Generation method and device of episode promotion video, equipment and medium

Abstract

The invention provides an episode promotion video generation method and device, equipment and a medium, and relates to the field of artificial intelligence, in particular to the technical field of computer vision and deep learning. The specific implementation scheme is as follows: segmenting a video in a target episode to obtain shot segments of the video; the target episode comprises at least one video; determining an initial shot segment, an ending shot segment and a middle shot segment from the shot segments of the videos; the time sequence of the middle lens segment is located between the initial lens segment and the ending lens segment; the initial shot segment is a shot segment with a highlight plot in the video, and the ending shot segment is a shot segment with a suspension plot in the video; and generating a promotion video of the target episode according to the initial shot fragment, the ending shot fragment and the middle shot fragment. The production efficiency of the episode promotion video can be improved, and the labor cost is saved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer vision and deep learning technology in the field of artificial intelligence, and in particular to a method, device, equipment and medium for generating a promotional video for a TV series. Background Art

[0002] Watching movies and TV series has become a way for people to relax in their daily lives, for example, watching movies and TV series, or watching short plays.

[0003] However, the content of the series is highly coherent, and users cannot quickly understand the content of a specific series in fragmented time. Therefore, it is necessary to produce attractive promotional videos for the series so that users can quickly understand the content of the series in fragmented time while attracting users to watch the full content of the series. Summary of the invention

[0004] The present disclosure provides a method, device, equipment and medium for generating a promotional video of a drama.

[0005] According to a first aspect of the present disclosure, a method for generating a promotional video of a drama is provided, comprising:

[0006] Segmenting the videos in the target series to obtain the shot segments of the videos; wherein the target series includes at least one video;

[0007] From the shot clips of each video, determine an initial shot clip, an ending shot clip, and an intermediate shot clip; the timing of the intermediate shot clip is between the initial shot clip and the ending shot clip; the initial shot clip is a shot clip with a highlight plot in the video, and the ending shot clip is a shot clip with a suspense plot in the video;

[0008] A promotional video of the target series is generated according to the initial shot segment, the ending shot segment, and the intermediate shot segments.

[0009] According to a second aspect of the present disclosure, a device for generating a promotional video of a drama is provided, comprising:

[0010] A segmentation unit, configured to segment the videos in the target series to obtain the shot segments of the videos; wherein the target series includes at least one video;

[0011] A determination unit is used to determine an initial shot segment, an ending shot segment, and an intermediate shot segment from the shot segments of each video; the timing of the intermediate shot segment is between the initial shot segment and the ending shot segment; the initial shot segment is a shot segment with a highlight plot in the video, and the ending shot segment is a shot segment with a suspense plot in the video;

[0012] A generating unit is used to generate a promotional video of the target series according to the initial shot segment, the ending shot segment, and the intermediate shot segments.

[0013] According to a third aspect of the present disclosure, a computer program product is provided, comprising: a computer program, wherein the computer program is stored in a readable storage medium, at least one processor of an electronic device can read the computer program from the readable storage medium, and the at least one processor executes the computer program so that the electronic device executes the method described in the first aspect.

[0014] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause the computer to execute the method described in the first aspect.

[0015] The technology disclosed herein improves the production efficiency of promotional videos for TV series and saves labor costs.

[0016] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0018] Figure 1 is a scene graph that can implement the embodiments of the present disclosure;

[0019] Figure 2 is a schematic diagram of an application interface of an application provided by the present disclosure;

[0020] Figure 3 It is a flowchart of a method for generating a promotional video of a drama provided by the first embodiment of the present disclosure;

[0021] Figure 4 It is a flowchart of a method for generating a promotional video of a drama provided by the second embodiment of the present disclosure;

[0022] Figure 5It is a flowchart of a method for generating a promotional video of a drama provided by the third embodiment of the present disclosure;

[0023] Figure 6 It is a schematic diagram of the correspondence between a lens segment and a vocal segment provided by the present disclosure;

[0024] Figure 7 It is a structural block diagram of a device for generating a promotional video of a drama provided by an embodiment of the present disclosure;

[0025] Figure 8 It is a block diagram of an electronic device used to implement the method for generating a promotional video of a drama in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0026] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0027] Watching TV series has become a part of people's daily leisure and entertainment life, but the content of TV series is usually coherent and there are many episodes, so users cannot understand all the contents of the series in a short time. Therefore, in the related technology, promotional videos of TV series are produced by manual editing, so that users can understand the content of the series in a short time and attract users to watch the full content of the series.

[0028] However, this method requires manual editing of the series, which has low production efficiency and high labor costs.

[0029] In view of this, the present disclosure provides a method for generating a promotional video for a TV series, which automatically edits the video of a target TV series to generate a promotional video for the target TV series, thereby improving production efficiency and saving labor costs.

[0030] The present invention discloses a method, device, equipment and medium for generating a promotional video for a TV series, which are applied to the fields of computer vision and deep learning technology in the field of artificial intelligence to improve the efficiency of producing promotional videos for the TV series and save labor costs.

[0031] It should be noted that in the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0032] Figure 1is a scene graph that can implement the embodiments of the present disclosure. Figure 1 As shown, the scene at least includes an electronic device 100, which can be an electronic device with processing capabilities such as a laptop, a smart phone, a desktop computer, a server, a tablet computer, etc. The electronic device 100 is configured with an application 101, a display screen 102, a memory 103, and a processor 104. Among them, the user can start and run the application 101 through the electronic device 100. After the application 101 is started, at least one video of the target episode selected by the user is received. The processor 104 processes the at least one video of the target episode based on the application 101, obtains a promotional video of the target episode, and displays it to the user through the display screen 102.

[0033] Optionally, the electronic device 100 provides the function of generating a promotional video of the target episode through the above-mentioned components working together. The communication mode between the above-mentioned components is usually realized through a bus system or a specific interface protocol inside the electronic device. The application 101 is an interface for the user to interact with the electronic device 100, which is used to display the content of the promotional video, receive user input (such as search, click, etc.) and call the functions of other components. The application 101 is usually installed on the operating system of the electronic device 100 and interacts with the underlying hardware and software through the system API (application programming interface). The display screen 102 is used to display the content of the promotional video and other visual information. The display screen 102 is usually connected to the processor 104 through a video output interface (such as HDMI, DisplayPort, LVDS, etc.), which can transmit high-quality video signals to ensure clear display of the promotional video content. The memory 103 is used to store video files, application data and various files of the operating system. The memory 103 can be an internal memory (such as RAM, ROM, flash memory, etc.) or an external memory (such as SD card, hard disk, etc.). The memory 103 communicates with the processor 104 via a memory bus or a specific storage interface (such as SATA, PCIe, etc.) to achieve fast data reading and writing. The processor 104 is the core component of the electronic device 100, responsible for executing program instructions, processing data, and controlling other components. The processor 104 communicates with each component through an internal bus system, including data exchange with the memory 103, video signal transmission with the display screen 102, and instruction execution with the application 101.

[0034] Optional, reference Figure 2 As shown, Figure 21 is a schematic diagram of an application interface of an application provided by the present disclosure. The application interface of the application 101 may be displayed on the display screen 102 of the electronic device 100. The application 101 may provide an input control for the user, and the input control is used to determine the video of the target episode. The present disclosure does not limit how to determine the video of the target episode. For example, the user may upload the video of the target episode through the input control, or the stored video of the corresponding episode interval may be extracted from the memory 103 based on the episode interval of the target episode input by the user through the input control. Then, at least one video of the target episode is obtained through the application interface of the application.

[0035] Understandably, Figure 1 The scene diagram shown is only an exemplary description. In actual applications, the scene diagram may also include other devices, such as network devices, audio devices, etc., which can be adjusted according to actual needs, and the present disclosure does not limit them. The embodiments of the present disclosure do not limit the actual forms of the various devices included in the application scenario, nor do they limit the interaction methods between the devices. In the specific application of the solution, they can be set according to actual needs.

[0036] The following specific embodiments are used to describe in detail the technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0037] Figure 3 : is a flowchart of a method for generating a promotional video of a drama provided by the first embodiment of the present disclosure. The execution subject of this embodiment may be the aforementioned electronic device. The method may include the following steps:

[0038] S301: Segment the video in the target series to obtain video clips.

[0039] For example, the episode represents a drama program played on a TV channel or an online video platform. The present disclosure does not limit the content and form of the episode. For example, the episode can be in the form of a TV series or a short play. The target episode is the episode for which a promotional video is to be generated. The target episode includes at least one video. For example, the target episode can be a short play A, which has 20 episodes.

[0040] It should be noted that the present disclosure does not limit the number of videos included in the target series. The target series may include all videos of the target series, for example, 20 episodes of the short series A; or the target series may include part of the videos of the target series, for example, 1-5 episodes of the short series A, or 10-18 episodes, which depends on the user's settings.

[0041] The shot segments represent the segments of different shots in the video, and there are shot boundaries between each shot segment. It should be noted that the present disclosure does not limit the duration of each shot segment, which is specifically related to the content of the video.

[0042] After receiving the video of the target series, the electronic device can segment each video to obtain the shot segments of the video. For example, the video can be input into a preset shot boundary detection model to segment the video to obtain multiple shot segments of the video. It should be noted that the present disclosure does not limit the type of shot boundary detection model, as long as the function of shot boundary detection can be realized, for example, it can be a Translation-Based Network Representation Learning for Social Relation Extraction (TransNet) model.

[0043] S302: Determine an initial shot segment, an ending shot segment, and an intermediate shot segment from the shot segments of each video.

[0044] Exemplarily, the initial shot segment represents the first shot segment of the promotional video of the target series, the ending shot segment represents the last shot segment of the promotional video of the target series, and the middle shot segment represents the shot segment between the initial shot segment and the ending shot segment, that is, the timing of the middle shot segment is between the initial shot segment and the ending shot segment. Among them, the initial shot segment is a shot segment with a highlight plot in the video, and the ending shot segment is a shot segment with a suspense plot in the video. It should be noted that the highlight plot represents a wonderful and interesting plot, which is more attractive to the audience than other plots. The suspense plot represents a plot with doubts and uncertainties, which attracts the audience to be curious and expectant about the development of the subsequent plot, and arouses the audience's interest in watching.

[0045] In some possible implementations, the electronic device may start from the shot clip of the first video among all the videos of the target series, and determine in sequence whether each shot clip has a highlight plot, and if so, determine the first shot clip with a highlight plot as the initial shot clip. If none of the shot clips of all the videos of the target series have a highlight plot, a prompt message is output to the user to prompt the user that there is no highlight plot in the video of the current target series.

[0046] Similarly, after determining the initial shot clip, the electronic device can start from the first shot clip with a highlight plot in all videos, and determine in sequence whether each shot clip after the initial shot clip has a suspense plot. If so, a shot clip with a suspense plot can be randomly determined as the ending shot clip; or a shot clip with a suspense plot can be determined as the ending shot clip based on the duration between the initial shot clip and the shot clip with a suspense plot, or a shot clip with a suspense plot adjacent to the initial shot clip can be determined as the ending shot clip. The present disclosure does not limit this, and can be specifically set according to actual needs. After determining the initial shot clip and the ending shot clip, the shot clip with a timing between the two is determined as the middle shot clip.

[0047] Among them, regarding how to determine whether a shot clip has a highlight plot, a possible implementation method is to input the shot clip into a preset video classification model, and output whether the shot clip is a shot clip with a highlight plot. Among them, the present disclosure does not limit the type of the video classification model, for example, it can be a video masked autoencoder (VideoMAE) model. Another possible implementation method is that the electronic device can obtain the text recognition information of the shot clip, and then based on a preset prompt word template, the dialogue information of the shot clip can be filled into the prompt word template to obtain the prompt word, and the prompt word is input into the preset large language model to output whether the shot clip has a highlight plot. Among them, the text recognition information represents the dialogue information of the shot clip, and the shot clip can be recognized by optical character recognition technology (Optical Character Recognition, OCR) to obtain the text recognition information of the shot clip.

[0048] How to determine whether the shot clip contains suspenseful plot is similar to the method of determining the highlight plot mentioned above, and will not be repeated here.

[0049] In some possible implementations, the electronic device may pre-process the lens segments of each video to remove invalid lens segments in the lens segments of each video to obtain valid lens segments of each video, wherein the invalid lens segments represent segments that are not related to the plot of the target series, such as the opening, ending, and advertisement lens segments. Furthermore, the above-mentioned implementation method may be adopted for the valid lens segments of each video to determine the initial lens segment, the ending lens segment, and the intermediate lens segment. Among them, as to how to identify the invalid lens segments in the lens segments, for example, it may be based on a preset lens classification model to identify the category of each lens segment, and then determine the invalid lens segment. The present disclosure does not limit the type of the lens classification model, such as a convolutional neural network model. In this way, the invalid lens segments in the video can be deleted, so that the generated promotional video only retains the lens segments related to the plot of the target series, thereby improving the effectiveness of the content in the promotional video.

[0050] S303: Generate a promotional video for the target series based on the initial shot segment, the ending shot segment, and the intermediate shot segments.

[0051] In a possible implementation, after obtaining the initial shot segment, the ending shot segment, and the middle shot segment, the electronic device can splice the three shot segments in time sequence to obtain a promotional video for the target series.

[0052] In another possible implementation, the electronic device may splice the above three types of footage in time sequence to obtain an initial promotional video, and then beautify the initial promotional video to obtain a promotional video of the target series. The beautification operation may include, for example, modifying the filter of the initial promotional video, adding background music, adding the name and episode number of the target series, generating copy information of the promotional video, etc., which is not limited in the present disclosure.

[0053] The promotional video obtained through this step not only has the initial footage of the highlight plot, but also includes the ending footage of the suspense plot, and the intermediate footage between the two has logical coherence with the initial footage and the ending footage, so that the audience can understand the content of the target series based on the coherent plot logic, and at the same time, the audience is attracted to watch the complete content of the target series based on the highlight plot and the suspense plot.

[0054] The method for generating a promotional video for a TV series provided in this embodiment is that the electronic device obtains the shot segments of each video after segmenting each video of the target TV series, and then determines the initial shot segment with a highlight plot, the ending shot segment with a suspense plot, and the middle shot segment between the initial shot segment and the ending shot segment based on the shot segments of all videos, and finally automatically generates a promotional video for the target TV series based on the initial shot segment, the ending shot segment, and the middle shot segment. In this way, the electronic device can automatically generate a promotional video based on the video of the target TV series without manual editing, thereby improving the production efficiency of the promotional video and saving labor costs.

[0055] Figure 4 1 is a flow chart of a method for generating a promotional video of a TV series provided by the second embodiment of the present disclosure. The execution subject of this embodiment may be the aforementioned electronic device. Based on the aforementioned embodiment, this embodiment describes in detail how to determine the initial shot segment, the ending shot segment, and the intermediate shot segment from the shot segments of each video. The method may include the following steps:

[0056] S401: Segment the video in the target series to obtain video clips.

[0057] It should be noted that this step is similar to the aforementioned step S301 and will not be repeated here.

[0058] S402: Identify the shot category of the shot segment.

[0059] For example, the shot category represents the category of the plot included in the shot clip, and the present disclosure does not limit the content of the shot category. For example, the shot category is any one of the following: a shot category representing a shot clip including a series cover, a shot category representing a shot clip including an ending, a shot category representing a shot clip including overlapping plots, and a shot category representing a shot clip including an advertisement.

[0060] By setting the shot category, each shot segment can be classified using the shot category, which is helpful for subsequently removing invalid shot segments accurately.

[0061] In some possible implementations, the electronic device may preset a shot classification model, which may be pre-trained based on shot segments with labels, the labels including the above-mentioned multiple shot categories. Therefore, the electronic device may input each shot segment into the shot classification model, and output the shot category of the shot segment.

[0062] In some possible implementations, the electronic device may set a different identification method for each shot category, so as to accurately determine the shot category of each shot segment, improve the accuracy of shot category identification, and prevent the phenomenon that the shot segment is accidentally deleted due to incorrect identification of the shot category.

[0063] By way of example, the following methods may be included:

[0064] Method 1: Extract the first frame of the first shot clip from the shot clips of the video; if it is determined that the first frame is the cover of the drama, determine that the shot category of the first shot clip is a shot category that includes the cover of the drama.

[0065] Among them, the electronic device can input the first frame image and the prompt word into a preset visual semantic model (Vision-Language Model, VLM), and output whether the first frame image is the cover of the series, wherein the prompt word can be, for example, "Please confirm whether this picture is the cover picture of the series", and the present disclosure does not limit the content of the prompt word. The electronic device can also pre-store the cover picture of the target series, respectively determine the feature vector of the cover picture of the target series and the feature vector of the first frame image, and then determine the similarity of the two feature vectors. If it is determined that the similarity is greater than or equal to a preset threshold, the first frame image is determined to be the cover of the series. If it is less than the preset threshold, it is determined that the first frame image is not the cover of the series. Among them, the method of determining the feature vector can be, for example, inputting the picture into a preset calculation model, and the calculation model can be, for example, a pre-training (Contrastive Language-Image Pre-Training, Clip) model based on contrastive text-image pairs, or other models, which are not limited by the present disclosure.

[0066] In this way, the electronic device can accurately identify whether the shot clip belongs to the shot category that includes the episode cover, and only the first shot clip of each video needs to be identified, which reduces the number of shot clips to be identified and improves identification efficiency.

[0067] Method 2: Obtain voice recognition information of the shot segments of the video; based on the voice recognition information, starting from the last shot segment of the video, determine in sequence whether there is voice recognition information in the shot segments of the video until a shot segment with voice recognition information is determined, and determine the shot category of the shot segments of the video that are located after the shot segment with voice recognition information as the shot category that includes the end credits.

[0068] Among them, the voice recognition information represents the text information of the voice included in the shot segment. It should be understood that there will be voice information in the shot segment, and the voice information in the shot segment can be recognized by automatic speech recognition technology (Automatic Speech Recognition, ASR) and converted into text information. It should be noted that the voice information in the shot segment does not include music information. The electronic device can determine whether there is voice recognition information in the shot segment of each video, starting from the last shot segment of the video. If not, it is determined whether the previous shot segment of the shot segment exists, until a shot segment with voice recognition information is determined, and the shot category of other shot segments of the video after the shot segment with voice recognition information is determined as a shot category including the end credits. For example, the video includes shot segments 1-10, wherein there is no voice recognition information in shot segments 8-10, and there is voice recognition information in shot segment 7, and the shot category of shot segments 8-10 is determined to be a shot category including the end credits.

[0069] In this way, the electronic device can accurately identify whether the shot segment is a shot category that includes the end credits, and there is no need to perform identification processing on all the shot segments of each video, thereby reducing the number of shot segments to be identified and improving identification efficiency.

[0070] Method 3: For each video of the target series, obtain the second text identification information of the shot clips of the video and the third text identification information of the shot clips of the previous episode of the video; for each shot clip of the video, if it is determined that the similarity between the second text identification information of the shot clip and the third text identification information of the shot clip of the previous episode of the video is greater than or equal to a preset threshold, then determine that the shot category of the shot clip is a shot category that includes overlapping plots.

[0071] The text recognition information represents the dialogue information in the shot segment. The method for obtaining the text recognition information can refer to the above-mentioned embodiment and will not be described in detail here. For example, taking the video as the third episode video as an example, the second text recognition information of the shot segment of the third episode video and the third text recognition information of the shot segment of the second episode video are obtained. For each shot segment of the third episode video, the similarity between the second text recognition information of the shot segment and the third text recognition information of each shot segment of the second episode is determined. If the similarity is greater than or equal to the preset threshold, the shot category of the shot segment of the third episode is determined to be a shot category including overlapping plots.

[0072] The electronic device may also preset m shot segments at the end of the second episode and n shot segments at the beginning of the third episode, and perform similarity judgment on the second text recognition information of the n shot segments at the beginning of the third episode and the third text recognition information of the shot segments at the end of the second episode, so as to determine whether the shot category of the shot segments of the third episode is a shot category including overlapping plots, thereby reducing the number of shot segments to be recognized and improving recognition efficiency. Wherein, m and n are both integers greater than or equal to 1.

[0073] In this way, the electronic device can accurately identify whether the shot segment is a shot category that includes overlapping plots.

[0074] Method 4: extract any frame of picture from the lens segment of the video; if it is determined that the any frame of picture is an advertisement picture, determine the lens category of the lens segment to be a lens category including advertisements.

[0075] Exemplarily, the electronic device can input the arbitrary frame of the picture and the prompt word into a preset visual semantic model (Vision-Language Model, VLM), and output whether the arbitrary frame of the picture is an advertising picture, where the prompt word can be, for example, "Please confirm whether this picture is an advertising picture." The present disclosure does not limit the content of the prompt word.

[0076] In this way, the electronic device can accurately identify whether the footage segment is a footage category including advertisements.

[0077] S403: Determine an initial shot segment, an ending shot segment, and an intermediate shot segment from the shot segments of each video according to the shot categories of the shot segments.

[0078] Exemplarily, the electronic device may delete the shot segments of each video whose shot categories are the above-mentioned shot categories according to the shot categories of the shot segments, and obtain valid shot segments in each video. The valid shot segments may represent shot segments that are related to the plot and do not repeat the plot, and then determine the initial shot segment, the ending shot segment, and the intermediate shot segments from the valid shot segments. This implementation method is similar to that in the aforementioned embodiment and will not be repeated here.

[0079] S404: splice the initial shot segment, the ending shot segment, and the intermediate shot segments to obtain a promotional video for the target series.

[0080] For example, after obtaining the initial shot segment, the ending shot segment, and the middle shot segment, the electronic device may splice the three shot segments in time sequence to obtain a promotional video for the target series.

[0081] In this way, the electronic device can automatically generate a promotional video with a coherent and logical plot content based on the time sequence, thereby improving the efficiency of generating the promotional video.

[0082] The method for generating a promotional video for a TV series provided in this embodiment is that the electronic device can obtain the lens segments of each video after segmenting each video of the target TV series, and then identify the lens category of the lens segment, delete the invalid lens segments in the video according to the lens category of the lens segment, obtain the valid lens segments of the video, and then determine the initial lens segment, the ending lens segment, and the middle lens segment from the valid lens segments of all videos, and finally automatically generate the promotional video of the target TV series according to the initial lens segment, the ending lens segment, and the middle lens segment. In this way, the electronic device can delete the invalid lens segments in the video based on the lens segment category, ensure that only the lens segments related to the plot and without repeated plots are retained in the generated promotional video, improve the production efficiency of the promotional video, save labor costs, and improve the effectiveness of the plot content in the promotional video.

[0083] Figure 5 : is a flowchart of a method for generating a promotional video of a drama provided by the third embodiment of the present disclosure. The execution subject of this embodiment may be the aforementioned electronic device. Based on the aforementioned embodiment, this embodiment describes in detail how to determine the initial shot segment, the ending shot segment, and the intermediate shot segment from the shot segments of each video according to the shot category of the shot segment. The method may include the following steps:

[0084] S501: Segment the video in the target series to obtain video clips.

[0085] It should be noted that this step is similar to the aforementioned step S301 and will not be repeated here.

[0086] S502: Identify the shot category of the shot segment.

[0087] It should be noted that this step is similar to the aforementioned step S401 and will not be repeated here.

[0088] S503: Identify the human voice in the audio corresponding to the video, and obtain the human voice segment corresponding to the video.

[0089] Exemplarily, each video has an audio corresponding to the video, and the audio is used to represent all sound information in the video, including voice information and background music. The vocal segment represents a segment of voice information in the video, and the vocal segment has a start and end time, for example, 0-5s is vocal segment 1, and 5-7s is vocal segment 2.

[0090] In a possible implementation, the electronic device may input the audio corresponding to the video into a preset speech recognition model, and output each vocal segment in the audio file. The present disclosure does not limit the type of the speech recognition model, for example, it may be a Voice Activity Detection (VAD) model.

[0091] S504: Determine the to-be-processed shot segments of the video from among the shot segments of the video according to the shot categories of the shot segments in the video and the vocal segments corresponding to the video.

[0092] Exemplarily, the shot segments to be processed represent shot segments that are aligned with the vocal segments of the video in time. It should be understood that when the electronic device performs shot segmentation processing on the video, it does not consider the continuity of the sound, resulting in the same voice being divided by the shot segments. If the shot segments are deleted, there will be a situation where the voice is discontinuous. Therefore, it is necessary to adjust the shot segments corresponding to the shot category according to the vocal segments to ensure the continuity of the voice.

[0093] For example, for each shot segment of a video, the shot segment corresponding to the shot category can be determined, and then the vocal segment corresponding to the shot segment corresponding to the shot category can be determined. Starting from the vocal segment, a vocal segment that is aligned with the shot segment in time after the vocal segment is determined. The shot segment aligned with the vocal segment and the shot segment after the shot segment aligned with the vocal segment are determined as the shot segments to be processed.

[0094] Specifically, according to the shot category of the shot clips in the video, a first vocal clip is determined from the vocal clips corresponding to the video; wherein the first vocal clip represents a vocal clip that temporally includes the shot clips corresponding to the shot category; according to the first vocal clip, the shot clips to be deleted in the video are determined; wherein the shot clips to be deleted represent shot clips that are not temporally aligned with other vocal clips located after the first vocal clip; from each shot clip of the video, the shot clips corresponding to the shot category and the shot clips to be deleted are deleted to obtain the shot clips to be processed of the video.

[0095] For example, refer to Figure 6 As shown, Figure 6This is a schematic diagram of the correspondence between a lens segment and a vocal segment provided by the present disclosure. Take the video including lens segment 1 (0-5s), lens segment 2 (5-15s), lens segment 3 (15-30s), lens segment 4 (30-60s) and lens segment 5 (60-70s), and the vocal segments corresponding to the video include vocal segment 1 (1-6s), vocal segment 2 (7-17s) and vocal segment 3 (30-69s) as an example. Among them, the lens category of lens segment 1 is the lens category including the cover of the drama.

[0096] refer to Figure 6 As shown, the electronic device can determine the first vocal segment that temporally includes the shot segment according to the shot category of the shot segment. Taking shot segment 1 as an example, the first vocal segment is vocal segment 1. Since shot segment 1 needs to be deleted in subsequent processing, it can be seen from vocal segment 1 that if only shot segment 1 is deleted, the voice information may be incomplete. Therefore, shot segment 2 corresponding to vocal segment 1 needs to be deleted as well. Figure 6 It can be seen that the shot segment 2 is still within the time range of the vocal segment 2. If the shot segment 2 is deleted, the voice information may still be incomplete. Therefore, it is necessary to identify the shot segments (shot segments 2 and shot segments 3) that are not aligned in time with the other vocal segments after the first vocal segment after the first vocal segment after the first vocal segment as the shot segments to be deleted. The shot segment 4 is aligned in time with the vocal segment 3, so it does not need to be deleted.

[0097] It should be noted that the above only takes the shot category including the series cover as an example to illustrate how to determine the shot segments to be deleted. The processing method for shot segments of other shot categories in the video is similar and will not be repeated here.

[0098] Through the above method, the electronic device can determine the shot segments to be deleted for each shot segment of the video, and then delete the shot segment corresponding to the shot category (shot segment 1) and the shot segments to be deleted (shot segment 2 and shot segment 3), and obtain the shot segments to be processed, wherein the shot segments to be processed are aligned with the vocal segments in time, ensuring that there is no incomplete voice information in the shot segments to be processed, thereby improving the quality of the promotional video.

[0099] S505 : Determine an initial shot segment, an ending shot segment, and an intermediate shot segment according to the shot segments to be processed of each video.

[0100] In this step, the footage clips to be processed are obtained by processing the footage clips of each video, and then the initial footage clip, the ending footage clip, and the middle footage clip are determined from the footage clips to be processed of each video, thereby reducing the number of footage clips that need to be identified and ensuring that the generated promotional video only includes valid footage clips, thereby improving the effectiveness of the content of the promotional video.

[0101] Exemplarily, the following method may be adopted to determine the initial shot segment, the ending shot segment, and the intermediate shot segment from the to-be-processed shot segments of each video. It should be noted that this step is similar to the method of the aforementioned embodiment, and the specific description may refer to the aforementioned embodiment.

[0102] (1) Initial shot segments: Starting from the first shot segment to be processed in each video, based on a preset video classification model, determine whether there is a shot segment with a highlight plot in each shot segment to be processed; if so, determine the first shot segment to be processed with a highlight plot as the initial shot segment.

[0103] In this way, the electronic device can determine the initial shot segments from the shot segments to be processed without identifying invalid shot segments of the video (titles, advertisements, repeated plots, etc.), thereby reducing the number of shot segments that need to be identified and improving recognition efficiency.

[0104] (2) The ended shot segment: obtaining the first text recognition information of the shot segment to be processed of each video; wherein the first text recognition information represents the dialogue information in the shot segment to be processed; from the shot segments to be processed of each video located after the initial shot segment, determining whether the shot segment to be processed is a shot segment with a suspenseful plot according to the text recognition information of the shot segment to be processed; if so, determining the ended shot segment according to the duration from the initial shot segment to the shot segment to be processed with a suspenseful plot.

[0105] For example, if the duration from the determined shot segment with suspense plot adjacent to the initial shot segment to the initial shot segment is less than a preset duration, such as 3 minutes, then you can continue to determine a shot segment with suspense plot that meets the preset duration based on the shot segment to be processed until the ending shot segment is determined.

[0106] In this way, the electronic device can determine the ending shot segment from the shot segments to be processed without identifying invalid shot segments of the video (ending credits, advertisements, etc.), thereby reducing the number of shot segments that need to be identified and improving identification efficiency.

[0107] (3) Middle shot segment: After the initial shot segment and the final shot segment are determined, the shot segment that is in between the two in terms of timing is determined as the middle shot segment.

[0108] S506: splice the initial shot segment, the ending shot segment, and the middle shot segment to obtain a promotional video for the target series.

[0109] It should be noted that this step is similar to the aforementioned step S404 and will not be repeated here.

[0110] In the method for generating a promotional video for a TV series provided in this embodiment, after the electronic device identifies the shot category for the shot segments of the video, it can determine the shot segments to be deleted according to the human voice segments and shot categories corresponding to the video, and delete the shot segments corresponding to the shot categories and the shot segments to be deleted from the shot segments corresponding to the video to obtain the shot segments to be processed, and then determine the initial shot segments, the ending shot segments, and the middle shot segments from the shot segments to be processed, and splice them to obtain the promotional video of the target TV series. In this way, the electronic device can ensure the integrity of the voice information in the shot segments to be processed based on the human voice segments, improve the quality of the promotional video, and thus enhance the user experience.

[0111] Figure 7 700 is a structural block diagram of a device for generating a promotional video of a TV series provided by an embodiment of the present disclosure. The device 700 includes the following units:

[0112] The segmentation unit 701 is used to segment the videos in the target series to obtain the shot segments of the videos; wherein the target series includes at least one video;

[0113] The determining unit 702 is used to determine an initial shot segment, an ending shot segment, and an intermediate shot segment from the shot segments of each video; the timing of the intermediate shot segment is between the initial shot segment and the ending shot segment; the initial shot segment is a shot segment with a highlight plot in the video, and the ending shot segment is a shot segment with a suspense plot in the video;

[0114] The generating unit 703 is used to generate a promotional video of the target series according to the initial shot segment, the ending shot segment, and the intermediate shot segments.

[0115] In some possible implementations, the determining unit 701 includes:

[0116] An identification module, used for identifying the shot category of the shot segment;

[0117] The determination module is used to determine the initial shot segment, the ending shot segment, and the middle shot segment from the shot segments of each of the videos according to the shot categories of the shot segments.

[0118] In some possible implementations, the shot category is any one of the following: a shot category representing a shot clip that includes a series cover, a shot category representing a shot clip that includes an ending credit, a shot category representing a shot clip that includes overlapping plots, and a shot category representing a shot clip that includes advertisements.

[0119] In some possible implementations, the video has audio corresponding to the video; and the determining module includes:

[0120] An identification submodule, used to identify the human voice in the audio corresponding to the video, and obtain the human voice segment corresponding to the video;

[0121] A first determination submodule is used to determine, from the various shot segments of the video, a shot segment to be processed of the video according to the shot categories of the shot segments in the video and the vocal segments corresponding to the video; wherein the shot segment to be processed represents a shot segment that is aligned with the vocal segment of the video in time;

[0122] The second determination submodule is used to determine the initial shot segment, the ending shot segment, and the middle shot segment according to the to-be-processed shot segments of each of the videos.

[0123] In some possible implementations, the first determining submodule is specifically configured to:

[0124] According to the shot category of the shot segment in the video, determining a first vocal segment from the vocal segments corresponding to the video; wherein the first vocal segment represents a vocal segment that temporally includes the shot segment corresponding to the shot category;

[0125] Determine, according to the first vocal segment, a shot segment to be deleted in the video; wherein the shot segment to be deleted represents a shot segment that is not aligned in time with other vocal segments that are located after the first vocal segment;

[0126] From each shot segment of the video, the shot segment corresponding to the shot category and the shot segment to be deleted are deleted to obtain the shot segment to be processed of the video.

[0127] In some possible implementations, the second determining submodule is specifically configured to:

[0128] Starting from the to-be-processed shot segment of the first video in each of the videos, determining in sequence whether each to-be-processed shot segment has a highlight plot based on a preset video classification model;

[0129] If so, the first shot segment to be processed with a highlight plot is determined as the initial shot segment.

[0130] In some possible implementations, the second determining submodule is specifically configured to:

[0131] Acquire first text recognition information of the to-be-processed shot segments of each of the videos; wherein the first text recognition information represents the dialogue information in the to-be-processed shot segments;

[0132] From the to-be-processed shot segments of each of the videos located after the initial shot segment, determining whether the to-be-processed shot segment is a shot segment with a suspenseful plot according to the text recognition information of the to-be-processed shot segment;

[0133] If so, the ending shot segment is determined according to the duration from the initial shot segment to the to-be-processed shot segment with suspenseful plot.

[0134] In some possible implementations, the identification module includes:

[0135] A first extraction submodule is used to extract the first frame image of the first shot segment from the shot segments of the video;

[0136] The third determination submodule is used to determine that the shot category of the first shot segment is a shot category including a drama cover when it is determined that the first frame image is a drama cover.

[0137] In some possible implementations, the identification module includes:

[0138] A first acquisition submodule is used to acquire speech recognition information of the shot segment of the video; wherein the speech recognition information represents text information of the speech included in the shot segment;

[0139] The fourth determination submodule is used to determine whether the shot segments of the video have the voice recognition information, starting from the last shot segment of the video, according to the voice recognition information, until a shot segment with voice recognition information is determined, and the shot category of the shot segment of the video that is located after the shot segment with voice recognition information is determined as a shot category that includes the end credits.

[0140] In some possible implementations, the identification module includes:

[0141] The second acquisition submodule is used to acquire, for each video of the target episode, second text recognition information of the shot segments of the video, and third text recognition information of the shot segments of the previous episode of the video; wherein the text recognition information represents the line information in the shot segments;

[0142] The fifth determination submodule is used to determine, for each shot segment of the video, that the shot category of the shot segment is a shot category that includes overlapping plots if the similarity between the second text identification information of the shot segment and the third text identification information of the shot segment of the previous video of the video is greater than or equal to a preset threshold.

[0143] In some possible implementations, the identification module includes:

[0144] A second extraction submodule is used to extract any frame of picture from the shot segment of the video;

[0145] The sixth determination submodule is used to determine that the shot category of the shot segment is a shot category including advertisement when determining that the any one frame of picture is an advertisement picture.

[0146] In some possible implementations, the generating unit includes:

[0147] The splicing module is used to splice the initial shot segment, the ending shot segment, and the intermediate shot segment to obtain a promotional video of the target series.

[0148] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0149] According to an embodiment of the present disclosure, the present disclosure also provides a computer program product, which includes: a computer program, the computer program is stored in a readable storage medium, at least one processor of an electronic device can read the computer program from the readable storage medium, and at least one processor executes the computer program so that the electronic device executes the solution provided by any of the above embodiments.

[0150] Figure 8A schematic block diagram of an example electronic device 800 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0151] like Figure 8 As shown, the device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the device 800 can also be stored. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0152] A number of components in the device 800 are connected to the I / O interface 805, including: an input unit 806, such as a keyboard, a mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a disk, an optical disk, etc.; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the device 800 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0153] The computing unit 801 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 801 performs the various methods and processes described above, such as the method for generating a promotional video for an episode. For example, in some embodiments, the method for generating a promotional video for an episode may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 808. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded into the RAM 803 and executed by the computing unit 801, one or more steps of the method for generating a promotional video for an episode described above may be performed. Alternatively, in other embodiments, the computing unit 801 may be configured to execute the method for generating a promotional video for an episode in any other appropriate manner (eg, by means of firmware).

[0154] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0155] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0156] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0157] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0158] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0159] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server may also be a server of a distributed system, or a server combined with a blockchain.

[0160] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0161] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for generating a promotional video for a TV series, comprising: Segmenting the videos in the target series to obtain the shot segments of the videos; wherein the target series includes at least one video; From the shot clips of each video, determine an initial shot clip, an ending shot clip, and an intermediate shot clip; the timing of the intermediate shot clip is between the initial shot clip and the ending shot clip; the initial shot clip is a shot clip with a highlight plot in the video, and the ending shot clip is a shot clip with a suspense plot in the video; A promotional video of the target series is generated according to the initial shot segment, the ending shot segment, and the intermediate shot segments.

2. The method according to claim 1, wherein: Determining an initial shot segment, an ending shot segment, and an intermediate shot segment from the shot segments of each of the videos, including: Identifying a shot category of the shot clip; According to the shot categories of the shot segments, the initial shot segment, the ending shot segment, and the middle shot segment are determined from the shot segments of each of the videos.

3. The method according to claim 2, wherein: The shot category is any one of the following: a shot category representing a shot clip including a series cover, a shot category representing a shot clip including an ending, a shot category representing a shot clip including overlapping plots, and a shot category representing a shot clip including advertisements.

4. The method according to claim 2 or 3, wherein: The video has audio corresponding to the video; according to the shot categories of the shot segments, the initial shot segment, the ending shot segment, and the intermediate shot segment are determined from each of the shot segments, including: Identify the human voice in the audio corresponding to the video, and obtain the human voice segment corresponding to the video; According to the shot categories of the shot segments in the video and the vocal segments corresponding to the video, determine the shot segments to be processed of the video from the shot segments of the video; wherein the shot segments to be processed represent shot segments that are aligned with the vocal segments of the video in time; According to the to-be-processed shot segments of each of the videos, the initial shot segment, the ending shot segment, and the middle shot segment are determined.

5. The method according to claim 4, wherein: Determining the to-be-processed shot segments of the video from the shot segments of the video according to the shot categories of the shot segments in the video and the vocal segments corresponding to the video, including: According to the shot category of the shot segment in the video, determining a first vocal segment from the vocal segments corresponding to the video; wherein the first vocal segment represents a vocal segment that temporally includes the shot segment corresponding to the shot category; Determine, according to the first vocal segment, a shot segment to be deleted in the video; wherein the shot segment to be deleted represents a shot segment that is not aligned in time with other vocal segments that are located after the first vocal segment; From each shot segment of the video, the shot segment corresponding to the shot category and the shot segment to be deleted are deleted to obtain the shot segment to be processed of the video.

6. The method according to claim 4 or 5, wherein: Determining the initial shot segments according to the to-be-processed shot segments of each of the videos includes: Starting from the to-be-processed shot segment of the first video in each of the videos, determining in sequence whether each to-be-processed shot segment has a highlight plot based on a preset video classification model; If so, the first shot segment to be processed with a highlight plot is determined as the initial shot segment.

7. The method according to claim 4 or 5, wherein: Determining the ended shot segments according to the to-be-processed shot segments of each of the videos includes: Acquire first text recognition information of the to-be-processed shot segments of each of the videos; wherein the first text recognition information represents the dialogue information in the to-be-processed shot segments; From the to-be-processed shot segments of each of the videos located after the initial shot segment, determining whether the to-be-processed shot segment is a shot segment with a suspenseful plot according to the text recognition information of the to-be-processed shot segment; If so, the ending shot segment is determined according to the duration from the initial shot segment to the to-be-processed shot segment with suspenseful plot.

8. The method according to any one of claims 2 to 7, wherein: Identify the shot category of the shot clip, including: Extracting the first frame of the first shot segment from the shot segments of the video; If it is determined that the first frame image is a drama cover, then the shot category of the first shot segment is determined to be a shot category that includes a drama cover.

9. The method according to any one of claims 2 to 7, wherein: Identify the shot category of the shot clip, including: Acquire speech recognition information of the shot segment of the video; wherein the speech recognition information represents text information of the speech included in the shot segment; According to the voice recognition information, starting from the last shot segment of the video, determine in sequence whether the shot segments of the video contain the voice recognition information, until a shot segment containing the voice recognition information is determined, and determine the shot category of the shot segments of the video that are sequentially located after the shot segment containing the voice recognition information as the shot category that includes the end credits.

10. The method according to any one of claims 2 to 7, wherein: Identify the shot category of the shot clip, including: For each video of the target episode, obtain the second text recognition information of the shot segment of the video, and the third text recognition information of the shot segment of the previous episode of the video; wherein the text recognition information represents the line information in the shot segment; For each shot segment of the video, if it is determined that the similarity between the second text identification information of the shot segment and the third text identification information of the shot segment of the previous episode of the video is greater than or equal to a preset threshold, then the shot category of the shot segment is determined to be a shot category that includes overlapping plots.

11. The method according to any one of claims 2 to 7, wherein: Identify the shot category of the shot clip, including: Extract any frame of picture from the video footage; If it is determined that the any one frame of picture is an advertisement picture, then the shot category of the shot segment is determined to be a shot category including advertisements.

12. The method according to any one of claims 1 to 11, wherein: Generating a promotional video of the target series according to the initial shot segment, the ending shot segment, and the intermediate shot segment, including: The initial shot segment, the ending shot segment, and the intermediate shot segment are spliced ​​together to obtain a promotional video for the target series.

13. A device for generating a promotional video of a TV series, comprising: A segmentation unit, configured to segment the videos in the target series to obtain the shot segments of the videos; wherein the target series includes at least one video; A determination unit is used to determine an initial shot segment, an ending shot segment, and an intermediate shot segment from the shot segments of each video; the timing of the intermediate shot segment is between the initial shot segment and the ending shot segment; the initial shot segment is a shot segment with a highlight plot in the video, and the ending shot segment is a shot segment with a suspense plot in the video; A generating unit is used to generate a promotional video of the target series according to the initial shot segment, the ending shot segment, and the intermediate shot segments.

14. The device according to claim 13, wherein: The determining unit comprises: An identification module, used for identifying the shot category of the shot segment; The determination module is used to determine the initial shot segment, the ending shot segment, and the middle shot segment from the shot segments of each of the videos according to the shot categories of the shot segments.

15. The device according to claim 14, wherein: The shot category is any one of the following: a shot category representing a shot clip including a series cover, a shot category representing a shot clip including an ending, a shot category representing a shot clip including overlapping plots, and a shot category representing a shot clip including advertisements.

16. The device according to claim 14 or 15, wherein: The video has audio corresponding to the video; The determining module comprises: An identification submodule, used to identify the human voice in the audio corresponding to the video, and obtain the human voice segment corresponding to the video; A first determination submodule is used to determine, from the various shot segments of the video, a shot segment to be processed of the video according to the shot categories of the shot segments in the video and the vocal segments corresponding to the video; wherein the shot segment to be processed represents a shot segment that is aligned with the vocal segment of the video in time; The second determination submodule is used to determine the initial shot segment, the ending shot segment, and the middle shot segment according to the to-be-processed shot segments of each of the videos.

17. The device according to claim 16, wherein: The first determining submodule is specifically used for: According to the shot category of the shot segment in the video, determining a first vocal segment from the vocal segments corresponding to the video; wherein the first vocal segment represents a vocal segment that temporally includes the shot segment corresponding to the shot category; Determine, according to the first vocal segment, a shot segment to be deleted in the video; wherein the shot segment to be deleted represents a shot segment that is not aligned in time with other vocal segments that are located after the first vocal segment; From each shot segment of the video, the shot segment corresponding to the shot category and the shot segment to be deleted are deleted to obtain the shot segment to be processed of the video.

18. The device according to claim 16 or 17, wherein: The second determining submodule is specifically used for: Starting from the to-be-processed shot segment of the first video in each of the videos, determining in sequence whether each to-be-processed shot segment has a highlight plot based on a preset video classification model; If so, the first shot segment to be processed with a highlight plot is determined as the initial shot segment.

19. The device according to claim 16 or 17, wherein: The second determining submodule is specifically used for: Acquire first text recognition information of the to-be-processed shot segments of each of the videos; wherein the first text recognition information represents the dialogue information in the to-be-processed shot segments; From the to-be-processed shot segments of each of the videos located after the initial shot segment, determining whether the to-be-processed shot segment is a shot segment with a suspenseful plot according to the text recognition information of the to-be-processed shot segment; If so, the ending shot segment is determined according to the duration from the initial shot segment to the to-be-processed shot segment with suspenseful plot.

20. The device according to any one of claims 14 to 19, wherein: The identification module comprises: A first extraction submodule is used to extract the first frame image of the first shot segment from the shot segments of the video; The third determination submodule is used to determine that the shot category of the first shot segment is a shot category including a drama cover when it is determined that the first frame image is a drama cover.

21. The device according to any one of claims 14 to 19, wherein: The identification module comprises: A first acquisition submodule is used to acquire speech recognition information of the shot segment of the video; wherein the speech recognition information represents text information of the speech included in the shot segment; The fourth determination submodule is used to determine whether the shot segments of the video have the voice recognition information, starting from the last shot segment of the video, according to the voice recognition information, until a shot segment with voice recognition information is determined, and the shot category of the shot segment of the video that is located after the shot segment with voice recognition information is determined as a shot category that includes the end credits.

22. The device according to any one of claims 14 to 19, wherein: The identification module comprises: The second acquisition submodule is used to acquire, for each video of the target episode, second text recognition information of the shot segments of the video, and third text recognition information of the shot segments of the previous episode of the video; wherein the text recognition information represents the line information in the shot segments; The fifth determination submodule is used to determine, for each shot segment of the video, that the shot category of the shot segment is a shot category that includes overlapping plots if the similarity between the second text identification information of the shot segment and the third text identification information of the shot segment of the previous video of the video is greater than or equal to a preset threshold.

23. The device according to any one of claims 14 to 19, wherein: The identification module comprises: A second extraction submodule is used to extract any frame of picture from the shot segment of the video; The sixth determination submodule is used to determine that the shot category of the shot segment is a shot category including advertisement when determining that the any one frame of picture is an advertisement picture.

24. The device according to any one of claims 13 to 23, wherein: The generating unit comprises: The splicing module is used to splice the initial shot segment, the ending shot segment, and the intermediate shot segment to obtain a promotional video of the target series.

25. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 12.

26. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-12.

27. A computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 12.

Images