the multimedia edit and the edit identifier are associated and stored in the multimedia database; and then outputting the editing identifier of the multimedia editing and the target cover to the target terminal. Accordingly, the target terminal may display the target cover on the user interface after receiving the editing identifier and the target cover. If the clicking operation on the target cover is detected, acquiring the multimedia edit from the server according to the edit identifier, and playing the multimedia edit. Because the memory occupation amount of the editing mark is less, when recommending the multimedia editing, the multimedia editing mark is stored in the multimedia database of the server, and only the editing mark is sent to the target terminal, so that the memory of the target terminal can be saved, the running speed of the target terminal can be improved, the playing efficiency of playing the multimedia editing by the target terminal can be improved, and the user experience can be improved.

Based on the above description, an embodiment of the present invention proposes a schematic flowchart of another multimedia recommendation method in fig. 3, which may be implemented by the above-mentioned server or terminal. For convenience of description, the embodiment of the present invention takes the server to execute the multimedia recommendation method as an example. The multimedia recommendation method may include the following steps S301 to S306:

s301, acquiring a candidate cover set of the multimedia editing.

In a specific implementation process, first, a multimedia collection may be acquired, where the multimedia collection includes a plurality of multimedia to be recommended. The server may count the play amount of each multimedia in a preset time period, where the preset time period may be set according to the service requirement, for example, the preset time period is set as follows: calculating a time period formed by one week forwards by taking the current time as a reference, wherein if the current time is 12 months and 10 days, the preset time period is 12 months and 3 days to 12 months and 9 days; determining candidate multimedia according to the play amount of each multimedia, wherein the play amount of the candidate multimedia is larger than a preset play amount threshold; taking the candidate multimedia as the multimedia to be recommended; and constructing a multimedia set by adopting the multimedia to be recommended. In one embodiment, if the number of candidate multimedia is greater, a preset number of candidate multimedia may be selected from the candidate multimedia according to the order from the high play amount to the low play amount as the multimedia to be recommended. It should be understood that the candidate multimedia may also be determined according to the number of effective plays or click rate of the multimedia to be recommended, and the principle is similar to that of determining the candidate multimedia according to the play amount, which is not described herein. It should be noted that, in other embodiments, the multimedia to be recommended may also be determined according to a preset multimedia style, for example, the preset multimedia style is "rock", and then the multimedia with the multimedia style of "rock" may be obtained as the multimedia to be recommended; in other embodiments, multimedia belonging to the same author (or singer) may also be acquired as multimedia to be recommended; in other embodiments, the multimedia related to the holiday may also be used as the multimedia to be recommended according to the actual requirement; for example, during mid-autumn festival, multimedia on topics of "mid-autumn", "pleasant", "reunion", etc. may be acquired as multimedia to be recommended.

And then, the corresponding multimedia fragments can be intercepted from each multimedia to be recommended, and at least one frame of image is intercepted from each multimedia fragment to form the cover of each multimedia fragment. Specifically, the server may first mark time information of the multimedia segments in each multimedia to be recommended by using a preset marking method, where the time information may include: a start time and an end time of the multimedia clip; the preset marking method may include: manual labeling methods and/or deep neural network-based labeling methods. And then intercepting the multimedia fragments from the multimedia to be recommended based on the time information of the multimedia fragments. From the foregoing, the cover of the multimedia clip may include: static covers or dynamic covers; then when at least one frame of image is cut out from each multimedia segment to form the cover of each multimedia segment, one frame of image can be cut out from any multimedia segment aiming at any multimedia segment, and the cut out frame of image is used as the static cover of any multimedia segment; or, intercepting multi-frame images from any multimedia segment, and adopting the intercepted multi-frame images to form the dynamic cover of any multimedia segment.

Then, a multimedia composition may be constructed using each multimedia clip, and a candidate cover set may be constructed using covers of each multimedia clip. When each multimedia segment is adopted to construct a multimedia composition, the server can splice each multimedia segment by using a multimedia splicing tool to obtain spliced multimedia, and the multimedia splicing tool can comprise: FFmpeg libraries (Fast Forward Mpeg, a suite of open source computer programs that record, convert digital audio, video, and convert it to streams); and taking the spliced multimedia as a multimedia editing. In one embodiment, the universal head multimedia and tail multimedia may be obtained, and the multimedia composition obtained by the embodiment includes: the head multimedia, the spliced multimedia and the tail multimedia.

In one embodiment, the multimedia clips in the multimedia composition are arranged sequentially; after the multimedia composition is obtained, a multimedia fragment pair in the multimedia composition can be obtained, wherein the multimedia fragment pair comprises a first multimedia fragment and a second multimedia fragment, and the first multimedia fragment and the second multimedia fragment are adjacent; and obtaining a transition animation, and inserting the transition animation between the first multimedia segment and the second multimedia segment. The transition animation can be any animation in a preset animation library, and can also be an animation related to the first multimedia segment and the second multimedia segment. As shown in fig. 4: the multimedia segment pair includes a multimedia segment b (first multimedia segment) and a multimedia segment c (second multimedia segment); and obtaining a transition animation, and then inserting the transition animation between the multimedia segment b and the multimedia segment c. According to the embodiment of the invention, the transition animation is inserted into two adjacent multimedia fragments (the first multimedia fragment and the second multimedia fragment), so that the transition from the first multimedia fragment to the second multimedia fragment is more natural and smoother.

S302, obtaining a user portrait.

S303, determining a target multimedia fragment according to the user portrait, wherein the first playing probability of the target multimedia fragment is larger than a probability threshold.

Specifically, the prediction of multimedia playing may be performed according to the historical multimedia information of the user portrait, and the target multimedia segment may be determined, where the historical multimedia information may include any one of the following: historical multimedia play information, historical multimedia search information, and historical multimedia collection information. Then, a target multimedia segment is obtained from the target multimedia, and the target multimedia segment comprises: segments of the target multimedia representing the subject matter of the target multimedia, such as climax segments, highlight segments, or logo segments of the target multimedia; or any fragment of the target multimedia. It should be noted that, since the target multimedia segment is obtained by capturing from the target multimedia, the first playing probability of the target multimedia segment is the same as the first playing probability of the target multimedia mentioned later.

In one embodiment, the historical multimedia information includes historical multimedia playing information, and correspondingly, the specific embodiment of the target multimedia determination may be that multimedia playing prediction is performed according to the historical multimedia information of the user portrait: firstly, historical multimedia played in history is determined according to the historical multimedia playing information, and single historical playing duration and historical playing times of each historical multimedia are determined. Then, calculating the favorite value of each historical multimedia according to the single historical play time length and the historical play times of each historical multimedia; in one embodiment, the total historical play time length of each historical multimedia can be calculated according to the single historical play time length and the historical play times of each historical multimedia, and the total historical play time length of each historical multimedia is used as the favorite value of each historical multimedia. And finally, determining a first playing probability of each historical multimedia according to the favorite value of each historical multimedia, and taking the historical multimedia with the first playing probability larger than a preset threshold value as the target multimedia. Specifically, a first correspondence table of favorite values and play probabilities can be obtained, play probabilities corresponding to the favorite values of each historical multimedia are respectively searched from the first correspondence table, and the searched play probabilities are respectively used as the first play probabilities of each historical multimedia. The first correspondence table of the preference value and the play probability may be as shown in table 1, where the preference value and the play probability are proportional.

TABLE 1

Preference value	Probability of play
		Greater than or equal to 2000	90％
[1500，2000)	80％
		[1000，1500)	70％
……	……

In still another embodiment, the historical multimedia information includes historical multimedia search information, and correspondingly, the specific embodiment of multimedia playing prediction according to the historical multimedia information of the user portrait, and determining the target multimedia may be: search multimedia of the historical search is determined according to the historical multimedia search information, and the historical search information of each search multimedia comprises the historical search times and/or the historical search time. Then, preference values of the respective search multimedia are determined based on the historical search information of the respective search multimedia. In one embodiment, the historical search times of each search multimedia can be used as the favorite value of each search multimedia; in still another embodiment, a search time difference between a historical search time and a current time of each search multimedia may be calculated separately; and obtaining a second corresponding relation table of the search time difference value and the favorite values, respectively searching the favorite values corresponding to the search time difference value of each search multimedia from the second corresponding relation table, and respectively taking the searched favorite values as the favorite values of each search multimedia. And finally, determining a first playing probability of each search multimedia according to the favorite value of each search multimedia, and taking the search multimedia with the first playing probability larger than a preset threshold value as the target multimedia. The second correspondence table of the search time difference value and the preference value may be as shown in table 2, where the search time difference value and the preference value are inversely proportional.

TABLE 2

Search time difference	Preference value
		Less than 24 hours	100
[24 hours, 48 hours ]	90
		[48 hours, 72 hours)	80
……	……

In still another embodiment, the historical multimedia information includes historical multimedia collection information, and correspondingly, the specific embodiment of the determining the target multimedia may be that: and determining the collected multimedia of the historical collection and the historical collection time of each collected multimedia according to the historical multimedia collection information. Then, a preference value of each of the plurality of collected multimedia is determined based on the historical collection time of each of the plurality of collected multimedia. Specifically, the collection time difference between the historical collection time and the current time of each collection multimedia can be calculated respectively; and acquiring a third corresponding relation table of the collection time difference values and the favorite values, respectively searching favorite values corresponding to the collection time difference values of the collected multimedia from the third corresponding relation table, and taking the searched favorite values as the favorite values of the collected multimedia respectively. And finally, determining a first playing probability of each collected multimedia according to the favorite value of each collected multimedia, and taking the collected multimedia with the first playing probability larger than a preset threshold value as the target multimedia.

S304, detecting whether the target multimedia fragment exists in the multimedia editing.

S305, if the target multimedia segment exists in the multimedia composition, selecting a cover corresponding to the target multimedia segment from the candidate cover set as a target cover of the multimedia composition.

In steps S304-S304, since the first play probability of the target multimedia clip is greater than the preset threshold value, which indicates that the user is interested in comparing the target multimedia clip, it is possible to detect whether the target multimedia clip is present in the multimedia composition after determining the target multimedia clip. If the target multimedia clip exists, searching the cover of the target multimedia clip from the candidate cover set, and taking the cover of the searched target multimedia clip as the target cover of the multimedia editing so as to attract the user to click the multimedia editing. As shown in fig. 5a, the multimedia composition includes 3 multimedia clips, which are respectively: multimedia segment a, multimedia segment b, and multimedia segment c; correspondingly, the cover of the multimedia segment a is cover 1, the cover of the multimedia segment b is cover 2, and the cover of the multimedia segment c is cover 3. The target multimedia fragment is multimedia fragment b, and if the target multimedia fragment b exists in the multimedia editing, the cover 2 is taken as the target cover.

It should be noted that, if there are multiple target multimedia fragments in the multimedia editing, one target multimedia fragment can be arbitrarily selected from the multiple target multimedia fragments, and the cover corresponding to the selected target multimedia fragment is used as the target cover of the multimedia editing; or selecting a target multimedia fragment with the largest first playing probability from the target multimedia fragments, and taking a cover corresponding to the selected target multimedia fragment with the largest first playing probability as a target cover of the multimedia editing; or combining and splicing the covers of the plurality of target multimedia fragments to obtain the target cover.

In one embodiment, if the target multimedia clip does not exist in the multimedia editing, each cover in the candidate cover set is spliced in a combined manner, and the combined and spliced cover is used as the target cover of the multimedia editing. Specifically, the specific implementation manner of performing the combination and the splicing on each cover in the candidate cover set may be: splicing the covers in the candidate cover set to obtain a spliced base map; acquiring background color shade and multimedia editing description information, wherein the transparency of the background color shade is larger than a preset transparency threshold; and covering the ground color shade and the description information on the spliced base map to obtain the combined and spliced cover. As shown in fig. 5b, the candidate cover set includes 5 covers, which are respectively: cover 1, cover 2, cover 3, cover 4, and cover 5, and the splice base map p can be obtained by splicing these 5 covers. And acquiring the background color shade m and the description information of the multimedia editing, and covering the background color shade m and the description information on the spliced base map p to obtain a combined and spliced cover s, wherein the cover s is the target cover.

In still another embodiment, if the target multimedia clip does not exist in the multimedia composition, the similarity of each multimedia clip in the target multimedia clip and the multimedia composition may be calculated first. Specifically, firstly, a feature matrix algorithm (such as a fourier transform algorithm, a mel-frequency cepstrum coefficient algorithm, etc.) can be adopted to calculate a feature vector matrix of a target multimedia segment and a feature vector matrix of each multimedia segment in the multimedia composition; and then, adopting a similarity algorithm (such as a cosine similarity algorithm, a Euclidean distance algorithm and the like) to respectively calculate the similarity between the feature vector matrix of the target multimedia fragment and each multimedia fragment in the multimedia editing so as to obtain the similarity of the target multimedia fragment and each multimedia fragment in the multimedia editing. After obtaining the similarity of the target multimedia segment and each multimedia segment in the multimedia composition, the target cover of the multimedia composition may be determined from the candidate cover set according to the similarity. Wherein, the specific implementation mode of determining the target cover of the multimedia editing from the candidate cover set according to the similarity can be as follows: determining the similar multimedia fragments from the multimedia composition according to the similarity, wherein the similar multimedia fragments are multimedia fragments with the maximum similarity in the multimedia composition; searching the covers of the similar multimedia fragments from the candidate cover set, and taking the searched covers of the similar multimedia fragments as target covers of the multimedia editing.

In still another embodiment, if the target multimedia clip does not exist in the multimedia clip, a second playing probability of each multimedia clip in the multimedia clip is predicted according to the user image, and the target cover of the multimedia clip is determined from the candidate cover set according to the second playing probability of each multimedia clip. The specific implementation manner of determining the target cover of the multimedia composition from the candidate cover set according to the second playing probability of each multimedia fragment may be: determining the multimedia segment with the maximum second play probability according to the second play probability of each multimedia segment; searching the cover of the multimedia fragment with the highest second playing probability from the candidate cover set, and taking the searched cover of the multimedia fragment with the highest second playing probability as a target cover of the multimedia editing.

S306, outputting the multimedia editing and the target cover so that the target cover is displayed on the user interface when the multimedia editing is recommended.

After determining the target cover of the multimedia edit, the server may output the multimedia edit and the target cover to the terminal. After receiving the multimedia edit and the target cover, the terminal may display the target cover on a user interface to attract the user to click to play the multimedia edit. And if the terminal detects the clicking operation on the target cover of the multimedia editing, playing the multimedia editing. In one embodiment, the server may further issue presentation information of each multimedia clip, so that when the terminal plays each multimedia clip in the multimedia clip, the corresponding presentation information may be displayed in an information display board for displaying the multimedia clip, where the presentation information may include, but is not limited to: the name of the multimedia to which the multimedia clip belongs, the corresponding author (e.g., singer), the region information to which the author belongs, the ranking of the multimedia to which the multimedia clip belongs, and so on.

Based on the above description, an embodiment of the present invention proposes a schematic flowchart of another multimedia recommendation method in fig. 6, which may be implemented by the above-mentioned server or terminal. For convenience of description, the embodiment of the present invention takes the server to execute the multimedia recommendation method as an example. The multimedia recommendation method may include the following steps S601 to S604:

s601, acquiring a candidate cover set of the multimedia composition.

The multimedia composition may comprise at least two multimedia segments of different multimedia, and the candidate cover set may comprise covers corresponding to each multimedia segment in the multimedia composition. Wherein, the multimedia fragments in the multimedia editing are: intercepting a segment obtained from multimedia meeting preset conditions, wherein the preset conditions comprise any one of the following: the play quantity is larger than a preset play quantity threshold, the effective play times is larger than a preset time threshold, and the click rate is larger than a preset click rate threshold. The cover corresponding to the multimedia fragment comprises: static covers or dynamic covers; the static cover is a cover formed by a frame of image intercepted from the multimedia segment, and the dynamic cover is a cover formed by a plurality of frames of image intercepted from the multimedia segment.

Optionally, each multimedia segment in the multimedia editing is arranged in sequence; after the multimedia composition is obtained, a multimedia fragment pair in the multimedia composition can be obtained, wherein the multimedia fragment pair comprises a first multimedia fragment and a second multimedia fragment, and the first multimedia fragment and the second multimedia fragment are adjacent; and obtaining a transition animation, and inserting the transition animation between the first multimedia segment and the second multimedia segment.

It should be noted that, the specific implementation manner of the step S601 may refer to the step S201 or the step S301 in the foregoing embodiment of the invention, which is not described herein again.

S602, obtaining a user portrait, and determining a target element according to the user portrait, wherein the target element is used for reflecting the multimedia preference of a user indicated by the user portrait.

The user representation may include a historical multimedia operations record, which may include, but is not limited to: historical multimedia play records, historical multimedia search records, historical multimedia collection records, and the like. After the user portrait is acquired, abstracting a target element according to a historical multimedia operation record in the user portrait; target elements herein may include, but are not limited to: target multimedia styles, target multimedia authors (e.g., target video authors, target song singers, etc.).

In one embodiment, the historical multimedia operations record includes a historical multimedia play record; accordingly, the specific implementation manner of abstracting the target element according to the historical multimedia operation record in the user portrait may be: determining historical multimedia of historical play according to the historical multimedia play records, and the total historical play duration of each historical multimedia and the elements of each historical multimedia; and taking the element of the history multimedia with the largest total history playing duration as a target element.

In yet another embodiment, the historical multimedia operations record includes a historical multimedia search record; accordingly, the specific implementation manner of abstracting the target element according to the historical multimedia operation record in the user portrait may be: determining search multimedia of historical search according to the historical multimedia search record, the historical search times of each search multimedia and elements of each search multimedia; and taking the element of the searching multimedia with the largest historical searching times as a target element.

In yet another embodiment, the historical multimedia operations record includes a historical multimedia collection record; accordingly, the specific implementation manner of abstracting the target element according to the historical multimedia operation record in the user portrait may be: determining the collected multimedia of the history collection according to the history multimedia collection record, the history collection time of each collected multimedia and the elements of each collected multimedia; and calculating the collection time difference value between the historical collection time and the current time of each collection multimedia, and taking the element of the collection multimedia with the smallest collection time difference value as a target element.

S603, selecting a target multimedia segment matched with the target element from the multimedia editing, and selecting a cover corresponding to the target multimedia segment from the candidate cover set as a target cover of the multimedia editing.

When selecting a target multimedia fragment matched with a target element from the multimedia composition, acquiring the element corresponding to each multimedia fragment in the multimedia composition; and if the element corresponding to each multimedia fragment comprises a target element, taking the multimedia fragment corresponding to the target element as a target multimedia fragment. If the elements corresponding to the multimedia fragments do not comprise the target elements, calculating the similarity between the elements corresponding to the multimedia fragments and the target elements; specifically, similarity between the element corresponding to each multimedia segment and the target element can be calculated by adopting a cosine similarity algorithm, a Euclidean distance algorithm, a Hamming distance algorithm and other similarity algorithms. Then, determining a target multimedia fragment according to the similarity, wherein the target multimedia fragment is a multimedia fragment corresponding to an element with the similarity larger than a similarity threshold value; the similarity threshold here may be set according to actual business requirements or empirical values. If the similarity of the plurality of multimedia fragments is larger than the similarity threshold, selecting the multimedia fragment with the largest similarity from the plurality of multimedia fragments larger than the similarity threshold as the target multimedia fragment; or selecting any multimedia segment from the multimedia segments larger than the similarity threshold as a target multimedia segment

Optionally, if the selection of the target multimedia segment fails, each cover in the candidate cover set may be spliced in a combined manner, and the combined and spliced cover is used as the target cover of the multimedia editing; the specific embodiment may be shown in fig. 5b, and will not be described herein.

S604, outputting the multimedia editing and the target cover so that the target cover is displayed on a user interface when the multimedia editing is recommended.

When the embodiment of the invention recommends multimedia editing, the user portrait can be obtained and the target element can be determined according to the user portrait; then, determining a target multimedia fragment according to the target element, and selecting a cover corresponding to the target multimedia fragment from the candidate cover set as a target cover of the multimedia editing; because the target element is used for reflecting the multimedia preference of the user indicated by the user portrait, the target multimedia fragment determined according to the target element can meet the preference of the user to a certain extent, so that the user can be attracted to click to play the multimedia editing when the target cover corresponding to the target multimedia fragment is displayed on the user interface, and the click rate and play quantity of the multimedia editing can be improved.

Based on the description of the method embodiment, in an embodiment, the embodiment of the present invention further provides a schematic structural diagram of the multimedia recommendation device shown in fig. 7. As shown in fig. 7, the multimedia recommending apparatus in the embodiment of the present invention may include:

an obtaining unit 101, configured to obtain a multimedia composition and a candidate cover set of the multimedia composition, where the multimedia composition includes at least two multimedia segments, and the candidate cover set includes covers corresponding to each multimedia segment in the multimedia composition;

the acquiring unit 101 is configured to acquire a user portrait, and determine a target cover of the multimedia editing from the candidate cover set according to the user portrait;

an output unit 102 for outputting the multimedia edit and the target cover so that the target cover is displayed on a user interface when the multimedia edit is recommended.

In one embodiment, the obtaining unit 101, when configured to determine the target cover of the multimedia edit from the candidate cover set according to the user portrait, may be specifically configured to:

determining a target multimedia fragment according to the user portrait, wherein the first playing probability of the target multimedia fragment is larger than a probability threshold;

And if the target multimedia fragment exists in the multimedia editing, selecting a cover corresponding to the target multimedia fragment from the candidate cover set as a target cover of the multimedia editing.

In yet another embodiment, the obtaining unit 101, when configured to determine the target multimedia clip according to the user representation, may be specifically configured to:

performing multimedia play prediction according to the historical multimedia information of the user portrait, and determining target multimedia, wherein the historical multimedia information comprises any one of the following: historical multimedia play information, historical multimedia search information and historical multimedia collection information;

and intercepting the target multimedia to obtain a target multimedia fragment.

In yet another embodiment, the obtaining unit 101 may further be configured to:

if the target multimedia fragment does not exist in the multimedia editing, combining and splicing all covers in the candidate cover set, and taking the combined and spliced cover as the target cover of the multimedia editing; or,

if the target multimedia fragment does not exist in the multimedia editing, calculating the similarity of the target multimedia fragment and each multimedia fragment in the multimedia editing, and determining a target cover of the multimedia editing from the candidate cover set according to the similarity; or,

If the target multimedia segment does not exist in the multimedia composition, predicting second playing probability of each multimedia segment in the multimedia composition according to the user portrait, and determining a target cover of the multimedia composition from the candidate cover set according to the second playing probability of each multimedia segment.

In still another embodiment, the obtaining unit 101, when used for determining the target cover of the multimedia edit from the candidate cover set according to the similarity, may be specifically configured to:

determining the similar multimedia fragments from the multimedia composition according to the similarity, wherein the similar multimedia fragments are multimedia fragments with the maximum similarity in the multimedia composition;

searching the covers of the similar multimedia fragments from the candidate cover set, and taking the searched covers of the similar multimedia fragments as target covers of the multimedia editing.

In yet another embodiment, the multimedia clips in the multimedia composition are: intercepting a segment obtained from multimedia meeting preset conditions, wherein the preset conditions comprise any one of the following: the play quantity is larger than a preset play quantity threshold, the effective play times is larger than a preset times threshold, and the click rate is larger than a preset click rate threshold;

The cover of the multimedia clip includes: static covers or dynamic covers; the static cover is a cover formed by a frame of image intercepted from the multimedia segment, and the dynamic cover is a cover formed by a plurality of frames of image intercepted from the multimedia segment.

In yet another embodiment, the multimedia clips in the multimedia composition are arranged sequentially; accordingly, the acquisition unit 101 may further be configured to:

acquiring a multimedia fragment pair in the multimedia editing, wherein the multimedia fragment pair comprises a first multimedia fragment and a second multimedia fragment, and the first multimedia fragment is adjacent to the second multimedia fragment;

and obtaining a transition animation, and inserting the transition animation between the first multimedia segment and the second multimedia segment.

According to one embodiment of the present invention, the steps involved in the methods shown in fig. 2 and 3 may be performed by the units in the multimedia recommendation device shown in fig. 7. For example, steps S201 and S202 shown in fig. 2 may be performed by the acquisition unit 101 shown in fig. 7, and step S203 may be performed by the output unit 102 shown in fig. 7; as another example, steps S301 to S305 shown in fig. 3 may be performed by the acquisition unit 101 shown in fig. 7, and step S306 may be performed by the output unit 102 shown in fig. 7.

According to another embodiment of the present invention, each unit in the multimedia recommendation apparatus shown in fig. 7 may be separately or completely combined into one or several other units, or some unit(s) thereof may be further split into a plurality of units with smaller functions, which may achieve the same operation without affecting the implementation of the technical effects of the embodiments of the present invention. The above units are divided based on logic functions, and in practical applications, the functions of one unit may be implemented by a plurality of units, or the functions of a plurality of units may be implemented by one unit. In other embodiments of the present invention, the multimedia recommendation device may also include other units, and in practical applications, these functions may also be implemented with assistance from other units, and may be implemented by cooperation of multiple units.

According to another embodiment of the present invention, a multimedia recommender device as shown in fig. 7 may be constructed by running a computer program (including program code) capable of executing the steps involved in the respective methods as shown in fig. 2 and 3 on a general purpose computing device such as a computer including a Central Processing Unit (CPU), a random access storage medium (RAM), a read only storage medium (ROM), etc., processing elements and storage elements, and implementing the multimedia recommendation method of the embodiment of the present invention. The computer program may be recorded on, for example, a computer-readable recording medium, and loaded into and executed by the above-described computing device via the computer-readable recording medium.

Based on the description of the method embodiment, in an embodiment, the embodiment of the present invention further provides a schematic structural diagram of the multimedia recommendation device shown in fig. 8. As shown in fig. 8, the multimedia recommending apparatus in the embodiment of the present invention may include:

an obtaining unit 201, configured to obtain a multimedia composition and a candidate cover set of the multimedia composition, where the multimedia composition includes at least two multimedia segments of different multimedia, and the candidate cover set includes covers corresponding to each multimedia segment in the multimedia composition;

an acquisition unit 201, configured to acquire a user portrait, and determine a target element according to the user portrait, where the target element is used to reflect a multimedia preference of a user indicated by the user portrait;

A selecting unit 202, configured to select a target multimedia segment matched with the target element from the multimedia composition, and select a cover corresponding to the target multimedia segment from the candidate cover set as a target cover of the multimedia composition;

an output unit 203 for outputting the multimedia edit and the target cover so that the target cover is displayed on a user interface when the multimedia edit is recommended.

In one embodiment, the selecting unit 202, when used for selecting a target multimedia clip matching the target element from the multimedia composition, may be specifically used for:

acquiring elements corresponding to each multimedia fragment in the multimedia editing;

if the element corresponding to each multimedia segment comprises the target element, taking the multimedia segment corresponding to the target element as a target multimedia segment;

if the elements corresponding to the multimedia fragments do not comprise the target elements, calculating the similarity between the elements corresponding to the multimedia fragments and the target elements, and determining a target multimedia fragment according to the similarity, wherein the target multimedia fragment is a multimedia fragment corresponding to an element with similarity greater than a similarity threshold.

In yet another embodiment, the selecting unit 202 may be further configured to:

if the selection of the target multimedia segment fails, each cover in the candidate cover set is spliced in a combined mode, and the combined and spliced cover is used as the target cover of the multimedia editing.

In yet another embodiment, the multimedia clips in the multimedia composition are arranged sequentially; accordingly, the obtaining unit 201 may further be configured to:

According to one embodiment of the present invention, the steps involved in the method shown in fig. 6 may be performed by the units in the multimedia recommendation device shown in fig. 8. For example, steps S601 and S602 shown in fig. 6 may be performed by the acquisition unit 201 shown in fig. 8, step S603 may be performed by the selection unit 202 shown in fig. 8, and step S604 may be performed by the output unit 203 shown in fig. 8.

According to another embodiment of the present invention, each unit in the multimedia recommendation apparatus shown in fig. 8 may be separately or completely combined into one or several other units, or some unit(s) thereof may be further split into a plurality of units with smaller functions, which may achieve the same operation without affecting the implementation of the technical effects of the embodiments of the present invention. The above units are divided based on logic functions, and in practical applications, the functions of one unit may be implemented by a plurality of units, or the functions of a plurality of units may be implemented by one unit. In other embodiments of the present invention, the multimedia recommendation device may also include other units, and in practical applications, these functions may also be implemented with assistance from other units, and may be implemented by cooperation of multiple units.

According to another embodiment of the present invention, a multimedia recommendation apparatus device as shown in fig. 8 may be constructed by running a computer program (including program code) capable of executing the steps involved in the respective methods as shown in fig. 6 on a general-purpose computing device such as a computer including a processing element such as a Central Processing Unit (CPU), a random access storage medium (RAM), a read only storage medium (ROM), and the like, and a storage element, and implementing the multimedia recommendation method of the embodiment of the present invention. The computer program may be recorded on, for example, a computer-readable recording medium, and loaded into and executed by the above-described computing device via the computer-readable recording medium.

Based on the description of the method embodiment and the device embodiment, the embodiment of the invention also provides a server. Referring to fig. 9, the server includes at least a processor 301, a transceiver 302 and a memory 303. Wherein the processor 301, transceiver 302, and memory 303 within the server may be connected by a bus or other means. The memory 303 is used for storing a computer program comprising program instructions, said processor 301 being configured for invoking said program instructions for performing the above mentioned multimedia recommendation method. The processor 301, or CPU (Central Processing Unit ), is a computing core as well as a control core of the server, adapted to implement one or more instructions, in particular to load and execute one or more instructions to implement a corresponding method flow or a corresponding function.

In one embodiment, the processor 301 according to the embodiment of the present invention may be configured to perform a series of multimedia recommendation processes, including: acquiring a multimedia edit and a candidate cover set of the multimedia edit, wherein the multimedia edit comprises at least two multimedia fragments of different multimedia, and the candidate cover set comprises covers corresponding to each multimedia fragment in the multimedia edit; acquiring a user portrait, and determining a target cover of the multimedia editing from the candidate cover set according to the user portrait, wherein the user portrait is used for reflecting the multimedia preference of a user indicated by the user portrait; outputting the multimedia edit and the target cover such that the target cover is displayed on a user interface when the multimedia edit is recommended, and so on. Alternatively, the processor 301 according to the embodiment of the present invention may be configured to perform a series of multimedia recommendation processes, including: acquiring a multimedia edit and a candidate cover set of the multimedia edit, wherein the multimedia edit comprises at least two multimedia fragments of different multimedia, and the candidate cover set comprises covers corresponding to each multimedia fragment in the multimedia edit; acquiring a user portrait, and determining a target element according to the user portrait, wherein the target element is used for reflecting the multimedia preference of a user indicated by the user portrait; selecting a target multimedia fragment matched with the target element from the multimedia editing, and selecting a cover corresponding to the target multimedia fragment from the candidate cover set as a target cover of the multimedia editing; outputting the multimedia edit and the target cover such that the target cover is displayed on a user interface when the multimedia edit is recommended, and so on.

The embodiment of the invention also provides a computer storage medium (Memory), which is a Memory device in a server and is used for storing programs and data. It is to be understood that the computer storage media herein may include built-in storage media in the server, or may include extended storage media supported by the server. The computer storage media provides storage space that stores the operating system of the server. Also stored in this memory space are one or more instructions, which may be one or more computer programs (including program code), adapted to be loaded and executed by the processor 301. The computer storage medium herein may be a high-speed RAM memory or a non-volatile memory (non-volatile memory), such as at least one magnetic disk memory; optionally, at least one computer storage medium remote from the processor may be present.

In one embodiment, first computer program instructions stored in a computer storage medium may be loaded and executed by the processor 301 to implement the corresponding steps of the methods described above in connection with the preferred embodiments of multimedia in fig. 2 or 3; in a specific implementation, first computer program instructions in a computer storage medium are loaded by the processor 301 and perform the steps of:

In one embodiment, the first computer program instructions are loaded by the processor 301 and executed in particular when determining a target cover of the multimedia composition from the candidate cover set based on the user representation:

In yet another embodiment, the first computer program instructions are loaded by the processor 301 and executed in particular in determining a target multimedia clip from the user representation:

and intercepting the target multimedia to obtain a target multimedia fragment.

In yet another embodiment, the first computer program instructions may also be loaded and executed by the processor 301:

In yet another embodiment, the first computer program instructions are loaded by the processor 301 and executed in particular when determining a recommended cover for the multimedia clip from the candidate cover set based on the similarity:

searching the covers of the similar multimedia fragments from the candidate cover set, and taking the searched covers of the similar multimedia fragments as recommended covers of the multimedia editing.

In yet another embodiment, the multimedia clips in the multimedia composition are arranged sequentially; accordingly, the first computer program instructions may also be loaded and executed by the processor 301:

In yet another embodiment, the processor 301 may further load and execute second computer program instructions stored in a computer storage medium to implement the corresponding steps described above in connection with the method in the preferred embodiment of multimedia in fig. 6; in a specific implementation, the second computer program instructions in the computer storage medium are loaded by the processor 301 and perform the steps of:

In one embodiment, the second computer program instructions are further loadable and executable by the processor 301 when selecting a target multimedia clip from the multimedia composition that matches the target element:

In yet another embodiment, the second computer program instructions may also be loaded and executed by the processor 301:

In yet another embodiment, the multimedia clips in the multimedia composition are arranged sequentially; accordingly, the second computer program instructions may also be loaded and executed by the processor 301:

The foregoing disclosure is illustrative of the present invention and is not to be construed as limiting the scope of the invention, which is defined by the appended claims.

Claims

1. A multimedia recommendation method, comprising:

acquiring a user portrait, and determining a target multimedia fragment according to the user portrait, wherein the first playing probability of the target multimedia fragment is larger than a probability threshold value, and the user portrait is used for reflecting the multimedia preference of a user indicated by the user portrait;

if the target multimedia fragment does not exist in the multimedia editing, calculating the similarity of the target multimedia fragment and each multimedia fragment in the multimedia editing, and determining a target cover of the multimedia editing from the candidate cover set according to the similarity;

2. The method of claim 1, wherein the method further comprises:

3. The method of claim 1, wherein said determining a target multimedia clip from said user representation comprises:

and intercepting the target multimedia to obtain a target multimedia fragment.

4. The method of claim 2, wherein the method further comprises:

5. The method of any one of claims 1-4, wherein the method further comprises:

The multimedia fragments in the multimedia editing are as follows: intercepting a segment obtained from multimedia meeting preset conditions, wherein the preset conditions comprise any one of the following: the play quantity is larger than a preset play quantity threshold, the effective play times is larger than a preset times threshold, and the click rate is larger than a preset click rate threshold;

6. The method of claim 5, wherein the multimedia clips in the multimedia composition are arranged sequentially; the method further comprises the steps of:

7. A multimedia recommendation method, comprising:

if the elements corresponding to the multimedia fragments do not comprise the target elements, calculating the similarity between the elements corresponding to the multimedia fragments and the target elements, and determining a target multimedia fragment according to the similarity, wherein the target multimedia fragment is a multimedia fragment corresponding to an element with similarity greater than a similarity threshold;

selecting a cover corresponding to the target multimedia fragment from the candidate cover set as a target cover of the multimedia editing;

8. The method of claim 7, wherein the method further comprises:

and if the element corresponding to each multimedia fragment comprises the target element, taking the multimedia fragment corresponding to the target element as a target multimedia fragment.

9. The method of claim 7, wherein the method further comprises:

10. The method of any one of claim 7 to 9, wherein,

the cover corresponding to the multimedia fragment comprises: static covers or dynamic covers; the static cover is a cover formed by a frame of image intercepted from the multimedia segment, and the dynamic cover is a cover formed by a plurality of frames of image intercepted from the multimedia segment.

11. The method of claim 10, wherein the multimedia clips in the multimedia composition are arranged sequentially; the method further comprises the steps of:

12. A multimedia recommendation device, comprising:

the acquisition unit is further used for acquiring a user portrait, determining a target multimedia fragment according to the user portrait, wherein the first playing probability of the target multimedia fragment is larger than a probability threshold, and the user portrait is used for reflecting the multimedia preference of a user indicated by the user portrait;

The obtaining unit is further configured to calculate a similarity of the target multimedia segment and each multimedia segment in the multimedia composition if the target multimedia segment does not exist in the multimedia composition, and determine a target cover of the multimedia composition from the candidate cover set according to the similarity;

13. A multimedia recommendation device, comprising:

the acquisition unit is also used for acquiring the user portrait and determining a target element according to the user portrait, wherein the target element is used for reflecting the multimedia preference of the user indicated by the user portrait;

the selecting unit is also used for acquiring elements corresponding to each multimedia fragment in the multimedia editing;

The selecting unit is further configured to calculate a similarity between the element corresponding to each multimedia segment and the target element if the element corresponding to each multimedia segment does not include the target element, and determine a target multimedia segment according to the similarity, where the target multimedia segment is a multimedia segment corresponding to an element whose similarity is greater than a similarity threshold;

the selecting unit is further configured to select, from the candidate cover set, a cover corresponding to the target multimedia clip as a target cover of the multimedia clip;

14. A server comprising a processor, a transceiver and a memory, the processor, transceiver and memory being interconnected, wherein the memory is adapted to store a computer program, the computer program comprising first program instructions, the processor being configured to invoke the first program instructions to perform the multimedia recommendation method of any of claims 1-6; alternatively, the computer program comprises second program instructions, the processor being configured to invoke the second program instructions to perform the multimedia recommendation method of any of claims 7-11.

15. A computer storage medium storing first computer program instructions adapted to be loaded by a processor and to perform the multimedia recommendation method of any one of claims 1-6; alternatively, the computer storage medium stores second computer program instructions adapted to be loaded by a processor and to perform the multimedia recommendation method according to any of the claims 7-11.