CN111314736B - Video copyright analysis method and device, electronic equipment and storage medium - Google Patents
Video copyright analysis method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN111314736B CN111314736B CN202010198381.4A CN202010198381A CN111314736B CN 111314736 B CN111314736 B CN 111314736B CN 202010198381 A CN202010198381 A CN 202010198381A CN 111314736 B CN111314736 B CN 111314736B
- Authority
- CN
- China
- Prior art keywords
- video
- analyzed
- type
- target
- continuous time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/234—Processing of video elementary streams, e.g. splicing of video streams, manipulating MPEG-4 scene graphs
- H04N21/23418—Processing of video elementary streams, e.g. splicing of video streams, manipulating MPEG-4 scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/234—Processing of video elementary streams, e.g. splicing of video streams, manipulating MPEG-4 scene graphs
- H04N21/23424—Processing of video elementary streams, e.g. splicing of video streams, manipulating MPEG-4 scene graphs involving splicing one content stream with another content stream, e.g. for inserting or substituting an advertisement
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs
- H04N21/44016—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs involving splicing one content stream with another content stream, e.g. for substituting a video clip
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
- H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Business, Economics & Management (AREA)
- Marketing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a video copyright analysis method, a video copyright analysis device, electronic equipment and a storage medium, and relates to the technical field of computers.
Description
Technical Field
The invention belongs to the technical field of computers, and particularly relates to a video copyright analysis method and device, electronic equipment and a storage medium.
Background
With the development of the internet technology, a user can obtain colorful videos from a multimedia resource platform through a client in a terminal device of the user, and can also edit and make the videos in a self-contained manner and upload the videos to the multimedia resource platform for sharing by other users.
However, the multimedia resource platform usually avoids the copyright risk by manually auditing videos issued by users, but since auditing personnel are required to inquire related copyright contents in the auditing process related to copyright videos, a large amount of manpower is consumed, and thus the efficiency of video auditing is reduced.
Disclosure of Invention
In view of this, the present invention provides a method for analyzing copyright of video, including:
acquiring a video to be analyzed;
inputting the video to be analyzed into a pre-trained video identification model, comparing the video with a copyright video in a preset copyright library, and determining a target copyright video and a picture time period which is matched with the content of the target copyright video in the video to be analyzed;
and determining whether the video to be analyzed is the illegal video of the copyright according to the target copyright video and the picture time period.
According to a second aspect of the present invention, there is provided a copyright analysis apparatus for video, comprising:
the acquisition module is used for acquiring a video to be analyzed;
the determining module is used for inputting the video to be analyzed to a pre-trained video recognition model so as to compare the video with the copyright video in a preset copyright library to determine a target copyright video and a picture time period which is matched with the content of the target copyright video in the video to be analyzed;
and the analysis module is used for determining whether the video to be analyzed is the illegal video of the copyright according to the target copyright video and the picture time period.
According to a third aspect of the present invention, there is provided an electronic device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for analyzing the copyright of a video according to any one of the first aspect when executing the computer program.
According to a fourth aspect of the present invention, there is provided a computer-readable storage medium on which a computer program is stored, the computer program, when executed by a processor, implementing the method for copyright analysis of video according to any of the first aspects.
Aiming at the prior art, the invention has the following advantages:
according to the video copyright analysis method, the video copyright analysis device, the electronic equipment and the storage medium, the target copyright video matched with the existing content of the video to be analyzed is obtained from the preset copyright library through the pre-trained video recognition model, and then whether the video to be analyzed is the illegal copyright video or not is determined according to the picture time period of the video to be analyzed matched with the content of the target copyright video, so that the manual participation in the video copyright analysis process is reduced, and the video copyright analysis efficiency is improved.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1 is a flowchart illustrating steps of a copyright analyzing method for a video according to a first embodiment of the present invention;
fig. 2 is a flowchart illustrating steps of a copyright analyzing method for a second video according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating the steps of a method for determining clip types according to an embodiment of the present invention;
FIG. 4 is a flowchart illustrating steps of a method for processing a frame time slot according to an embodiment of the present invention;
fig. 5 is a block diagram of a copyright analyzing apparatus for video according to an embodiment of the present invention;
fig. 6 is a block diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
Fig. 1 is a flowchart illustrating steps of a copyright analyzing method for a video according to an embodiment of the present invention, where the method includes:
In the embodiment of the invention, the video to be analyzed is any video which needs to be determined whether the copyright violation exists, for example, the video to be analyzed can be video data which is uploaded by a user and needs to be issued in a multimedia resource platform. Because the video data may contain the content of the copyright video, if the content is not identified, the content is directly released on the multimedia resource platform, which may infringe the interests of the copyright content owners, so that the multimedia resource platform bears the risk of infringing the copyright of others. Therefore, when video data to be published uploaded by a user is received, copyright analysis needs to be performed on the video data serving as a video to be analyzed, and whether copyright violation exists or not is determined.
In the embodiment of the invention, the copyright video can be the copyright content acquired from the internet through a network capture technology, can also be the copyright content locally acquired from a multimedia resource platform through a synchronous message, and can also be the copyright content of a small-size work uploaded manually, and the copyright video is collected in a preset copyright library in advance for storage so as to be used for copyright analysis.
Comparing the video content of the video to be analyzed with the video content of the copyright video in the preset copyright library frame by frame, and taking the copyright video with the same content as the video to be analyzed as the target copyright video. The video to be analyzed may be matched with a partial video content of the target copyrighted video, and one or more target copyrighted videos may exist. It can be understood that the video to be analyzed may be obtained by cutting and splicing a part of video segments in one or more copyright videos, but there is a risk of copyright violation if the content in the copyright video is included, so that a target copyright video corresponding to the copyright content included in the video to be analyzed needs to be screened from a preset copyright library for subsequent judgment on whether the copyright is violated.
Specifically, the video to be analyzed and the copyright video in the preset copyright library can be identified through the pre-trained video identification model, so that the human participation in copyright content comparison is reduced. The pre-trained video identification model is obtained by pre-training through a pre-collected sample copyright video and a sample analysis video marked with copyright content by adopting a neural network algorithm, and the pre-trained video identification model can be trained by utilizing the updated copyright video according to a preset period along with updating of the copyright video in a preset copyright library so as to improve the accuracy of the pre-trained video identification model.
In the embodiment of the invention, the picture time period is a time period which is matched with the content of the target copyright video in the returned video to be analyzed after the video to be analyzed and the target copyright video are identified and compared by the pre-trained video identification model. It can be understood that the content of the target copyright video included in the video to be analyzed only has a risk of copyright violation, and further, the picture time period included in the target copyright video needs to be analyzed in one step according to a preset copyright violation rule, so as to determine whether the video to be analyzed is the copyright violation video, that is, the duration and the occupation ratio of the picture time period can be used as a basis for determining whether the video to be analyzed is the copyright violation, for example: if the duration of the picture time period exceeds 3min, the video can be judged as the copyright violation video, or if the ratio of the duration of the picture time period to the total duration of the target copyright video exceeds 5%, or the ratio of the duration of the picture time period to the video to be analyzed exceeds 80%, or if the difference between the duration of the picture time period and the duration of the target copyright video is less than 30min, the video can be judged as the copyright violation video; the position of the picture time period in the target copyrighted video can also be determined according to the location of the picture time period, for example: if the picture time period is at the starting position in the target copyrighted video, it may be only the opening picture referring to the target copyrighted video and is the substantive content related to the target copyrighted video, and thus it may be determined that the video is not a copyright violation video. The above is only an example description of judging whether the video is a copyright violation video according to the picture time period, and may be determined specifically according to actual requirements, and is not specifically limited here.
According to the first video copyright analysis method, the pre-trained video identification model is used for obtaining the target copyright video matched with the existing content of the video to be analyzed from the preset copyright library, and then whether the video to be analyzed is the illegal copyright video or not is determined according to the picture time period of the video to be analyzed matched with the content of the target copyright video, so that manual participation in the video copyright analysis process is reduced, and the video copyright analysis efficiency is improved.
Fig. 2 is a flowchart of steps of a copyright analyzing method for a second video according to an embodiment of the present invention, where the method includes:
This step can refer to the detailed description of step 101, which is not repeated herein.
In the embodiment of the invention, a preset trained video identification model can extract each first video frame in the video to be analyzed and each second video frame of the copyright video in a preset copyright library to perform frame-by-frame matching so as to obtain a candidate video frame with matched content in the video to be analyzed and a corresponding matching degree, wherein the matching degree of the candidate video frame refers to the feature similarity of the first video frame corresponding to the candidate video frame, the matching degree of the candidate copyright video is obtained according to the matching degree corresponding to the candidate video frame, and the matching degree of the candidate copyright video is obtained by summarizing the matching degrees of the candidate video frames.
In the embodiment of the present invention, since the video picture of the video to be analyzed may be a partial video picture in the candidate copyrighted video, for example: the to-be-analyzed video obtained by splicing the figure, the scene picture and the like in the copyright video with other pictures is only quoted for the scene or the figure in the copyright video and does not necessarily relate to the actual content of the copyright video, so that the matching degree of the candidate copyright video and the to-be-analyzed video needs to be further judged to determine whether the to-be-analyzed video contains the actual content of the candidate copyright video. Specifically, a matching degree threshold value can be preset according to actual requirements, so that candidate copyright videos with the corresponding matching degree smaller than the matching degree threshold value are removed, and the target copyright video is obtained. And in the first video frame of the video to be analyzed, the video frame matched with the content of the second video frame is used as a candidate video frame, and then the timestamps corresponding to the adjacent candidate video frames are spliced, so that the picture time period of the video to be analyzed can be obtained. If a plurality of groups of video frames with continuous timestamps exist in the candidate video frames, a plurality of image time periods exist, and if only one group of video frames with continuous timestamps exists in the candidate video frames, only one image time period exists.
In the embodiment of the invention, a plurality of picture time periods of the video to be analyzed aiming at a single target copyright video may exist, so that the duration of the picture time period with the longest duration is taken as the longest continuous duration, and the sum of the durations of all the picture time periods is taken as the total continuous duration.
In the embodiment of the invention, if the video to be analyzed is clipped, the video to be analyzed is obtained by splicing the pictures in different time periods in the target copyright video, the picture time periods are not integrally continuous, the longest continuous time period is shorter, and the ratio of the time period to the time period of the target copyright video is smaller, so that whether the video to be analyzed is clipped or not can be determined according to the longest continuous time period and the time ratio of the longest continuous time period to the time period of the target copyright video.
And step 206, determining that the video to be analyzed is a copyright violation video under the condition that the total continuous time, the clip type of the video to be analyzed and the copyright type of the target copyright video are determined to meet preset copyright violation rules.
In the embodiment of the invention, the copyright type of the copyright video is determined when the copyright video is uploaded to the preset copyright library by a user, and the user can set the corresponding copyright type for each copyright video according to all parties of the copyright contained in the copyright video, the copyright importance degree, the content type and the like, so that the target copyright video and the corresponding copyright type can be directly obtained from the preset copyright library in the subsequent inquiry process.
For example: when the copyright type is divided according to all parties of the copyright, the copyright type can be divided into a self-owned copyright type and a non-self-owned copyright type, specifically, the self-owned copyright type refers to the copyright type owned by a multimedia resource platform, and the multimedia resource platform enjoys the reprogramming right, the use right and the like of the copyright video of the type; the non-self copyright type refers to a copyright type which is not owned by the multimedia resource platform, and the multimedia resource platform does not enjoy the arrangement right, the use right and the like of the copyright video of the type;
when the copyright type is divided according to the importance degree of the copyright, the copyright type can be divided into a high-risk copyright type and a low-risk copyright type, namely, the copyright video of the high-risk copyright type has higher infringement risk, and the copyright video of the low-risk copyright type has lower infringement risk;
in the copyright type division according to the content type of the copyright, the copyright type may be divided into a movie type, an animation type, a sports type, a variety type, and the like. The copyright type division mode can determine the contained content according to the copyright type corresponding to the copyright video.
Of course, the specific dividing manner of the copyright type may be determined according to the actual requirement of the user, and the embodiment of the present invention is not particularly limited.
In practical application, the preset copyright violation rules established for the owned copyright and the non-owned copyright in the multimedia platform can be different, and as the video clip is an adaptation mode of the target copyright video, for the video to be analyzed of the non-owned copyright type, if the video to be analyzed is of the clipped type, the risk of copyright violation exists in the video to be analyzed is higher, therefore, for the video to be analyzed of the clipped type, the total continuous time corresponding to the video to be analyzed is not longer, and the video to be analyzed with the longer total continuous time is determined to be the copyright violation video. For the video to be analyzed of the self-copyrighted type, the multimedia resource platform can appropriately relax the limitation, for example, set a longer total continuous duration limit compared to the total continuous duration of the non-self-copyrighted video.
Specifically, for the non-owned copyright of the multimedia resource platform, the preset copyright violation rule can be set according to the relevant regulation for judging the copyright infringement of the audio-video product in the area where the multimedia resource platform is located, for the owned copyright of the multimedia resource platform, the preset copyright violation rule can be set according to the own copyright use requirement of the platform, the preset copyright violation rule can be determined according to the actual requirement, and the realization of the scheme is taken as the standard, and no specific limitation is made here.
In the embodiment of the present invention, since the multimedia resource platform enjoys the legal rights such as the adaptation right and the usage right of the copyrighted video, although the copyright of the multimedia resource platform is violated by some illegal videos, some high-traffic users are considered, for example: the uploaded copyright violation videos of famous people such as stars and industry leaders are beneficial to the multimedia resource platform to exert the own copyright value, and the multimedia resource platform also has the ability to authorize the users to release the copyright violation analysis videos. Therefore, when the copyright violation video is of the self-owned copyright type, whether further manual review is needed or not needs to be determined according to the uploaded user account information of the user.
And 208, prompting to perform manual copyright analysis on the video to be analyzed when the user account information is of the target type.
In the embodiment of the present invention, the multimedia resource platform divides users into different types according to the user account information, such as industry, traffic level, and other dimensions, for example: known video authors, stars, general users, etc. The target type may be determined according to actual requirements of the multimedia resource platform, that is, the multimedia platform may allow some types of users to use copyrighted videos in consideration of its own interests, and take these types as the target type. Of course, in order to avoid malicious use of the copyright video of the own copyright type by the target type user, further manual review of the copyright violation video is required.
According to the embodiment of the invention, the copyright violation videos with own copyright types are further subjected to manual review according to the user account information, so that the videos to be analyzed with the own copyright types and the videos not with the own copyright types are treated in a distinguishing manner, and the own copyright of the multimedia resource platform can be fully utilized.
In the embodiment of the invention, when the total continuous time, the clip type of the video to be analyzed and the copyright type of the target copyright video do not accord with the preset copyright violation rule, the video to be analyzed is determined not to be the copyright violation video.
Optionally, the clip types include: clipped type, non-clipped type, referring to fig. 3, the step 205 includes:
In the embodiment of the present invention, since the longest continuous duration of the video to be analyzed obtained by clipping the target copyright video is usually shorter, and the longest continuous duration occupying the duration of the target copyright video is also smaller, the longest continuous duration threshold and the longest continuous duration occupying threshold may be set according to actual requirements, for example: the longest continuous time threshold is 5min, and the longest continuous time ratio threshold is 70%, so that when the longest continuous time is less than 5min and the longest continuous time ratio is less than 70%, the video to be analyzed is determined to be the edited type, otherwise, the video is the non-edited type. The maximum continuous time threshold and the maximum continuous time threshold may be determined according to actual requirements, and are not limited herein.
The method and the device for judging the copyright violation determine the type of the video to be analyzed according to the longest continuous time so as to be used for judging whether the video is the copyright violation video or not in the follow-up process, and improve the accuracy of judging the copyright violation.
Optionally, the copyright type includes: the self-copyright type and the non-self-copyright type, and the step 206 includes: and under the condition that the total continuous time, the clip type of the video to be analyzed and the copyright type of the target copyright video meet any one of the following conditions, confirming that a preset copyright violation rule is met:
conditional T1, the copyright type of the target copyrighted video is a non-owned copyright type, and the total continuous time period is greater than a first total continuous time period threshold.
In the embodiment of the present invention, for a multimedia resource platform, the infringement risk brought by a video to be analyzed related to a video to be analyzed of an owned copyright type is significantly lower than that caused by a video to be analyzed related to a non-owned copyright type, so that a preset copyright violation rule related to an owned copyright type can be set with a more relaxed restriction with respect to the non-owned copyright type when determining whether a copyright violation occurs, for example: and limiting the picture time period of the video to be analyzed of the non-self copyright type to be less than 3min, and then limiting the picture time period of the video to be analyzed of the self copyright type to be less than 5 min. The specific value can be determined according to actual requirements, and is not particularly limited herein.
In the embodiment of the present invention, when the copyright type of the target copyright video corresponding to the video to be analyzed is a non-owned copyright type, no matter whether the video to be analyzed is edited or not, if the total continuous time length of the video to be analyzed exceeds a first total continuous time length threshold value, the video to be analyzed is determined as the copyright violation type, and the first total continuous time length threshold value may be determined according to an actual requirement.
And under the condition T2, the copyright type of the target copyright video is the self-owned copyright type, the total continuous time length ratio is greater than a total continuous time length ratio threshold, the clip type of the video to be analyzed is the non-clip type, and the total continuous time length ratio is the ratio of the total continuous time length to the time length of the target copyright video.
In the embodiment of the invention, when the copyright type of the target copyright video corresponding to the video to be analyzed is the self-owned copyright type, if the video to be analyzed is the non-clipped type, that is, the video to be analyzed is not subjected to the recomposition, but the total continuous time length ratio is greater than the total continuous time length ratio threshold value, the video to be analyzed is determined to be the copyright violation video.
The embodiment of the invention determines whether the video to be analyzed is the illegal copyright video according to the total continuous time length, the clipping type and the related copyright type of the video to be analyzed, integrates multiple factors of the video to be analyzed, and improves the accuracy of judging the illegal copyright video.
Optionally, the copyright type further includes: high risk copyright type, said step 208, comprising: and under the condition that the copyright type of the target copyright video is a high-risk copyright type, the copyright type of the video to be analyzed is an edited type, and the total continuous time length is smaller than a second total continuous time length threshold value, determining that the video to be analyzed is not a copyright violation video, and adding copyright prompt information to the video to be analyzed.
In the embodiment of the invention, the high-risk copyright type can be a copyright type that the number of copyright violation videos corresponding to the history exceeds a number threshold, or a copyright with higher network heat in the current period, and the risk caused by copyright violation possibly caused by the copyright is higher, so that special auditing is required, and the specific high-risk copyright type can be determined according to the actual demand of a multimedia resource platform.
And for the video to be analyzed containing the content of the high-risk copyright type, even if the video is of the edited type, the total continuous time length is smaller than a second total continuous time length threshold value, determining that the video to be analyzed is not the copyright violation video, and issuing the video to be analyzed after adding copyright prompt information to the video to be analyzed. The copyright prompt information refers to copyright content of a target copyright video of a high-risk copyright type related to the video to be analyzed, for example: in the mode, videos to be analyzed related to high-risk copyright types can be classified at certain popular events and certain popular television dramas, so that subsequent distinction is facilitated and pushed to users.
The embodiment of the invention reduces the copyright infringement risk of the video to be analyzed by setting the copyright type with higher copyright infringement risk as the high-risk copyright type and setting a special copyright violation rule aiming at the video to be analyzed with the high-risk copyright type.
Optionally, referring to fig. 4, the step 204 includes:
in the embodiment of the present invention, because the pre-trained video recognition model has a defect of time delay when returning to the picture time period, some human eyes may determine the picture time period that is continuous, as a plurality of discontinuous picture time periods, and therefore, for the obtained picture time periods, if the time interval between adjacent picture time periods is too small, that is, smaller than the time interval threshold, for example, 0.5s, the adjacent picture time periods may be further spliced to be used as one continuous picture time period.
In the embodiment of the invention, the longest continuous time and the total continuous time are determined by using the picture time periods with the adjacent time intervals smaller than the time interval threshold, the condition that the continuous time is inaccurate due to the time delay existing in the picture time periods is overcome, the accuracy of determining the longest continuous time and the total continuous time is improved, and the accuracy of the subsequent copyright violation analysis is improved.
According to the second video copyright analysis method provided by the invention, the target copyright video matched with the existing content of the video to be analyzed is obtained from the preset copyright library through the pre-trained video identification model, then the clipping type is determined according to the longest continuous time length of the video to be analyzed matched with the content of the target copyright video, and finally whether the video to be analyzed is the illegal copyright video is determined according to the total continuous time length, the clipping type and the copyright type, so that the manual participation in the video copyright analysis process is reduced, and the video copyright analysis efficiency is improved. And the obtained adjacent picture time periods are spliced according to a time interval threshold, and special copyright violation rules are set for the video to be analyzed related to the high-risk copyright type, so that the accuracy of copyright violation analysis is improved.
Fig. 5 is a block diagram of a copyright analyzing apparatus 30 for video according to an embodiment of the present invention, including:
an obtaining module 301, configured to obtain a video to be analyzed.
The determining module 302 is configured to input the video to be analyzed into a pre-trained video recognition model, compare the video with a copyright video in a preset copyright library, and determine a target copyright video and a picture time period that is matched with the content of the target copyright video in the video to be analyzed.
And the analysis module 303 is configured to determine whether the video to be analyzed is a copyright violation video according to the target copyright video and the picture time period.
Optionally, the analysis module 303 includes:
the first determining submodule 3031 is configured to determine the longest continuous time length and the total continuous time length in the picture time period.
A second determining submodule 3032, configured to determine the clip type of the video to be analyzed according to the longest continuous time.
And a first analysis submodule 3033, configured to determine that the video to be analyzed is a copyright violation video under the condition that it is determined that the total duration, the clip type of the video to be analyzed, and the copyright type of the target copyright video meet preset copyright violation rules.
And the second analysis submodule 3034 is configured to determine that the video to be analyzed is not the copyright violation video if the video to be analyzed is not the copyright violation video.
Optionally, the clip types include: a clipped type and an unclipped type, and the second determining submodule 3032 is further configured to:
and under the condition that the longest continuous time length is less than a longest continuous time length threshold value and the longest continuous time length ratio is less than a longest continuous time length ratio threshold value, determining that the type of the video to be analyzed is a clipped type, wherein the longest continuous time length ratio is the ratio of the longest continuous time length to the time length of the target copyright video.
Otherwise, determining the type of the clip of the video to be analyzed as a non-clip type.
Optionally, the copyright type includes: the self-copyrighted type and the non-self-copyrighted type, and the first analysis sub-module 3033 is further configured to:
and under the condition that the total continuous time, the clip type of the video to be analyzed and the copyright type of the target copyright video meet any one of the following conditions, confirming that a preset copyright violation rule is met:
the copyright type of the target copyright video is a non-owned copyright type, and the total continuous time length is greater than a first total continuous time length threshold value.
The copyright type of the target copyright video is the self-owned copyright type, the total continuous time length occupation ratio is larger than the total continuous time length occupation ratio threshold value, the clip type of the video to be analyzed is the non-clip type, and the total continuous time length occupation ratio is the ratio of the total continuous time length to the time length of the target copyright video.
Optionally, the first analysis submodule 3033 is further configured to:
and under the condition that the target copyright video corresponding to the copyright violation video is of the own copyright type, acquiring user account information corresponding to the video to be analyzed.
And prompting to perform manual copyright analysis on the video to be analyzed when the user account information is of the target type.
Optionally, the copyright type further includes: the high-risk copyright type, the second analysis submodule is further used for:
and under the condition that the copyright type of the target copyright video is a high-risk copyright type, the copyright type of the video to be analyzed is an edited type, and the total continuous time length is smaller than a second total continuous time length threshold value, determining that the video to be analyzed is not a copyright violation video, and adding copyright prompt information to the video to be analyzed.
Optionally, the first determining submodule 3031 is further configured to:
and splicing the picture time sections of which the adjacent time intervals are less than or equal to the time interval threshold value.
And determining the longest continuous time and the total continuous time according to the spliced picture time period.
Optionally, the determining module 302 includes:
the third determining submodule 3021 is configured to input the video to be analyzed into a pre-trained video recognition model, so as to compare each first video frame in the video to be analyzed with each second video frame of the copyrighted video in the preset copyright library, and obtain a candidate copyrighted video in which the contents of the included second video frames are matched with the contents of the first video frames, and a corresponding matching degree.
The fourth determining sub-module 3022 is configured to determine, as the target copyrighted video, the candidate copyrighted video whose matching degree is greater than or equal to the matching degree threshold, and determine, according to the first video frame matched with the content of the second video frame, the picture time period of the video to be analyzed.
According to the video copyright analysis device, the target copyright video matched with the existing content of the video to be analyzed is obtained from the preset copyright library through the pre-trained video identification model, and then whether the video to be analyzed is the illegal copyright video or not is determined according to the picture time period of the video to be analyzed matched with the content of the target copyright video, so that manual participation in the video copyright analysis process is reduced, and the video copyright analysis efficiency is improved.
For the embodiment of the server, since it is basically similar to the method embodiment, the description is relatively simple, and for relevant points, reference may be made to part of the description of the method embodiment.
An embodiment of the present invention further provides an electronic device, as shown in fig. 6, including a processor 401, a communication interface 402, a memory 403, and a communication bus 404, where the processor 401, the communication interface 402, and the memory 403 complete mutual communication through the communication bus 404,
a memory 403 for storing a computer program;
the processor 401, when executing the program stored in the memory 403, implements the following steps:
acquiring a video to be analyzed; inputting the video to be analyzed into a pre-trained video identification model, comparing the video with a copyright video in a preset copyright library, and determining a target copyright video and a picture time period which is matched with the content of the target copyright video in the video to be analyzed; and determining whether the video to be analyzed is the illegal video of the copyright according to the target copyright video and the picture time period.
According to the embodiment of the invention, the target copyright video matched with the existing content of the video to be analyzed is obtained from the preset copyright library through the pre-trained video identification model, and then whether the video to be analyzed is the illegal copyright video is determined according to the picture time period matched with the content of the video to be analyzed and the target copyright video, so that the manual participation in the video copyright analysis process is reduced, and the efficiency of the video copyright analysis is improved.
The communication bus mentioned in the above terminal may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.
The communication interface is used for communication between the terminal and other equipment.
The Memory may include a Random Access Memory (RAM), or may also include a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the Integrated Circuit may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete gate or transistor logic device, or discrete hardware components.
In another embodiment of the present invention, a computer-readable storage medium is further provided, which stores instructions that, when executed on a computer, cause the computer to execute the copyright analysis method for video described in any of the above embodiments.
In yet another embodiment of the present invention, there is also provided a computer program product containing instructions which, when run on a computer, cause the computer to perform the method for copyright analysis of video as described in any of the above embodiments.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (8)
1. A method for analyzing copyright of video, comprising:
acquiring a video to be analyzed; the clip type of the video to be analyzed comprises a clipped type and an uncut type;
inputting the video to be analyzed into a pre-trained video identification model, comparing the video with a copyright video in a preset copyright library, and determining a target copyright video and a picture time period which is matched with the content of the target copyright video in the video to be analyzed;
determining the longest continuous duration and the total continuous duration in the picture time period;
determining the video to be analyzed as the edited type under the condition that the longest continuous time length is less than a longest continuous time length threshold value and the longest continuous time length ratio is less than a longest continuous time length ratio threshold value, and determining the video to be analyzed as the un-edited type under the condition that the longest continuous time length is less than the longest continuous time length threshold value and the longest continuous time length ratio is less than the longest continuous time length ratio threshold value; the longest continuous time length ratio is the ratio of the longest continuous time length to the time length of the target copyright video; and under the condition that the total continuous time, the clip type of the video to be analyzed and the copyright type of the target copyright video are determined to accord with preset copyright violation rules, determining that the video to be analyzed is the copyright violation video.
2. The method of claim 1, wherein the copyright type comprises: the step of confirming that the total continuous time, the clip type of the video to be analyzed and the copyright type of the target copyright video conform to preset copyright violation rules comprises the following steps:
and under the condition that the total continuous time, the clip type of the video to be analyzed and the copyright type of the target copyright video meet any one of the following conditions, confirming that a preset copyright violation rule is met:
the copyright type of the target copyright video is a non-owned copyright type, and the total continuous time length is greater than a first total continuous time length threshold value;
the copyright type of the target copyright video is the self-owned copyright type, the total continuous time length occupation ratio is larger than the total continuous time length occupation ratio threshold value, the clip type of the video to be analyzed is the non-clip type, and the total continuous time length occupation ratio is the ratio of the total continuous time length to the time length of the target copyright video.
3. The method of claim 1, wherein the copyright type further comprises: the step of determining that the video to be analyzed is not a copyright violation video comprises the following steps:
and under the condition that the copyright type of the target copyright video is a high-risk copyright type, the copyright type of the video to be analyzed is an edited type, and the total continuous time length is smaller than a second total continuous time length threshold value, determining that the video to be analyzed is not a copyright violation video, and adding copyright prompt information to the video to be analyzed.
4. The method of claim 1, wherein the step of determining the longest continuous duration and the total continuous duration in the picture period comprises:
splicing the picture time periods of which the adjacent time intervals are less than or equal to a time interval threshold value;
and determining the longest continuous time and the total continuous time according to the spliced picture time period.
5. The method according to claim 1, wherein the step of inputting the video to be analyzed into a pre-trained video recognition model to compare with the copyrighted video in a preset copyright library to determine a target copyrighted video, and the presence of a picture time period matching the target copyrighted video content in the video to be analyzed comprises:
inputting the video to be analyzed into a pre-trained video identification model so as to compare each first video frame in the video to be analyzed with each second video frame of the copyright video in the preset copyright library, and obtaining candidate copyright video and corresponding matching degree, wherein the content of the included second video frame is matched with that of the first video frame;
and determining the candidate copyright video with the matching degree larger than or equal to the matching degree threshold value as a target copyright video, and determining the picture time period of the video to be analyzed according to the first video frame matched with the second video frame content.
6. A copyright analyzing apparatus for video, comprising:
the acquisition module is used for acquiring a video to be analyzed; the clip type of the video to be analyzed comprises a clipped type and an uncut type;
the determining module is used for inputting the video to be analyzed to a pre-trained video recognition model so as to compare the video with the copyright video in a preset copyright library to determine a target copyright video and a picture time period which is matched with the content of the target copyright video in the video to be analyzed;
the first determining submodule is used for determining the longest continuous time and the total continuous time in the picture time period;
a second determining submodule, configured to determine that the video to be analyzed is the edited type when the longest continuous duration is less than a longest continuous duration threshold and a longest continuous duration ratio is less than a longest continuous duration ratio threshold, and determine that the video to be analyzed is the unetched type when the longest continuous duration is less than the longest continuous duration threshold and the longest continuous duration ratio is less than the longest continuous duration ratio threshold, where the longest continuous duration ratio is a ratio of the longest continuous duration to a duration of the target copyrighted video;
and the first analysis submodule is used for determining that the video to be analyzed is the copyright violation video under the condition that the total continuous time length, the clip type of the video to be analyzed and the copyright type of the target copyright video are determined to accord with preset copyright violation rules.
7. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of copyright analysis of video according to any one of claims 1 to 5 when executing the computer program.
8. A computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the copyright analysis method for video according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010198381.4A CN111314736B (en) | 2020-03-19 | 2020-03-19 | Video copyright analysis method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010198381.4A CN111314736B (en) | 2020-03-19 | 2020-03-19 | Video copyright analysis method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111314736A CN111314736A (en) | 2020-06-19 |
| CN111314736B true CN111314736B (en) | 2022-03-04 |
Family
ID=71148135
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010198381.4A Active CN111314736B (en) | 2020-03-19 | 2020-03-19 | Video copyright analysis method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111314736B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112383770A (en) * | 2020-11-02 | 2021-02-19 | 杭州当虹科技股份有限公司 | Film and television copyright monitoring and comparing method through voice recognition technology |
| CN113360709B (en) * | 2021-05-28 | 2023-02-17 | 维沃移动通信（杭州）有限公司 | Method and device for detecting short video infringement risk and electronic equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109145529A (en) * | 2018-09-12 | 2019-01-04 | 重庆工业职业技术学院 | A kind of text similarity analysis method and system for copyright authentication |
| CN110418161A (en) * | 2019-08-02 | 2019-11-05 | 广州虎牙科技有限公司 | Video reviewing method and device, electronic equipment and readable storage medium storing program for executing |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090028517A1 (en) * | 2007-07-27 | 2009-01-29 | The University Of Queensland | Real-time near duplicate video clip detection method |
| CN102314562A (en) * | 2010-07-01 | 2012-01-11 | 盛乐信息技术（上海）有限公司 | Video copyright detecting system and method of audio-video sharing website |
| CN102467552B (en) * | 2010-11-16 | 2014-01-22 | Tcl集团股份有限公司 | Internet device with copyright protection function and copyright protection method |
| US9996769B2 (en) * | 2016-06-08 | 2018-06-12 | International Business Machines Corporation | Detecting usage of copyrighted video content using object recognition |
| CN107750015B (en) * | 2017-11-02 | 2019-05-07 | 腾讯科技（深圳）有限公司 | Detection method, device, storage medium and the equipment of video copy |
| CN108124191B (en) * | 2017-12-22 | 2019-07-12 | 北京百度网讯科技有限公司 | A kind of video reviewing method, device and server |
| CN108647245B (en) * | 2018-04-13 | 2023-04-18 | 腾讯科技（深圳）有限公司 | Multimedia resource matching method and device, storage medium and electronic device |
| CN109660823A (en) * | 2018-12-28 | 2019-04-19 | 广州华多网络科技有限公司 | Video distribution method, apparatus, electronic equipment and storage medium |
| CN110324729B (en) * | 2019-07-18 | 2021-08-27 | 北京奇艺世纪科技有限公司 | Method, device, electronic equipment and medium for identifying infringement video link |
-
2020
- 2020-03-19 CN CN202010198381.4A patent/CN111314736B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109145529A (en) * | 2018-09-12 | 2019-01-04 | 重庆工业职业技术学院 | A kind of text similarity analysis method and system for copyright authentication |
| CN110418161A (en) * | 2019-08-02 | 2019-11-05 | 广州虎牙科技有限公司 | Video reviewing method and device, electronic equipment and readable storage medium storing program for executing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111314736A (en) | 2020-06-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10437965B2 (en) | Detection of unauthorized content in live multiuser composite streams | |
| CN110909205B (en) | Video cover determination method and device, electronic equipment and readable storage medium | |
| CN110324729B (en) | Method, device, electronic equipment and medium for identifying infringement video link | |
| CN107273508B (en) | Information processing method and device based on artificial intelligence | |
| US6961467B2 (en) | Identifying image content | |
| CN111314736B (en) | Video copyright analysis method and device, electronic equipment and storage medium | |
| US10366211B2 (en) | Method, device and apparatus for recognizing picture | |
| CN109857908B (en) | Method and apparatus for matching videos | |
| US9652551B2 (en) | Automated effort judgement of user generated content | |
| CN109685144B (en) | Method and device for evaluating video model and electronic equipment | |
| CN110149529B (en) | Media information processing method, server and storage medium | |
| CN112860943A (en) | Teaching video auditing method, device, equipment and medium | |
| CN109672936B (en) | Method and device for determining video evaluation set and electronic equipment | |
| CN112565775B (en) | Method, device and storage medium for audio and video transcoding | |
| CN111314732A (en) | Method for determining video label, server and storage medium | |
| WO2022007626A1 (en) | Video content recommendation method and apparatus, and computer device | |
| CN104486649A (en) | Video content rating method and device | |
| CN110941738A (en) | Recommendation method and device, electronic equipment and computer-readable storage medium | |
| CN112686165A (en) | Method and device for identifying target object in video, electronic equipment and storage medium | |
| CN111125429A (en) | Video pushing method and device and computer readable storage medium | |
| CN108924630B (en) | Method for displaying cache progress and playing device | |
| CN110996142B (en) | Video recall method and device, electronic equipment and storage medium | |
| US10387490B2 (en) | Video inspector | |
| CN110708361B (en) | System, method and device for determining grade of digital content publishing user and server | |
| CN111031397B (en) | Method, device, equipment and storage medium for collecting clip comments |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |