CN114302174A - Video editing method and device, computing equipment and storage medium - Google Patents

Abstract

The application discloses a video clipping method, a video clipping device, a computing device and a storage medium, wherein the method comprises the following steps: the method comprises the steps of obtaining an original video to be processed, identifying a plurality of key positions and transition positions in the original video, wherein the key positions are used for indicating video clips in the original video, so that the plurality of video clips can be obtained by segmenting the original video according to the plurality of key positions and the transition positions, and splicing the video clips to obtain a target video based on the plurality of video clips, wherein the playing time length of the generated target video is less than the playing time length of the original video. Since the target video is generated according to the transition position, the video content of each clip segment in the target video at the start position and/or the end position can be generally made complete. Therefore, when the user watches the target video, the user usually considers that the consistency of the appearance is high because the content of the video is complete, so that the experience of watching the clipped target video by the user can be improved.

Description

Video editing method, device, computing device and storage medium

technical field

The present application relates to the technical field of video processing, and in particular, to a video editing method, apparatus, computing device and storage medium.

Background technique

In practical application scenarios, for a video with a long playback duration, the video can usually be edited to generate a clipped video with a relatively short playback duration and including core video content. For example, under the feature film of a talk show variety show on an Internet video website, some video clips of the laughing point collection obtained by editing the feature film are usually released, so that the audience can quickly watch the whole laughing point clip.

Since the edited video is generated by manual editing, not only the labor cost is high, but also the efficiency of manually editing the video is usually low. Therefore, an artificial intelligence (Artificial Intelligence, AI) editing method can be used to automatically generate a clipped video. However, the video clips generated based on AI usually have the problem of incoherent look and feel. For example, there is a character in the clip video that is cut off without finishing a sentence, which affects the user's viewing experience of the clip video.

SUMMARY OF THE INVENTION

Embodiments of the present application provide a video editing method, apparatus, computing device, and storage medium, which aim to improve the look-and-feel consistency of the automatically generated edited video, thereby improving the user's viewing experience of the edited video.

In a first aspect, an embodiment of the present application provides a video editing method, and the method includes:

Get the raw video to be processed;

Identifying multiple key positions and transition positions in the original video, where the key positions are used to indicate video clips in the original video;

According to the plurality of key positions and the transition positions, a plurality of video clips are obtained by segmenting the original video;

Based on the plurality of video clips, a target video is obtained by splicing, and the playback duration of the target video is shorter than the playback duration of the original video.

In a possible implementation manner, the identifying the transition position in the original video includes:

Calculate the similarity between the adjacent first frame images and the second frame images in the original video;

When the similarity between the first frame image and the second frame image is less than a preset threshold, the position of the first frame image or the second frame image in the original video is determined as the the transition location.

In a possible implementation manner, according to the multiple key positions and the transition positions, multiple video clips are obtained by segmenting the original video, including:

According to a plurality of key positions in the original video, determine the start division point and the end division point corresponding to the plurality of candidate video segments in the original video;

Determine whether a transition position is included in the multi-frame first video images whose distance from the original video and the starting segmentation point of the target candidate video segment does not exceed a first preset distance, and whether the original video is different from the Whether the distance between the termination points of the target candidate video segment does not exceed the second preset distance includes transition positions in the multi-frame second video images, and the target candidate video segment is any of the plurality of candidate video segments. a candidate video segment;

When the multi-frame first video images include transition positions, and/or, the multi-frame second video images include transition positions, according to the transition positions in the multi-frame first video images and/or Or at the transition positions in the second video images of the multiple frames, the target candidate video segment is obtained by segmenting the original video.

In a possible implementation manner, the original video is a first type of video, and key positions in the original video are identified by audio features in the original video.

In a possible implementation manner, the audio content corresponding to the key positions in the original video is laughter and/or applause, and the identifying multiple key positions in the original video includes:

Inputting the original video into an artificial intelligence AI model to obtain multiple key positions in the original video output by the AI model, and the AI model is pre-completed with sample videos marked with laughter and/or applause train;

Or, performing voiceprint feature matching between the audio data in the original video and the audio data corresponding to laughter and/or applause to obtain the multiple key positions matched by the voiceprint feature.

In a possible implementation manner, the original video is a second type of video, and key positions in the original video are identified by image features in the original video.

In a possible implementation manner, the identifying multiple key positions in the original video includes:

determining a plurality of initial key positions from the original video;

The multiple initial key positions are adjusted by using the optical character recognition technology to obtain the multiple key positions, so that each of the key positions is the position where the subtitles start to be displayed or the subtitles end to be displayed.

In a second aspect, an embodiment of the present application further provides a video editing device, the device comprising:

The acquisition module is used to acquire the raw video to be processed;

a position identification module for identifying multiple key positions and transition positions in the original video, where the key positions are used to indicate video clips in the original video;

A segmentation module, configured to obtain a plurality of video clips by segmentation from the original video according to the multiple key positions and the transition positions;

A splicing module, configured to splicing a target video based on the plurality of video clips, where the playback duration of the target video is shorter than the playback duration of the original video.

In a possible implementation, the location identification module includes:

a calculation unit, used for calculating the similarity between the adjacent first frame images and the second frame images in the original video;

A first determining unit, configured to, when the similarity between the first frame image and the second frame image is less than a preset threshold, place the first frame image or the second frame image in the original The position in the video is determined as the transition position.

In a possible implementation, the segmentation module includes:

a second determining unit, configured to determine, according to a plurality of key positions in the original video, a start split point and an end split point corresponding to a plurality of candidate video segments in the original video;

a third determining unit, configured to determine whether a transition position is included in the multi-frame first video images whose distances from the original video and the starting segmentation point of the target candidate video segment do not exceed the first preset distance, and Whether a transition position is included in the multi-frame second video images whose distance between the original video and the termination point of the target candidate video segment does not exceed the second preset distance, and the target candidate video segment is the multi-frame second video image. any candidate video segment among the candidate video segments;

The segmentation unit is configured to, when the multi-frame first video images include transition positions, and/or, the multi-frame second video images include transition positions, according to the multi-frame first video images and/or the transition position in the multi-frame second video images, the target candidate video segment is obtained by segmenting the original video.

In a possible implementation manner, the original video is a first type of video, and key positions in the original video are identified by audio features in the original video.

In a possible implementation manner, the audio content corresponding to key positions in the original video is laughter and/or applause, and the position identification module includes:

The first identification unit is used to input the original video into an artificial intelligence AI model, and obtain a plurality of key positions in the original video output by the AI model, and the AI model is pre-marked with laughter and/or Or applause marked sample videos to complete the training;

or,

The second identification unit is configured to perform voiceprint feature matching between the audio data in the original video and the audio data corresponding to laughter and/or applause to obtain the multiple key positions matched by the voiceprint feature.

In a possible implementation manner, the original video is a second type of video, and key positions in the original video are identified by image features in the original video.

In a possible implementation, the location identification module includes:

a fourth determining unit, configured to determine a plurality of initial key positions from the original video;

The adjustment unit is configured to adjust the multiple initial key positions by using the optical character recognition technology to obtain the multiple key positions, so that each of the key positions is the position where the subtitles start to be displayed or the subtitles end to be displayed.

In a third aspect, an embodiment of the present application further provides a computing device, and the computing device may include a processor and a memory:

the memory is used to store computer programs;

The processor is configured to execute the method according to the first aspect and any one of the implementation manners of the first aspect according to the computer program.

In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium is used to store a computer program, and the computer program is used to execute the first aspect and the first The method of any embodiment of the aspect.

In the above implementation manner of the embodiment of the present application, the original video to be processed is acquired, and a plurality of key positions and transition positions in the original video are identified, and the key positions are used to indicate the video clips in the original video, so that the The multiple key positions and transition positions are divided into multiple video clips from the original video, and a target video is obtained by splicing based on the multiple video clips, wherein the playback duration of the generated target video is shorter than the playback duration of the original video. .

Due to the transition position in the original video, it usually indicates that a continuous video content ends at the transition position. Therefore, when the target video is generated according to the transition position, each clip in the target video can usually be made at the starting position. and/or the video content at the end position is relatively complete. In this way, when the user watches the target video, the user usually thinks that the video content is complete and has a high look and feel consistency, so that the user's experience of watching the edited target video can be improved.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments. Obviously, the drawings in the following description are only some implementations described in the present application. For example, for those skilled in the art, other drawings can also be obtained from these drawings.

FIG. 1 is a schematic diagram of an exemplary application scenario in an embodiment of the present application;

2 is a schematic flowchart of a video editing method in an embodiment of the present application;

3 is a schematic structural diagram of a video editing device in an embodiment of the application;

FIG. 4 is a schematic diagram of a hardware structure of a computing device according to an embodiment of the present application.

Detailed ways

Referring to FIG. 1 , it is a schematic diagram of an application scenario provided by an embodiment of the present application. In the application scenario shown in FIG. 1 , the client 101 may have a communication connection with the computing device 102 . In addition, the client 101 can receive the video provided by the user (such as a video editor), and send it to the computing device 102; the computing device 102 is used to perform AI editing on one or more received videos, generate a clipped video, and The clipped video is presented to the user through the client 101 .

The computing device 102 refers to a device with data processing capability, such as a terminal, a server, and the like. Client 101 may be implemented in a physical device separate from computing device 102 . For example, when the computing device 102 is implemented by a server, the client 101 may run on a device such as a user terminal on the user side. Alternatively, client 101 may also run on computing device 102 .

In practical application, the video clip generated by the computing device 102 based on the preset AI algorithm usually has the problem of incoherent look and feel, which will affect the user's look and feel of the clip video. For example, suppose the original video includes a dialogue in which Person A asks, "Have you not been to the gym recently?" Person B replies: "No," and the computing device 102 may speak to Person A based on an AI algorithm. The truncation starts at the end position, then the edited video clip only includes the answer content (“no”) of character B, which makes the user feel confused when watching the content of character B’s speech “no” in the edited video Cloud, that is, I don't know what character B is answering "no" to. For another example, character A speaks multiple sentences continuously, but the computing device 102 only intercepts video clips of some sentences spoken by the character A based on the AI algorithm, resulting in obvious lack of speech content of the character A.

Based on this, an embodiment of the present application provides a video editing method, which aims to improve the look-and-feel consistency of the generated edited video, so as to improve the user's viewing experience for the edited video. In specific implementation, the computing device 102 acquires the original video to be processed, and identifies multiple key positions and transition positions in the original video, where the key positions are used to indicate video segments in the original video, so that the computing device 102 can A plurality of key positions and transition positions are obtained by dividing a plurality of video clips from the original video, and a target video is obtained by splicing based on the plurality of video clips, wherein the playback duration of the generated target video is shorter than the playback duration of the original video.

Due to the transition position in the original video, it usually indicates that a continuous video content ends at the transition position. Therefore, when the target video is generated according to the transition position (that is, the aforementioned clip video is generated), it is usually possible to make the target video The video content at the start position and/or the end position of each clip in the video is relatively complete. In this way, when the user watches the target video, the user usually thinks that the video content is complete and has a high look and feel consistency, so that the user's experience of watching the edited target video can be improved.

It should be noted that the video in this embodiment refers to a video having both image and audio content, that is, a video file includes not only video images of multiple consecutive frames, but also audio data synchronized with the video images.

It can be understood that the architecture of the application scenario shown in FIG. 1 is only an example provided by the embodiment of the present application. In practical application, the embodiment of the present application can also be applied to other applicable scenarios. For example, the computing device 102 can automatically Obtain one or more videos from the Internet, and automatically generate clip videos corresponding to each video through the above implementation manner. In conclusion, the embodiments of the present application can be applied to any applicable scenario, and are not limited to the above scenario examples.

In order to make the above objects, features and advantages of the present application more clearly understood, various non-limiting implementations in the embodiments of the present application will be exemplarily described below with reference to the accompanying drawings. Obviously, the described embodiments are some, but not all, embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

Referring to FIG. 2, FIG. 2 shows a schematic flowchart of a video editing method in an embodiment of the present application. The method may be applied to the application scenario shown in FIG. 1, or may be applied to other applicable application scenarios. For ease of description and understanding, the following description is given by applying the application scenario shown in FIG. 1 as an example. Specifically, the method may include:

S201: Obtain the original video to be processed.

For the convenience of distinction and description, in this embodiment, the video to be edited is referred to as the original video, and the video generated by the clip is referred to as the target video.

In one possible implementation, the original video may be provided to computing device 102 by the user. Specifically, the client 101 can present a video import interface to the user, so that the user can import the original video into the client 101 by performing corresponding operations on the video import interface. Then, the client 101 can transmit the original video provided by the user to the computing device 102 through the network connection with the computing device 102 .

In another possible implementation, the original video may also be obtained by the computing device 102 from the Internet. For example, the user can send an instruction to generate a clip video to the computing device 102 through the client 101, so that the computing device 102 can download a specific type of video from the Internet based on the instruction, such as downloading a talk show type video or a cross talk type video, Download videos of viewing variety shows, etc., and use these videos as original videos for subsequent editing of these original videos.

It is worth noting that the original video acquired by the computing device 102 may be one video or multiple videos. For example, the computing device 102 may generate a target video based on multiple original videos and clipping, etc. This embodiment does not Not limited. For ease of understanding and description, in this embodiment, the original video is taken as an example for description. When the original video includes multiple videos, the implementation is similar to this embodiment, and the difference lies in the subsequent splicing of multiple video segments. Sourced from multiple different raw videos.

S202: Identify multiple key positions and transition positions in the original video, where the key positions are used to indicate video segments in the original video.

Wherein, the key position refers to the position used to indicate the video clip in the original video, and the start and end split points of the clipped video clip can be determined based on the key position when the original video is edited. And, for different types of original videos, different categories of locations can be used as key locations.

In one example, computing device 102 may identify key locations through audio features in the raw video. For example, when the original video is the first type of video, such as talk show, cross talk and other funny type videos, due to the actual application scenario, the video content with laughter and/or applause is usually more attractive to users. Therefore, The computing device 102 may determine a position in the original video where the audio content is laughter and/or applause as a key position.

In one embodiment of identifying key locations, computing device 102 may utilize an AI model to identify multiple key locations in the raw video. When specifically implemented, the computing device 102 may be pre-configured with an AI model, and the AI model is pre-trained through video samples marked with "laughter" and/or "applause", so that the trained AI model can identify "Laughter" and/or "Applause". Therefore, for the original video of the first type, the computing device 102 may input the original video into the AI model that has completed the training, and obtain multiple key positions in the original video output by the AI model.

In another embodiment of identifying key positions, the computing device 102 may determine multiple key positions in the original video by comparing voiceprint features. Specifically, the computing device 102 may acquire audio data with the content of "laughter" and/or "applause", and extract the voiceprint features of the "laughter" and/or "applause", and then the computing device 102 may Compare the voiceprint feature with the voiceprint feature corresponding to the audio data in the original video segment by segment, and determine the location of the audio data with the same voiceprint feature as the key location, so as to determine multiple key locations in the original video. .

It should be noted that the above-mentioned implementation manner of determining the key position is provided as some exemplary descriptions. In practical application, the computing device 102 may also determine the key position in the original video by other means, which is not limited in this embodiment. .

In yet another example, computing device 102 may identify key locations through image features in the raw video. For example, when the original video is a video of the second type, such as a video of viewing variety shows, the position where the subtitles of one of the subtitles in the original video start to be displayed and/or the subtitles are displayed may end may be determined as the key position. Certainly, in other embodiments, the key position may also be other possible implementation manners, which are not limited in this embodiment.

In a possible implementation, the computing device 102 may determine a plurality of initial key positions from the original video, and adjust the plurality of initial key positions through an optical character recognition (Optical Character Recognition, OCR) technology to obtain a plurality of initial key positions. Key positions, so that each key position is the position where the subtitles start to be displayed or the subtitles end to be displayed. For example, the computing device 102 may randomly select two positions as the start position and the end position of a video according to the playback duration (eg, 30 seconds, etc.), and obtain two initial key positions corresponding to the video. Then, the computing device 102 can identify the subtitle in the video image at the starting position by using the OCR technology, and use the starting display position of the subtitle in the original video as a key position. In addition, the computing device 102 can also use the OCR technology to identify the subtitle in the video image at the end position, and use the end display position of the subtitle in the original video as a key position and the like. In this way, the computing device 102 can determine the key positions corresponding to the multiple video clips in the above manner.

The transition position refers to the position where the character type in the original video is switched, such as the position where the performer switches to the audience or the guest seat, or the position where the performer switches to a scene that does not include characters, which may be due to the shooting scene. The switching of the camera leads to the switching of the type of person to be photographed, etc. In practical applications, when the character type in the original video is switched, it usually means that a piece of video content related to the character ends temporarily. For example, when an argument/argument made by the character ends, the shooting lens can usually be switched to the auditorium. In order to capture the audience's reaction to the speech content of the character (such as laughing, nodding, shaking his head, etc.), the computing device 102 can use the transition position to determine the edited video when editing the original video. The boundary point of the fragment. Of course, in other possible embodiments, the transition position may also be a position where other information in the original video is switched, such as scene switching, which is not limited in this embodiment.

In a possible implementation, the computing device 102 may determine the transition position by comparing the difference between two adjacent frames of images. During specific implementation, the computing device 102 may calculate the similarity between the adjacent first frame image and the second frame image in the original video, and when the display degree between the first frame image and the second frame image is smaller than the preset When the threshold is set, the computing device 102 may determine the position of the first frame image or the second frame image in the original video as the transition position. For example, the first frame of image is an image obtained by photographing a performer, and the second frame image is an image obtained by photographing an audience. The computing device 102 can determine all transition positions in the original video by traversing the calculation method. For example, the computing device 102 can determine the transition by sequentially comparing the image similarity between two consecutive adjacent video images. point. Alternatively, the computing device 102 may perform similarity calculation on multiple frames of images in the vicinity of the key position to determine whether the transition position is included in the vicinity of the key position.

Exemplarily, when calculating the similarity between two frames of images, the computing device 102 may first reduce the two frames of images to a size of 8 pixels*8 pixels, that is, each frame of images after the reduction has 64 pixels. The function of this step is to remove the details of the image, and only retain the basic information such as structure/light and shade in the image to reduce the amount of subsequent calculations. Then, the computing device 102 may perform grayscale processing on the reduced two frames of images, and respectively calculate the average grayscale value of each frame image (ie, the average value of 64 grayscale values in each frame image). Next, the computing device 102 compares the gray value of each pixel in each frame of image with the average gray value corresponding to the frame image, and, when the gray value of the pixel is greater than or equal to the average gray value, marks the pixel is 1, and when the gray value of the pixel is less than the average gray value, the pixel is marked as 0, so that the 64 pixels in each frame of image are combined according to uniform rules, and a 64-bit hash value can be generated (by 1 and 0), the hash value can be used as the fingerprint of the frame image. In this way, the computing device 102 can compare the 64-bit hash values corresponding to the two frames of images, and when the number of bits that differ between the two hash values exceeds a preset value (eg, 5, etc.), the computing device 102 determines that the two frames The similarity of the images is relatively small, and when the number of bits that differ between the two hash values does not exceed the preset value, the computing device 102 determines that the similarity of the two frames of images is relatively high. In practical application, the computing device 102 may also determine the similarity between two frames of images in other ways, which is not limited in this embodiment.

S203: According to the identified multiple key positions and transition positions, segment the original video to obtain multiple video segments.

In this embodiment, the computing device 102 may segment the original video according to multiple key positions and transition positions to obtain multiple video segments.

In a possible implementation manner, the computing device 102 may first determine, according to multiple key positions in the original video, start and end division points corresponding to multiple candidate video segments in the original video. The starting segmentation point refers to the starting point of the candidate video segment, and the video image at the starting segmentation point is the first frame image of the candidate video segment. Correspondingly, the termination division point refers to the termination point of the candidate video segment, and the video image at the termination division point is the last frame image of the candidate video segment. For example, when the original video is a video of the first type, the computing device 102 may determine the playback position at the first 15 (or other numerical values) seconds of the key position as the starting split point of the candidate video segment, and the post-key position The playback position at 1 (or other numerical value) seconds is determined as the termination split point of the candidate video segment.

Then, the computing device 102 may determine whether a transition position is included in the multi-frame first video images whose distance from the starting segmentation point of the target candidate video segment does not exceed the first preset distance in the original video, and whether the transition position is included in the original video. Whether a transition position is included in the multi-frame second video images of which the distance between the termination points of the target candidate video segment does not exceed the second preset distance, the target candidate video segment is any of the above-mentioned candidate video segments a candidate video segment. Wherein, the distance from the starting split point can be characterized as the number of video frames included between the position of the video image in the original video and the starting and ending split points (or the playback duration corresponding to the number of these video frames, etc.), and , the larger the number of video frames, the farther the representation distance, the smaller the number of video frames, the closer the representation distance. Correspondingly, the first preset distance may be, for example, a preset number of video frames (or a playback duration corresponding to the number of video frames, etc.), such as 450 consecutive frames of images. Similarly, the distance from the end split point can be represented by the number of video frames included between the position of the video image in the original video and the start and end split point, or the playback duration corresponding to the number of video frames. Correspondingly, the second preset distance may be a preset number of video frames, or a playback duration corresponding to the number of video frames. The size of the first preset distance and the size of the second preset distance may be the same or different.

When the multiple frames of first video images include transition positions, and/or the multiple frames of second video images include transition positions, the computing device 102 Frame the transition position in the second video image, and segment the original video to obtain the target candidate video segment. Specifically, when the multi-frame first video image includes a transition position, the computing device 102 may update the starting segmentation point of the target candidate video segment to the transition position in the multi-frame first video image, and according to the transition position The field position and the end segmentation point of the target candidate video segment are segmented from the original video to obtain the target candidate video segment. When a transition position is included in the multiple frames of the second video image, the computing device 102 may update the end split point of the target candidate video segment to the transition position in the multiple frames of the second video image, and according to the transition position and The starting segmentation point of the target candidate video segment, which is segmented from the original video to obtain the target candidate video segment. When both the multi-frame first video images and the multi-frame second video images include transition positions, the computing device 102 may update the starting segmentation point of the target candidate video segment to the transition in the multi-frame first video images position, update the end split point of the target candidate video segment to the transition position in the multi-frame second video image, and segment the original video to obtain the target candidate video segment according to the two transition positions. In this way, the computing device 102 can obtain multiple video clips by segmenting according to multiple key positions and transition positions.

Since the computing device 102 can adjust the start split point and/or the end split point of the target candidate video clip according to the transition position, the video clip edited based on the transition position can usually include a relatively complete video content. In this way, when the user watches the video clip, the problem of low consistency of look and feel can be avoided. For example, suppose the original video includes a conversation about Person A and Person B, Person A asks: "Have you been to the gym recently?" Person B replies: "No", and, after Person B finishes speaking When a scene transition occurs, the computing device 102 can determine the end split point of the video clip as after the character B has said "no" according to the identified transition position. Including the dialogue content of character A and character B, therefore, the visual continuity of this video clip is relatively high.

S204: Based on a plurality of video clips, a target video is obtained by splicing, and the playback duration of the target video is shorter than the playback duration of the original video.

After editing and including multiple video segments from the original video, the computing device 102 may splicing the multiple video segments to generate a target video. The computing device 102 may perform sequential splicing according to the playing order of each video clip in the original video, or may perform splicing in other sequences, etc., which is not limited in this embodiment.

In practical applications, the playback duration of the target video generated by clipping is limited. For example, for an original video with a playback duration of 2 hours, the playback duration of a target video generated by editing it may not exceed 10 minutes. Therefore, the playback duration of the target video generated by the video clip is usually shorter than the playback duration of the original video.

Further, if the total playing duration corresponding to the multiple video clips obtained by segmentation is greater than the maximum playing duration of the target video to be generated, the computing device 102 selects some video clips from the multiple video clips to generate the target video. For example, the computing device 102 may select some video segments with a relatively long playback duration among the plurality of video segments to generate the target video, or may randomly select some video segments to generate the target video, etc., which is not limited in this embodiment.

In this embodiment, due to the transition position in the original video, it usually indicates that a piece of continuous video content ends at the transition position. Therefore, when the target video is generated according to the transition position, each clip in the target video can usually be The video content of the segment is relatively complete at the start position and/or end position. In this way, when the user watches the target video, the user usually thinks that the video content is complete and has a high look and feel consistency, so that the user's experience of watching the edited target video can be improved.

For ease of understanding and explanation, the following is an example of the original video being a talk show variety show video and an observation variety show video.

After acquiring the talk show variety show video, the computing device 102 can identify the location where laughter and/or applause exists in the talk show video through the AI model, or determine that the talk show video exists in the talk show video by comparing voiceprint features. Key locations for laughter and/or applause, etc. Then, for each key position, the computing device 102 may further determine the playback position at the first 15 seconds of the key position (ie the start split point) and the playback position at the last 3 seconds (the end split point), and identify the Whether the transition position is included in the multi-frame first video images between the key position and the playback position in the first 15 seconds, and whether the multi-frame second video image between the key position and the playback position in the next 3 seconds is identified Including transition positions. If a transition position is included in the multiple frames of the first video image, the computing device 102 may adjust the starting split point of the video clip to the transition position, and/or if the multiple frames of the second video image include a transition position , the end split point of the transition position of the video clip is adjusted to the transition position, so that the video clip is cut out from the original video based on the adjusted start split point and/or end split point. As such, for multiple key locations, the computing device 102 may segment multiple video segments including laughter and/or applause. Finally, the computing device 102 may splicing the segmented video segments to generate a talk show video collection, that is, a clip video desired by the user.

After acquiring the observation type variety show video, the computing device 102 can intercept a candidate video segment with a preset duration (such as a 30-second playback duration) from the observation type variety show video, and the start and end segmentation points of the candidate video segment can be. Preliminarily determined by OCR technology, specifically, through OCR technology, the position where a certain subtitle starts to be displayed is used as the starting segmentation point of the candidate video segment, and the position where another subtitle is displayed at the end is used as the termination segmentation point of the candidate video segment, And the playback duration corresponding to the multi-frame video images between the start split point and the end split point is close to the preset duration. Then, the computing device 102 can identify whether a transition position is included in the multiple frames of first video images whose distances from the start dividing point in the observed variety show video do not exceed a first preset distance (eg, 90 frames of video images, etc.). , and identify whether a transition position is included in the multi-frame first video images in which the distance between the viewing variety video and the termination point does not exceed the second preset distance (such as 60 frames of video images, etc.) If the frame of the first video image includes a transition position, the computing device 102 may adjust the starting split point of the video clip to the transition position, and/or, if multiple frames of the second video image include the transition position, then The end split point of the transition position of the video clip is adjusted to the transition position, so that the video clip is split from the original video based on the adjusted start split point and/or end split point. In this way, multiple video clips can be cut out from the observation type variety show video. Finally, the computing device 102 may splicing the divided video segments to generate a clipped video of the viewing variety video.

In addition, the embodiments of the present application also provide a video editing apparatus. Referring to FIG. 3, FIG. 3 shows a schematic structural diagram of a video editing apparatus in an embodiment of the present application. The video editing apparatus 300 includes:

an acquisition module 301, configured to acquire the original video to be processed;

a position identification module 302, configured to identify multiple key positions and transition positions in the original video, where the key positions are used to indicate video clips in the original video;

A segmentation module 303, configured to segment the original video to obtain multiple video segments according to the multiple key positions and the transition positions;

The splicing module 304 is used for splicing and obtaining a target video based on the plurality of video clips, where the playback duration of the target video is shorter than the playback duration of the original video.

In a possible implementation manner, the location identification module 302 includes:

a calculation unit, used for calculating the similarity between the adjacent first frame images and the second frame images in the original video;

A first determining unit, configured to, when the similarity between the first frame image and the second frame image is less than a preset threshold, place the first frame image or the second frame image in the original The position in the video is determined as the transition position.

In a possible implementation manner, the segmentation module 303 includes:

a second determining unit, configured to determine, according to a plurality of key positions in the original video, a start split point and an end split point corresponding to a plurality of candidate video segments in the original video;

a third determining unit, configured to determine whether a transition position is included in the multi-frame first video images whose distances from the original video and the starting segmentation point of the target candidate video segment do not exceed the first preset distance, and Whether a transition position is included in the multi-frame second video images whose distance between the original video and the termination point of the target candidate video segment does not exceed the second preset distance, and the target candidate video segment is the multi-frame second video image. any candidate video segment among the candidate video segments;

The segmentation unit is configured to, when the multi-frame first video images include transition positions, and/or, the multi-frame second video images include transition positions, according to the multi-frame first video images and/or the transition position in the multi-frame second video images, and obtain the target candidate video segment by segmenting the original video.

In a possible implementation manner, the original video is a first type of video, and key positions in the original video are identified by audio features in the original video.

In a possible implementation manner, the audio content corresponding to the key positions in the original video is laughter and/or applause, and the position identification module 302 includes:

The first identification unit is used to input the original video into an artificial intelligence AI model, and obtain a plurality of key positions in the original video output by the AI model, and the AI model is pre-marked with laughter and/or Or applause marked sample videos to complete the training;

or,

The second identification unit is configured to perform voiceprint feature matching between the audio data in the original video and the audio data corresponding to laughter and/or applause to obtain the multiple key positions matched by the voiceprint feature.

In a possible implementation manner, the original video is a second type of video, and key positions in the original video are identified by image features in the original video.

In a possible implementation manner, the location identification module 302 includes:

a fourth determining unit, configured to determine a plurality of initial key positions from the original video;

The adjustment unit is configured to adjust the multiple initial key positions by using the optical character recognition technology to obtain the multiple key positions, so that each of the key positions is the position where the subtitles start to be displayed or the subtitles end to be displayed.

It should be noted that the information exchange, execution process and other contents between the modules and units of the above-mentioned apparatus are based on the same concept as the method embodiments in the embodiments of the present application, and the technical effects brought by them are the same as those of the method embodiments in the embodiments of the present application. The examples are the same, and the specific content can refer to the descriptions in the method embodiments shown in the foregoing embodiments of the present application, and details are not repeated here.

In addition, the embodiments of the present application also provide a computing device. Referring to FIG. 4 , FIG. 4 shows a schematic diagram of a hardware structure of a computing device in an embodiment of the present application. The computing device 400 may include a processor 401 and a memory 402 .

Wherein, the memory 402 is used to store computer programs;

The processor 401 is configured to perform the following steps according to the computer program:

Get the raw video to be processed;

Identifying multiple key positions and transition positions in the original video, where the key positions are used to indicate video clips in the original video;

According to the plurality of key positions and the transition positions, a plurality of video clips are obtained by segmenting the original video;

Based on the plurality of video clips, a target video is obtained by splicing, and the playback duration of the target video is shorter than the playback duration of the original video.

In a possible implementation manner, the processor 401 is specifically configured to perform the following steps according to the computer program:

Calculate the similarity between the adjacent first frame images and the second frame images in the original video;

When the similarity between the first frame image and the second frame image is less than a preset threshold, the position of the first frame image or the second frame image in the original video is determined as the the transition location.

In a possible implementation manner, the processor 401 is specifically configured to perform the following steps according to the computer program:

According to a plurality of key positions in the original video, determine the start division point and the end division point corresponding to the plurality of candidate video segments in the original video;

Determine whether a transition position is included in the multi-frame first video images whose distance from the original video and the starting segmentation point of the target candidate video segment does not exceed a first preset distance, and whether the original video is different from the Whether the distance between the termination points of the target candidate video segment does not exceed the second preset distance includes transition positions in the multi-frame second video images, and the target candidate video segment is any of the plurality of candidate video segments. a candidate video segment;

When the multi-frame first video images include transition positions, and/or, the multi-frame second video images include transition positions, according to the transition positions in the multi-frame first video images and/or Or at the transition positions in the second video images of the multiple frames, the target candidate video segment is obtained by segmenting the original video.

In a possible implementation manner, the original video is a first type of video, and key positions in the original video are identified by audio features in the original video.

In a possible implementation manner, the audio content corresponding to the key position in the original video is laughter and/or applause, and the processor 401 is specifically configured to perform the following steps according to the computer program:

Inputting the original video into an artificial intelligence AI model to obtain multiple key positions in the original video output by the AI model, and the AI model is completed in advance through sample videos marked with laughter and/or applause train;

Or, performing voiceprint feature matching between the audio data in the original video and the audio data corresponding to laughter and/or applause to obtain the multiple key positions matched by the voiceprint feature.

In a possible implementation manner, the original video is a second type of video, and key positions in the original video are identified by image features in the original video.

In a possible implementation manner, the processor 401 is specifically configured to perform the following steps according to the computer program:

determining a plurality of initial key positions from the original video;

The multiple initial key positions are adjusted by using the optical character recognition technology to obtain the multiple key positions, so that each of the key positions is the position where the subtitles start to be displayed or the subtitles end to be displayed.

In addition, an embodiment of the present application further provides a computer-readable storage medium, where the computer-readable storage medium is used to store a computer program, and the computer program is used to execute the method described in the foregoing method embodiment.

From the description of the above embodiments, those skilled in the art can clearly understand that all or part of the steps in the methods of the above embodiments can be implemented by means of software plus a general hardware platform. Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, and the computer software product can be stored in a storage medium, such as read-only memory (English: read-only memory, ROM)/RAM, magnetic disk, An optical disc, etc., includes several instructions for causing a computer device (which may be a personal computer, a server, or a network communication device such as a router) to execute the methods described in various embodiments or some parts of the embodiments of the present application.

The various embodiments in this specification are described in a progressive manner, and the same and similar parts between the various embodiments may be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the apparatus embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and reference may be made to some descriptions of the method embodiments for related parts. The device embodiments described above are only illustrative, wherein the modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place , or distributed to multiple network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.

The above descriptions are only exemplary embodiments of the present application, and are not intended to limit the protection scope of the present application.

Claims (10)

1. a video editing method, is characterized in that, described method comprises: Get the raw video to be processed; Identifying multiple key positions and transition positions in the original video, where the key positions are used to indicate video clips in the original video; According to the plurality of key positions and the transition positions, a plurality of video clips are obtained by segmenting the original video; Based on the plurality of video clips, a target video is obtained by splicing, and the playback duration of the target video is shorter than the playback duration of the original video. 2. The method according to claim 1, wherein the identifying the transition position in the original video comprises: calculating the similarity between the adjacent first frame images and the second frame images in the original video; When the similarity between the first frame image and the second frame image is less than a preset threshold, the position of the first frame image or the second frame image in the original video is determined as the the transition location. 3. The method according to claim 1, characterized in that, according to the multiple key positions and the transition positions, obtaining multiple video clips from the original video by dividing, comprising: According to a plurality of key positions in the original video, determine the start division point and the end division point corresponding to the plurality of candidate video segments in the original video; Determine whether a transition position is included in the multi-frame first video images whose distance from the original video and the starting segmentation point of the target candidate video segment does not exceed a first preset distance, and whether the original video is different from the Whether the distance between the termination points of the target candidate video segment does not exceed the second preset distance includes transition positions in the multi-frame second video images, and the target candidate video segment is any of the plurality of candidate video segments. a candidate video segment; When the multi-frame first video images include transition positions, and/or, the multi-frame second video images include transition positions, according to the transition positions in the multi-frame first video images and/or Or at the transition positions in the second video images of the multiple frames, the target candidate video segment is obtained by segmenting the original video. 4. The method according to claim 1, wherein the original video is a first type of video, and key positions in the original video are identified by audio features in the original video. 5. The method according to claim 4, wherein the audio content corresponding to the key positions in the original video is laughter and/or applause, and the identifying a plurality of key positions in the original video includes : Inputting the original video into an artificial intelligence AI model to obtain multiple key positions in the original video output by the AI model, and the AI model is pre-completed with sample videos marked with laughter and/or applause train; Or, performing voiceprint feature matching between the audio data in the original video and the audio data corresponding to laughter and/or applause to obtain the multiple key positions matched by the voiceprint feature. 6 . The method according to claim 1 , wherein the original video is of the second type, and key positions in the original video are identified by image features in the original video. 7 . 7. The method according to claim 6, wherein the identifying a plurality of key positions in the original video comprises: determining a plurality of initial key positions from the original video; The multiple initial key positions are adjusted by using the optical character recognition technology to obtain the multiple key positions, so that each of the key positions is the position where the subtitles start to be displayed or the subtitles end to be displayed. 8. A video editing device, wherein the device comprises: The acquisition module is used to acquire the raw video to be processed; a position identification module for identifying multiple key positions and transition positions in the original video, where the key positions are used to indicate video clips in the original video; A segmentation module, configured to obtain a plurality of video clips by segmentation from the original video according to the multiple key positions and the transition positions; A splicing module, configured to splicing a target video based on the plurality of video clips, where the playback duration of the target video is shorter than the playback duration of the original video. 9. A computing device, wherein the device comprises a processor and a memory: the memory is used to store computer programs; The processor is adapted to perform the method of any one of claims 1-7 according to the computer program. 10. A computer-readable storage medium, wherein the computer-readable storage medium is used to store a computer program, and the computer program is used to execute the method of any one of claims 1-7.

Images