CN112597335B - Output device and output method for selecting drama - Google Patents
Output device and output method for selecting drama Download PDFInfo
- Publication number
- CN112597335B CN112597335B CN202011533971.4A CN202011533971A CN112597335B CN 112597335 B CN112597335 B CN 112597335B CN 202011533971 A CN202011533971 A CN 202011533971A CN 112597335 B CN112597335 B CN 112597335B
- Authority
- CN
- China
- Prior art keywords
- video
- processed
- time point
- module
- characteristic information
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/73—Querying
- G06F16/735—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/73—Querying
- G06F16/738—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/783—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
Abstract
The invention belongs to the technical field of multimedia, and particularly relates to an output device and an output method for selecting a drama segment, wherein the device comprises: the acquisition module is used for acquiring a key frame set of a video to be processed and characteristic information in the key frame set; the searching module is used for searching a video library according to the characteristic information and judging whether a matched video consistent with the characteristic information of the video to be processed exists in the video library or not; the matching video processing device is also used for calculating the similarity between the matching video and the video to be processed one by one according to the characteristic information and judging whether the content of the matching video is consistent with that of the video to be processed according to the similarity; the determining module is used for acquiring the expected output duration information and the time point set of the key frame set, and acquiring a clipping time point set in the time point set according to the characteristic information and the expected output duration information; and the clipping module is used for clipping the video to be processed according to the clipping time point to obtain the finished product video. The device takes the video library as a reference, and improves the production efficiency of video output.
Description
Technical Field
The invention belongs to the technical field of multimedia, and particularly relates to an output device and an output method for selecting a drama segment.
Background
Along with the popularization of mobile terminals and the acceleration of networks, the mass-flow transmission content of short videos is favored by various large platforms, vermicelli and capital, so that the traditional content is conveniently popularized by processing audio and video materials through fragments, and the method is a development direction.
Compared with the conventional movie and television series, part of the drama music is long in birth year, the old and old audio and video quality of the drama content is not high, and the parameters are different; meanwhile, the problems that the quality of the music video is influenced by the recording environment and recording equipment, and the like are also caused by noises; in addition, in order to meet the requirements on compliance of network audio-visual programs, contents such as those not included in a copyright library or newly filed drama should not be distributed, so that it is necessary to automatically screen production contents and output high-quality compliance contents. These problems all restrict the propagation of drama video.
In the related art at present, there is no method for identifying and segmenting specific characteristics of dramas and simultaneously reducing noise and improving segment selection quality. There are also various problems of low automation, poor effect, and the like when a technique commonly used in the field of approximate speech and sound separation, such as CN110493637A, is used for drama video segmentation. Therefore, at present, the drama is manually segmented mainly in a manual mode, noise content is manually marked to reduce noise, a large amount of labor is consumed for distinguishing, screening and manufacturing, time is consumed, and the finished product effect is unstable.
Therefore, in view of the above disadvantages, the present invention is urgently needed to provide an output device and an output method for selecting a drama segment.
Disclosure of Invention
The invention aims to provide an output device and an output method for selecting a drama section, which are used for solving the problems of low efficiency and unstable effect of outputting short videos of drama in the prior art.
The invention provides an output device for selecting a section of a drama, comprising: the acquisition module is used for acquiring a key frame set of a video to be processed and characteristic information in the key frame set; the searching module is used for searching a video library according to the characteristic information and judging whether a matched video consistent with the characteristic information of the video to be processed exists in the video library or not; the matching video processing device is also used for calculating the similarity between the matching video and the video to be processed one by one according to the characteristic information and judging whether the content of the matching video is consistent with that of the video to be processed or not according to the similarity; the determining module is used for acquiring the expected output duration information and the time point set of the key frame set, and acquiring a clipping time point set in the time point set according to the characteristic information and the expected output duration information; and the clipping module is used for clipping the video to be processed according to the clipping time points in the clipping time point set to obtain the finished product video.
The output device for selecting the drama passages as described above further preferably further comprises a video library, wherein the video library stores a plurality of tracks, characteristic information and a noise library; the characteristic information comprises role character names, costume styles, drama and accompaniment instruments, chapter plot introduction and chapter complete duration information, and the role character names, the costume styles, the drama and accompaniment instruments, the chapter plot introduction and the chapter complete duration information are all associated with the songs.
In the above-described output device for a drama selection chapter, it is further preferable that the acquisition module includes: a first extraction module: the system comprises a key frame set, a video processing unit and a video processing unit, wherein the key frame set is used for acquiring a video to be processed, extracting key frames of the video to be processed at equal frame intervals and forming the key frame set; a second extraction module: the method is used for identifying the key frames in the key frame set frame by frame, acquiring the characteristic information in the key frames and constructing a set for each item of characteristic information.
The above-mentioned apparatus for outputting a drama selection chapter further preferably includes: the first analysis module is used for searching a video library according to the clothing style, the role character name, the drama and the accompaniment instrument of the video to be processed to obtain matched videos which have the same clothing style and/or the role character name and/or the drama and the accompaniment instrument with the video to be processed, and a plurality of matched videos form a track set; and the second analysis module is used for calculating the similarity between the video to be processed and each element in the song set one by one according to the characteristic information and judging whether the content of the matched video is consistent with that of the video to be processed or not according to the similarity.
In the output device for a drama chapter, it is preferable that the similarity calculation formula in the second analysis module is:
wherein D is the similarity, j is the serial number of the track in the track set, and j is 1,2,3,4 … n; x is the number of jfk Is an element of a set Xj, where Xj is a set of times of occurrence of the set Fj in the key frame set of the track with sequence number j in the track set, and Xj ═ xjf1, xjf2, … …, xjfk }; x is the number of 0fk Is an element of the set X0, where X0 is a set of times of occurrence of the Fj set in the key frame set of the video to be processed, and X0 ═ X0f1, X0f2, … …, X0fk }; fj is the union of the video clothing style set to be processed and the clothing style set of the track with sequence number j in the track set, and Fj ═ fa1, fa2, … …, fak }; k is 1,2,3,4 … e;
y jfi the element is an element of a set Yj, where Yj is a set of times of occurrence of a set Rj in a key frame set of a track with sequence number j in the track set, and Yj ═ yjf1, yjf2, … …, yjfi }; y is 0fi Is an element of the set Y0, where Y0 is the set of times of occurrence of the Rj set in the key frame set of the video to be processed, and Y0 ═ Y0f1, Y0f2, … …, Y0fi }; rj is a union of the character name sets of the videos to be processed and the character name sets of the tracks with the sequence number j in the track sets, and Rj is { ra1, ra2, … … and rai }; i is 1,2,3,4 … t.
In the aforementioned output device for a drama selection piece, it is further preferable that the determination module includes: the time length confirming module is used for acquiring the expected output time length and the floating range of the expected output time length; the node extraction module is used for acquiring a time point set of a key frame of a video to be processed and obtaining a time point subset of the time point set of the key frame according to the characteristic information, wherein the time point set of the key frame comprises a clothing style change time point and a clothing style change time point of a character name of a role; and the node confirmation module is used for extracting a plurality of pairs of elements in the time point subset to form a clipping time point set, and the time difference of each element of the clipping time point set falls into the range formed by the expected output duration and the floating range thereof.
The output device for selecting the drama passages as described above, further preferably, the determining module further comprises a duration comparing module, and the duration comparing module is configured to determine whether the expected output duration is less than the duration of the video to be processed; if not, the clipping time point is the time of the start and the end of the video to be processed; and if so, acquiring a clipping time point set according to the node extraction module and the node confirmation module.
The apparatus for outputting a drama selection segment as described above, preferably, the clip module includes: the video splitting module is used for cutting the video to be processed according to the cutting time point to obtain an output video set; the audio processing module is used for acquiring the audio waveform of each element in the output video set, comparing the audio waveform of each element with the noise waveform and removing the noise of the audio waveform to obtain a processed audio; and the video output module is used for mixing the processed audio with the corresponding output video to obtain a finished product video.
The invention also provides a method for selecting the segment of the output quality of the ultra-high definition audio/video, which is used for the output device for selecting the segment of the drama, and comprises the following steps: s1: the method comprises the steps that an acquisition module acquires a key frame set of a video to be processed and characteristic information in the key frame set; s2: the searching module is used for searching a video library according to the characteristic information and judging whether a matched video with the same characteristic information as the video to be processed exists in the video library or not; s3: the searching module calculates the similarity between the matched video and the video to be processed and judges whether the matched video is consistent with the video to be processed; s4: the determining module acquires time point sets of the expected output duration information and the key frame set and acquires a clipping time point set in the time point sets according to the characteristic information and the expected output duration information; s5: and the clipping module clips the video to be processed according to the clipping time point to obtain a finished product video.
In the method for selecting the ultra-high-definition audio/video output quality segment, S4 further preferably includes: s41: the time length confirming module obtains expected output time length and a floating range of the expected output time length; s42: the time length comparison module judges whether the expected output time length is less than the time length of the video to be processed; if not, the clipping time point is the time of the start and the end of the video to be processed; if so, acquiring a time point set of the key frame set, and carrying out the next step; s43: acquiring a time point subset which simultaneously comprises a clothing style change time point and a role character name clothing style change time point according to the characteristic information; s44: and extracting a plurality of pairs of elements of which the time difference in the time point subset falls in a range formed by the expected output duration and the floating range of the expected output duration to obtain a clipping time point set.
The device comprises an acquisition module, a search module, a confirmation module and a clipping module, wherein the search module can search a video library according to characteristic information, judge whether a matched video consistent with the characteristic information of a video to be processed exists in the video library, calculate the similarity between the matched video and the video to be processed one by one according to the characteristic information and judge whether the content of the matched video is consistent with that of the video to be processed according to the similarity; the confirmation module can acquire the expected output duration information and the time point set of the key frame set, and acquire the clipping time point set in the time point set according to the characteristic information and the expected output duration information. Through the device, the plot transition point can be obtained according to the characteristic message of the video to be processed, and the editing time point is further obtained, so that the selection of the drama songs is automatically and efficiently carried out according to the output drama, the obtained editing time points are coherent, the content of the selected segments is high in quality, and meanwhile, the output high-quality selected segments can be ensured not to exceed the preset range, and the related requirements of the law and regulations for propagation can be met.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a block diagram of an output device for a selection of a drama in the present invention;
FIG. 2 is a flow chart of the method for outputting the selection of drama;
FIG. 3 is a flowchart of obtaining a clipping time point according to the present invention.
Detailed Description
Example 1:
fig. 1 discloses a module connection diagram of an output device for a drama selection in the present embodiment, and specifically, as shown in fig. 1, the output device for the drama selection in the present embodiment includes:
the acquisition module is used for acquiring a key frame set of a video to be processed and characteristic information in the key frame set;
the searching module is used for searching a video library according to the characteristic information and judging whether a matched video consistent with the characteristic information of the video to be processed exists in the video library or not; the matching video processing device is also used for calculating the similarity between the matching video and the video to be processed one by one according to the characteristic information and judging whether the content of the matching video is consistent with that of the video to be processed or not according to the similarity;
the determining module is used for acquiring the expected output duration information and the time point set of the key frame set, and acquiring a clipping time point set in the time point set according to the characteristic information and the expected output duration information;
and the clipping module is used for clipping the video to be processed according to the clipping time points in the clipping time point set to obtain the finished product video.
Further, the system also comprises a video library, wherein a plurality of tracks, characteristic information and noise libraries are stored in the video library; the feature information comprises character names, costume styles, dramas and accompanying musical instruments, chapter scenario introduction and chapter complete duration information, and the character names, the costume styles, the dramas and accompanying musical instruments, the chapter scenario introduction and the chapter complete duration information are all associated with the songs.
Specifically, the feature information includes a character name, a clothing style, a genre, an accompaniment instrument, a chapter scenario introduction, and chapter complete duration information, and the character name, the clothing style, the genre, the accompaniment instrument, the chapter scenario introduction, and the chapter complete duration information are all associated with the track. Specifically, every song can correspond a drama information, chapter plot introduction, chapter complete time length information and a plurality of role personage names, multiple dress style, multiple accompaniment musical instrument, according to drama information, chapter plot introduction, chapter complete time length information and arbitrary a role personage name, dress style, accompaniment musical instrument homoenergetic obtain rather than the song of being correlated with, and in the same way, also can obtain characteristic information according to the song. The above arrangement is suitable for bidirectional retrieval, i.e., retrieving the corresponding track according to the feature information, and retrieving the corresponding feature information according to the track information.
The noise library includes a plurality of noise effects.
Further, the acquiring module includes: a first extraction module: the system comprises a video acquisition module, a key frame matching module and a video processing module, wherein the video acquisition module is used for acquiring a video to be processed and extracting key frames of the video to be processed at equal frame intervals to form a key frame set; a second extraction module: the method is used for identifying the key frames in the key frame set frame by frame, acquiring the characteristic information in the key frames and constructing a set for each item of characteristic information.
Specifically, the first extraction module obtains the video h0 to be processed, and extracts key frames from the video h0 with a medium frame distance according to a control factor K, where the extracted key frame set is B0. The size of the control factor K controls the total number of key frames extracted from the original video, and the value of the control factor K can be input or given a fixed value.
The second extraction module is configured to extract feature information in the key frame set B0, specifically, the extracted costume style information constitutes a costume style set F0, the extracted character name information constitutes a character set R0, and the extracted accompaniment instrument information constitutes an accompaniment instrument set P0.
Further, the searching module comprises: the first analysis module is used for searching a video library according to the clothing style, the role character name, the drama and the accompaniment instrument of the video to be processed to obtain matched videos which have the same clothing style and/or the role character name and/or the drama and the accompaniment instrument with the video to be processed, and a plurality of matched videos form a track set; and the second analysis module is used for calculating the similarity between the video to be processed and each element in the track set one by one according to the characteristic information and judging whether the content of the matched video is consistent with that of the video to be processed or not according to the similarity.
Specifically, the first analysis module is configured to search a video library according to the clothing style set F0, the character person set R0, and the accompaniment instrument set P0, further obtain a track set Hf corresponding to the clothing style set F0, a track set Hr corresponding to the character person set R0, and a track set Hb corresponding to the accompaniment instrument set P0, and further obtain a track set H1 according to the track set Hf, the track set Hr, and the track set Hb, where H1 ═ Hf ═ Hr ═ Hb ═ H1, H2, … …, hn }. If the track set H1 is empty, the video to be processed is not contained in the video library, and the subsequent processing is stopped; if the track set H1 is not empty, then the tracks in the track set H1 need to be matched and filtered.
The second analysis module is used for matching and screening the tracks in the track set H1, specifically, the second analysis module calculates the similarity between each element in the track set H1 and the original video one by one, wherein the calculation formula of the similarity is as follows:
the calculation formula is as follows:
wherein D is the similarity, j is the serial number of the track in the track set H1, and j is 1,2,3,4 … n;
x jfk is an element of a set Xj, where Xj is a set of times of occurrence of the set Fj in the key frame set of the track with sequence number j in the track set, and Xj ═ xjf1, xjf2, … …, xjfk };
x 0fk is an element of the set X0, where X0 is a set of times of occurrence of the Fj set in the key frame set of the video to be processed, and X0 ═ X0f1, X0f2, … …, X0fk };
fj is the union of the video costume style set F0 to be processed and the costume style set Fjw of the track with sequence number j in the track set, and Fj ═ F0 ═ Fjw ═ fa1, fa2, … …, fak }; k is 1,2,3,4 … e;
y jfi is an element of a set Yj, where Yj is a key frame of a track with sequence number j in the track setThe number of occurrences of the Rj set in the set, Yj ═ yjf1, yjf2, … …, yjfi };
y 0fi is an element of the set Y0, where Y0 is the set of times of occurrence of the Rj set in the key frame set of the video to be processed, and Y0 ═ Y0f1, Y0f2, … …, Y0fi };
rj is a union of the character name set R0 of the video to be processed and the character name set Rjc of the track with sequence number j in the track set, Rj ═ R0 ═ Rjc ═ ra1, ra2, … …, rai }; i is 1,2,3,4 … t.
And obtaining the similarity values D1, D2, … … and Dn of the track set H1 according to a similarity formula, wherein the smaller the similarity value is, the more similar the elements in the track set H1 are proved to be to the video to be processed. Comparing the sizes of D1, D2, … … and Dn, and selecting the element of the minimum Dx in the track set H1 as the track most similar to the original video. Setting a similarity threshold value W, and when Dx is larger than W, indicating that the core song library does not contain the original video song, and ending the processing; when Dx is smaller than W, it indicates that the track hx corresponding to Dx is the track corresponding to the original video h0, and the process continues.
Further, the determining includes: the time length confirming module is used for acquiring the expected output time length and the floating range of the expected output time length; the node extraction module is used for acquiring a time point set of the key frames of the video to be processed and obtaining a time point subset of the time point set of the key frames according to the characteristic information, wherein the time point set of the key frames comprises a clothing style change time point and a clothing style change time point of the role character name; and the node confirmation module is used for extracting a plurality of pairs of elements in the time point subset to form a clipping time point set, and the time difference of each element of the clipping time point set falls into the range formed by the expected output duration and the floating range thereof.
Specifically, the duration confirmation module is configured to obtain a duration of the desired output segment, i.e., the desired output duration S1, and an acceptable floating range P of the desired output duration S1.
The time point set corresponding to the key frame set B0 extracted from the video to be processed is T, T ═ T1, T2, … …, tn }. In key frame set B0, the set of time points for the clothing style change is Tx, Tx ═ Tx1, Tx2, … …, txn }; in the key frame set B0, the set of time points at which the dress style change occurs in the character name change is Ty, Ty ═ Ty1, Ty2, … …, tyn }. Meanwhile, the clothing style change time point and the time point subset Tz of the person name change time point are the intersection of the clothing style change time point set and the person name change time point set, namely Tz equals to Tx ═ Ty.
The node validation module is used for extracting the clipping time points of which the time difference values of the elements in the time point subset Tz are between S1-P to S1+ P, and forming a clipping time point set A.
Preferably, the determining module further comprises a duration comparing module, and the duration comparing module is configured to determine whether the expected output duration is less than the duration of the video to be processed; if not, the clipping time point is the time of the start and the end of the video to be processed; and if so, acquiring a clipping time point set according to the node extraction module and the node confirmation module. The time length comparison module is arranged in front of the node extraction module and used for avoiding subsequent processing in the condition of S1 ≧ S0 and further used for improving the processing speed.
Further, the clipping module includes: the video splitting module is used for cutting a video to be processed according to the cutting time point to obtain an output video set; the audio processing module is used for acquiring the audio waveform of each element in the output video set, comparing the audio waveform of each element with the noise waveform and removing the noise of the audio waveform to obtain a processed audio; and the video output module is used for mixing the processed audio with the corresponding output video to obtain a finished product video.
Example 2:
fig. 2 is a flowchart of an output method of a drama chapter according to the present invention, and as shown in fig. 2, the present embodiment discloses an output method of a drama chapter, which is applied to an output device of a drama chapter according to any one of embodiments 1, and includes:
s1: the method comprises the steps that an acquisition module acquires a key frame set of a video to be processed and characteristic information in the key frame set;
s2: the searching module is used for searching a video library according to the characteristic information and judging whether a matched video with the same characteristic information as the video to be processed exists in the video library or not;
s3: the searching module calculates the similarity between the matched video and the video to be processed and judges whether the matched video is consistent with the video to be processed;
s4: the determining module acquires the expected output duration information and a time point set of the key frame set, and acquires a clipping time point set in the time point set according to the characteristic information and the expected output duration information;
s5: and the clipping module clips the video to be processed according to the clipping time point to obtain a finished product video.
Fig. 3 discloses a flowchart of acquiring a clip time point, and as shown in fig. 3, S4 further includes:
s41: the time length confirming module obtains expected output time length and a floating range of the expected output time length;
s42: the time length comparison module judges whether the expected output time length is less than the time length of the video to be processed; if not, the clipping time point is the time of the start and the end of the video to be processed; if so, acquiring a time point set of the key frame set, and carrying out the next step;
s43: acquiring a time point subset which simultaneously comprises a clothing style change time point and a role character name clothing style change time point according to the characteristic information;
s44: and extracting a plurality of pairs of elements of which the time difference in the time point subset falls in a range formed by the expected output duration and the floating range of the expected output duration to obtain a clipping time point set.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (7)
1. An output device for selecting a chapter of a drama, comprising:
the acquisition module is used for acquiring a key frame set of a video to be processed and characteristic information in the key frame set;
the searching module is used for searching a video library according to the characteristic information and judging whether a matched video consistent with the characteristic information of the video to be processed exists in the video library or not; the matching video processing device is also used for calculating the similarity between the matching video and the video to be processed one by one according to the characteristic information and judging whether the content of the matching video is consistent with that of the video to be processed or not according to the similarity;
the determining module is used for acquiring the expected output duration information and the time point set of the key frame set, and acquiring a clipping time point set in the time point set according to the characteristic information and the expected output duration information;
the editing module is used for editing the video to be processed according to the editing time points in the editing time point set to obtain a finished product video;
the determining module comprises:
the time length confirming module is used for acquiring the expected output time length and the floating range of the expected output time length;
the node extraction module is used for acquiring a time point set of the key frames of the video to be processed and obtaining a time point subset of the time point set of the key frames according to the characteristic information, wherein the time point set of the key frames comprises a clothing style change time point and a clothing style change time point of the role character name;
the node confirmation module is used for extracting a plurality of pairs of elements in the time point subset to form a clipping time point set, and the time difference of each element in the clipping time point set falls in a range formed by the expected output duration and the floating range of the expected output duration;
the determining module further comprises a time length comparing module, and the time length comparing module is used for judging whether the expected output time length is less than the time length of the video to be processed; if not, the clipping time point is the time of the start and the end of the video to be processed; and if so, acquiring a clipping time point set according to the node extraction module and the node confirmation module.
2. The apparatus as claimed in claim 1, further comprising a video library, wherein a plurality of video, feature information and noise libraries are stored in the video library; the characteristic information comprises character names, costume styles, dramas and accompanying musical instruments, chapter scenario introduction and chapter complete duration information, and the character names, the costume styles, the dramas and accompanying musical instruments, the chapter scenario introduction and the chapter complete duration information are all associated with the videos.
3. The apparatus as claimed in claim 2, wherein the acquiring module comprises:
a first extraction module: the system comprises a key frame set, a video processing unit and a video processing unit, wherein the key frame set is used for acquiring a video to be processed, extracting key frames of the video to be processed at equal frame intervals and forming the key frame set;
a second extraction module: the method is used for identifying the key frames in the key frame set frame by frame, acquiring the characteristic information in the key frames and constructing a set for each item of characteristic information.
4. The apparatus as claimed in claim 3, wherein the search module comprises:
the first analysis module is used for searching a video library according to the clothing style, the role character name, the drama and the accompaniment instrument of the video to be processed to obtain matched videos which have the same clothing style and/or role character name and/or drama and accompaniment instrument with the video to be processed, and a plurality of matched videos form a track set;
and the second analysis module is used for calculating the similarity between the video to be processed and each element in the song set one by one according to the characteristic information and judging whether the content of the matched video is consistent with that of the video to be processed or not according to the similarity.
5. The apparatus as claimed in claim 4, wherein the similarity in the second analysis module is calculated by the following formula:
wherein D is the similarity, j is the serial number of the track in the track set, and j is 1,2,3,4 … n;
x jfk the element is a set Xj, where Xj is a set of times of occurrence of the Fj set in the key frame set of the track with the sequence number j in the track set, and Xj ═ xjf1, xjf2, … …, xjfk };
x 0fk is an element of the set X0, where X0 is a set of times of occurrence of the Fj set in the key frame set of the video to be processed, and X0 ═ X0f1, X0f2, … …, X0fk };
fj is the union of the video costume style set to be processed and the costume style set of the song with sequence number j in the song set, and Fj ═ fa1, fa2, … …, fak }; k is 1,2,3,4 … e;
y jfi the element is an element of a set Yj, where Yj is a set of times of occurrence of a set Rj in a key frame set of a track with sequence number j in the track set, and Yj ═ yjf1, yjf2, … …, yjfi };
y 0fi is an element of the set Y0, where Y0 is the set of times of occurrence of the Rj set in the key frame set of the video to be processed, and Y0 ═ Y0f1, Y0f2, … …, Y0fi };
rj is the union of the role character name set of the video to be processed and the role character name set of the track with the sequence number j in the track set, and Rj is { ra1, ra2, … …, rai }; i is 1,2,3,4 … t.
6. The apparatus as claimed in claim 5, wherein the editing module comprises:
the video splitting module is used for cutting a video to be processed according to the cutting time point to obtain an output video set;
the audio processing module is used for acquiring the audio waveform of each element in the output video set, comparing the audio waveform of each element with the noise waveform and removing the noise of the audio waveform to obtain a processed audio;
and the video output module is used for mixing the processed audio with the corresponding output video to obtain a finished product video.
7. An output method of a drama selection, which is applied to the output device of the drama selection of any one of claims 1 to 6, comprising:
s1: the method comprises the steps that an acquisition module acquires a key frame set of a video to be processed and characteristic information in the key frame set;
s2: the searching module is used for searching a video library according to the characteristic information and judging whether a matched video with the same characteristic information as the video to be processed exists in the video library or not;
s3: the searching module calculates the similarity between the matched video and the video to be processed and judges whether the content of the matched video is consistent with that of the video to be processed;
s4: the determining module acquires the expected output duration information and a time point set of the key frame set, and acquires a clipping time point set in the time point set according to the characteristic information and the expected output duration information;
s5: the editing module edits the video to be processed according to the editing time point to obtain a finished product video;
s4 further includes:
the time length confirming module obtains expected output time length and a floating range of the expected output time length;
the node extraction module acquires a time point set of a key frame of a video to be processed, and obtains a time point subset of the time point set of the key frame according to the characteristic information, wherein the time point set of the key frame comprises a clothing style change time point and a role figure name clothing style change time point;
the node confirmation module extracts a plurality of pairs of elements in the time point subset to form a clipping time point set, and the time difference of each element in the clipping time point set falls in a range formed by the expected output duration and the floating range of the expected output duration;
the time length comparison module judges whether the expected output time length is less than the time length of the video to be processed; if not, the clipping time point is the time of the start and the end of the video to be processed; and if so, acquiring a clipping time point set according to the node extraction module and the node confirmation module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011533971.4A CN112597335B (en) | 2020-12-21 | 2020-12-21 | Output device and output method for selecting drama |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011533971.4A CN112597335B (en) | 2020-12-21 | 2020-12-21 | Output device and output method for selecting drama |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112597335A CN112597335A (en) | 2021-04-02 |
| CN112597335B true CN112597335B (en) | 2022-08-19 |
Family
ID=75200350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011533971.4A Active CN112597335B (en) | 2020-12-21 | 2020-12-21 | Output device and output method for selecting drama |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112597335B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113422981B (en) * | 2021-06-30 | 2023-03-10 | 北京华录新媒信息技术有限公司 | Method and device for identifying opera based on ultra-high definition opera video |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6774917B1 (en) * | 1999-03-11 | 2004-08-10 | Fuji Xerox Co., Ltd. | Methods and apparatuses for interactive similarity searching, retrieval, and browsing of video |
| CN104166685A (en) * | 2014-07-24 | 2014-11-26 | 北京捷成世纪科技股份有限公司 | Video clip detecting method and device |
| CN106570165A (en) * | 2016-11-07 | 2017-04-19 | 北京航空航天大学 | Content-based video retrieval method and apparatus |
| CN109889882A (en) * | 2019-01-24 | 2019-06-14 | 北京亿幕信息技术有限公司 | A kind of video clipping synthetic method and system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101859585B (en) * | 2010-07-01 | 2011-08-03 | 福建省三奥信息科技股份有限公司 | System and method for frame-accuracy cutting of video material |
| US8731292B2 (en) * | 2011-01-07 | 2014-05-20 | Alcatel Lucent | Method and apparatus for comparing videos |
-
2020
- 2020-12-21 CN CN202011533971.4A patent/CN112597335B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6774917B1 (en) * | 1999-03-11 | 2004-08-10 | Fuji Xerox Co., Ltd. | Methods and apparatuses for interactive similarity searching, retrieval, and browsing of video |
| CN104166685A (en) * | 2014-07-24 | 2014-11-26 | 北京捷成世纪科技股份有限公司 | Video clip detecting method and device |
| CN106570165A (en) * | 2016-11-07 | 2017-04-19 | 北京航空航天大学 | Content-based video retrieval method and apparatus |
| CN109889882A (en) * | 2019-01-24 | 2019-06-14 | 北京亿幕信息技术有限公司 | A kind of video clipping synthetic method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112597335A (en) | 2021-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10497378B2 (en) | Systems and methods for recognizing sound and music signals in high noise and distortion | |
| CN105868397B (en) | Song determination method and device | |
| EP1081960B1 (en) | Signal processing method and video/voice processing device | |
| US7826708B2 (en) | System and method for automatically customizing a buffered media stream | |
| US20060080356A1 (en) | System and method for inferring similarities between media objects | |
| JP2003517786A (en) | Video production system and method | |
| KR20060020114A (en) | System and method for providing music search service | |
| CN111242110B (en) | Training method of self-adaptive conditional random field algorithm for automatically breaking news items | |
| US7349477B2 (en) | Audio-assisted video segmentation and summarization | |
| CN112597335B (en) | Output device and output method for selecting drama | |
| CN110933459B (en) | Event video clipping method, device, server and readable storage medium | |
| CN113115055A (en) | User portrait and live video file editing method based on viewing behavior | |
| KR101389730B1 (en) | Method to create split position accordance with subjects for the video file | |
| JP2001147697A (en) | Method and device for acoustic data analysis | |
| JP5257356B2 (en) | Content division position determination device, content viewing control device, and program | |
| CN116489449A (en) | Video redundancy fragment detection method and system | |
| JP2000285242A (en) | Signal processing method and video sound processing device | |
| CN113569086B (en) | Method, device, terminal equipment and readable storage medium for aggregating curved libraries | |
| CN113012723B (en) | Multimedia file playing method and device and electronic equipment | |
| Shao et al. | Automatically generating summaries for musical video | |
| Haloi et al. | Unsupervised story segmentation and indexing of broadcast news video | |
| US20060092327A1 (en) | Story segmentation method for video | |
| CN117750148A (en) | Method and system for identifying dramatic subtitles, electronic equipment and storage medium | |
| JP5424306B2 (en) | Information processing apparatus and method, program, and recording medium | |
| CN117651157A (en) | Lyric file generation method, computer device, and computer-readable storage medium |
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 |