CN103473220B - The automatic merogenesis of documentary film content based on subtitle file and the automatic generation method of subhead thereof - Google Patents
The automatic merogenesis of documentary film content based on subtitle file and the automatic generation method of subhead thereof Download PDFInfo
- Publication number
- CN103473220B CN103473220B CN201310417087.8A CN201310417087A CN103473220B CN 103473220 B CN103473220 B CN 103473220B CN 201310417087 A CN201310417087 A CN 201310417087A CN 103473220 B CN103473220 B CN 103473220B
- Authority
- CN
- China
- Prior art keywords
- subhead
- phrase
- chapters
- automatic
- sections
- 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
Abstract
A kind of automatic merogenesis of documentary film content based on subtitle file of the present invention and the automatic generation method of subhead thereof, belong to Chinese text field of information processing, automatically generated for two stages form by the automatic merogenesis of content and subhead. In the automatic merogenesis of the content stage, the size in the time interval occurring according to the adjacent captions that record in documentary film subtitle file, selects several the longest positions of interval as cut-point, and whole documentary film is divided into some chapters and sections in time. At the automatic generation phase of chapters and sections subhead, according to previous stage in definite chapters and sections start-stop moment, under the guidance of certain selecting rule, in the commentary of these chapters and sections, automatically select a phrase combination with " attribute+head " this syntactic structure, as the subhead of these chapters and sections. The time that mode is carried out segmentation to longer documentary film according to semanteme has manually been shortened in this invention comparatively significantly, has greatly improved teacher and has adopted the documentary film fragment perfect in workmanship enthusiasm as teaching material.
Description
Technical field
The invention belongs to Chinese text processing field, be specifically related to a kind of documentary film content based on subtitle file and automatically divideThe automatic generation method of joint and subhead thereof.
Background technology
Modernization, high-caliber Activities for Teaching need teacher extensively to collect for instructional objective and are well-organized variousThe teaching material of various kinds. The science and education channel of American National geography, Discovery channel and Chinese Central Television, documentary film frequencyThe famous documentary film brand column such as road is undoubtedly one of important sources of high-quality teaching material. Adopt such multimediaTeaching material not only can enrich teacher's the content of courses, can also increase student's classroom impression and excite student's study emergingInterest.
But the time span of documentary film is generally more than 45 minutes, quite a few reaches 1 and a half hours, even 2Hour, considerably beyond the time of a class. All finish whole documentary film and not only easily depart from instructional objective, and can takeToo much time in classroom. In general, the content that meets the current instructional objective of teacher in certain section of documentary film is approximately 3-5 minute,Only account for this section of very little part of the whole durations of documentary film. If there is no one automatic mode easily, only depend on human eye to browseThis section of material of each width picture removal search of documentary film, will inevitably significantly reduce the efficiency of prepare lesson.
At present, carry out segmentation to a video file, a long large film of time is divided into when a series ofBetween shorter little film, had the method for comparative maturity. The difference of these methods is the foundation of cutting apart. Fairly simpleTo carry out a segmentation according to the thinking of decile in time, or according to being carried out segmentation by the fragment popularity of program request. RelativelyComplicated is to carry out segmentation according to the switching of the variation of scene or camera lens. But existing method is subject to the restriction (ratio of its demandAs film, segmentation is the displacement in order to promote video server buffer memory or extracts certain section of certain content as soccer goal-shooting), also depositTwo defects relatively more outstanding: the one, be unfavorable for semantic meaning representation, unless finished watching from the beginning to the end, user is unclear to be splitCertain section of film there is which type of content; The 2nd, be unfavorable for content search, unless finished watching from the beginning to the end unclear point of userWhether certain section of film that cuts out has other content that oneself is concerned about.
The appearance of subtitle file, has prepared condition for using computer to make up above-mentioned defect.
For ensureing that hearing has the rights and interests of impedient people, the relevant laws of the United Nations's pact and China are all encouraged and are advocated at soundProvide captions as goods. Now, in all channels and column of the Chinese Central Television (CCTV), except only a few part programme televised live, broadcastThe TV programme that go out all have captions.
In general, captions are divided into two kinds of " soft ", " firmly ". So-called " soft " captions, refer to that audio-visual product is in compacting,Using caption content as a track (track) independently, separate with Voice & Video content, be encapsulated into a file (as MKVThe file of form) in, " plug-in captions " that those need title plug-ins support, can preserve separately also belong to this class, withImage content adheres to different video playback stream separately. And " firmly " captions refer to that caption content is taken as watermark and is embedded in videoRong Zhong, in the same data group of image content in video playback stream. Now, existing ripe method is extracted video contentIn " firmly " captions, its thinking is roughly divided into two kinds: the extraction based on font edge and carrying based on font color connected regionGet. Both common ground are that final step all will, through the identification of OCR word, form character cell, has recorded word thereby generateThere is the subtitle file of start-stop moment and caption content in curtain.
Summary of the invention
The present invention is exactly for above-mentioned the deficiencies in the prior art, and a kind of documentary film content based on subtitle file providingAutomatically the automatic generation method of merogenesis and subhead thereof, for improving prepare lesson efficiency, enriching class teaching content and improve and learnRaw interest provides technical support.
The present invention seeks to realize by following technological means.
The automatic merogenesis of documentary film content based on subtitle file and an automatic generation method for subhead thereof, the method is dividedFor the automatic merogenesis of content and subhead two stages of generation automatically, use respectively time and the Word message in subtitle file.Subtitle file, taking sentence as unit, is stored it for information about, and common basic format is as follows.
The common basic format of table 1 subtitle file
The object in automatic merogenesis stage of content is in order to be divided into some small videos to the documentary film generally with larger durationChapters and sections. Documentary film is read aloud the literary composition matching with image content conventionally in the mode of offscreen voice, with gentle keynote by an announcerWord script, some documentary film is also aided with background music or interspersed personage dialogue. In the process of reading aloud announcer, pause longerThe time interval often can be used as the position that certain section of explanation comes to an end, is about to have a new paragraph. This special phenomenon reflectionIn subtitle file, showing as certain captions, to start moment of occurring larger with the difference in the moment of upper captions disappearance.Merogenesis is not only simple and easy to do accordingly, and has the significance from semantically carrying out segmentation. Concrete steps are as follows:
The first step, every a line of traversal subtitle file, always total N bar captions. If find that its initial character is numeral, saysBright this is the first row of new captions, as shown in table 1. Record its numbering i and extract and start accordingly T nowbegin(i) and disappear moment Tend(i)。
Second step, calculates current captions and starts now and upper captions and disappear time difference in moment:，（i>1）。
The 3rd step, sorted from big to small to (N-1) individual such time difference. Obtain several i that come above. ThisA little i are the position of content merogenesis. Advise that such i is no more than 9, is divided at most 10 chapters and sections.
The object of the automatic generation phase of subhead is for the some little video to forming in the automatic merogenesis of the content stageChapters and sections are set up navigation or index. Automatically generate not only a grammaticalness but also proper subhead, specifically at subtitle fileWord content in excavate the character string of " nominal endocentric phrase ", as the subhead of corresponding chapters and sections. Its step is as follows:
The first step, according to the chapters and sections temporal information that the automatic merogenesis of the content stage divides, extracts in the time of this chapters and sections start-stopCaption character content between quarter.
Second step, the caption character content that uses automatic word segmentation instrument to extract the first step carries out participle, and to separatingWord mark part of speech.
The 3rd step, from first to last scans the word that second step separates in order one by one, if find that this word is noun or shapeHold word, and next word is noun, extracts with regard to the phrase that sets it as " nominal endocentric phrase ". These phrases are as these chapters and sectionsCandidate's phrase of subhead.
The 4th step through certain selecting rule, is selected a subhead as these chapters and sections from multiple candidate's phrases.These selecting rules are listed as follows, and the rule that numbering comes above is preferentially used.
(1), if candidate's phrase only has 1, it just directly becomes subhead, and needn't use rule below.
(2) if candidate's phrase more than 1, that is its head just relatively, " noun+()+noun " or " adjective+()+noun " second notional word in structure or the word that finally that part of speech is noun, in these chapters and sections, occur timeNumber. If candidate's phrase that head occurrence number is maximum only has 1, it just becomes subhead so, and after needn't usingRule. If the maximum candidate's phrase of generative center language occurrence number is more than 1, so such candidate's phrase will enter downOne take turns select.
(3) attribute of these candidate's phrases of comparison, i.e. " noun+()+noun " or " adjective+()+noun " structureIn first notional word, namely first noun or " adjective+()+noun " in " noun+()+noun " structureAdjective, the number of times of the appearance in these chapters and sections. If candidate's phrase that attribute head occurrence number is maximum only has 1,It just becomes subhead so, and needn't use rule below. If candidate's phrase that attribute head occurrence number is maximumMore than 1, so such candidate's phrase will enter selecting of next round.
(4) number of words of these candidate's phrases of comparison. If candidate's phrase that number of words is maximum only has 1, it just becomes soSubhead, and needn't use rule below. If the maximum candidate's phrase of number of words is more than 1, so such candidate's phraseSelecting that last takes turns will be entered.
(5) order that relatively these candidate's phrases occur, what first occur just becomes subhead.
The present invention proposes a kind of automatic merogenesis of documentary film content based on subtitle file and the automatic generation of subhead thereofMethod, has solved and has adopted mode manually longer documentary film to be needed to the technical problem of long period according to semantic segmentation.This invention was made up of the automatic merogenesis of content and subhead two stages of automatic generation. In the automatic merogenesis of the content stage, according to documentary filmThe size in the time interval that the adjacent captions that record in subtitle file occur, selects several the longest positions of interval as cutting apartPoint, is divided into some chapters and sections in time by whole documentary film. Because this dividing mode is actually according to documentary film commentaryDivide this documentary film reading aloud dead time length while dubbing, thereby these chapters and sections of dividing out possess in terms of contentCertain semantic consistency. At the automatic generation phase of chapters and sections subhead, according to previous stage in definite chapters and sections start-stop moment,Under the guidance of certain selecting rule, in the commentary of these chapters and sections, automatically selecting one, to have " attribute+head " thisThe phrase combination of syntactic structure, as the subhead of these chapters and sections. This invention shortened comparatively significantly manually mode according toThe semantic time that longer documentary film is carried out to segmentation, greatly improve teacher and adopted documentary film fragment perfect in workmanship as teachingThe enthusiasm of material.
Brief description of the drawings
Fig. 1 is the schematic flow sheet in automatic merogenesis stage of content in the present invention.
Fig. 2 is the automatic generation phase schematic flow sheet of subhead in the present invention.
Detailed description of the invention
Below in conjunction with drawings and Examples, the present invention is described in further detail.
The documentary film " home (Home) " of directing taking France director YannArthus-Bertrand, as example, is introduced base in detailIn the detailed description of the invention of the automatic merogenesis of documentary film content of subtitle file and the automatic generation method of subhead thereof.
This documentary film combines by captured all over the world high-quality aerial image, shown on the earth biologicalHow diversity and the mankind's activity endangers the tellurian ecological balance.
What in the present embodiment, adopt is 90 minutes versions of this documentary film. The subtitle file corresponding with this version comprises altogetherArticle 778, captions. Remove the 1st article of captions and provide outside translator's information, other 777 is all that documentary film interpreter aside is usedCommentary. The subtitling format of this subtitle file meets table 1. If run into the subtitle file of extended formatting, can use captionsFile format crossover tool, as Subresync, just can carry out ssa, ass, srt, smi, idx, sub etc. conventional easilyTurning mutually and other caption editing function of subtitle file form, is transferred to the subtitle file consistent with table 1 form.
First enter the automatic merogenesis stage of content. As shown in Figure 1, comprise the following steps:
The first step, every a line of traversal subtitle file, always total N bar captions. If find that its initial character is numeral, saysBright this is the first row of new captions, as shown in table 1. Record its numbering i and extract and start accordingly T nowbegin(i) and disappear moment Tend(i)。
Second step, calculates current captions and starts now and upper captions and disappear time difference in moment:，（i>1）。
The 3rd step, sorted from big to small to (N-1) individual such time difference. Obtain several i that come above. ThisA little i are the position of content merogenesis. Advise that such i is no more than 9, is divided at most 10 chapters and sections.
According to above-mentioned steps, by extract without captions occur 9 the longest positions of the time interval, by Homeland in the timeOn be divided into 10 little chapters and sections. As shown in table 2.
10 little chapters and sections of table 2. Homeland
| Numbering of section | The numbering of initial captions | This chapters and sections zero hours | This chapters and sections finish times |
| 1 | 2 | 00:01:18,268 | 00:02:32,518 |
| 2 | 16 | 00:03:18,184 | 00:08:38,018 |
| 3 | 81 | 00:09:11,393 | 00:14:08,934 |
| 4 | 124 | 00:14:48,393 | 00:17:33,309 |
| 5 | 150 | 00:18:31,226 | 00:35:19,268 |
| 6 | 327 | 00:36:24,309 | 00:55:44,132 |
| 7 | 510 | 00:56:17,215 | 00:57:48,423 |
| 8 | 520 | 00:58:20,340 | 01:01:28,757 |
| 9 | 559 | 01:02:59,798 | 01:04:59,965 |
| 10 | 581 | 01:05:43,257 | 01:24:59,507 |
Then enter the automatic generation phase of subhead. As shown in Figure 2. Comprise the following steps:
The first step, according to the chapters and sections temporal information that the automatic merogenesis of the content stage divides, extracts in the time of this chapters and sections start-stopCaption character content between quarter.
Second step, the caption character content that uses automatic word segmentation instrument to extract the first step carries out participle, and to separatingWord mark part of speech.
The 3rd step, from first to last scans the word that second step separates in order one by one, if find that this word is noun or shapeHold word, and next word is noun, extracts with regard to the phrase that sets it as " nominal endocentric phrase ". These phrases are as these chapters and sectionsCandidate's phrase of subhead.
The 4th step through certain selecting rule, is selected a subhead as these chapters and sections from multiple candidate's phrases.These selecting rules are listed as follows, and the rule that numbering comes above is preferentially used.
(1), if candidate's phrase only has 1, it just directly becomes subhead, and needn't use rule below.
(2) if candidate's phrase more than 1, that is its head just relatively, " noun+()+noun " or " adjective+()+noun " second notional word in structure or the word that finally that part of speech is noun, in these chapters and sections, occur timeNumber. If candidate's phrase that head occurrence number is maximum only has 1, it just becomes subhead so, and after needn't usingRule. If the maximum candidate's phrase of generative center language occurrence number is more than 1, so such candidate's phrase will enter downOne take turns select.
(3) attribute of these candidate's phrases of comparison, i.e. " noun+()+noun " or " adjective+()+noun " structureIn first notional word, namely first noun or " adjective+()+noun " in " noun+()+noun " structureAdjective, the number of times of the appearance in these chapters and sections. If candidate's phrase that attribute head occurrence number is maximum only has 1,It just becomes subhead so, and needn't use rule below. If candidate's phrase that attribute head occurrence number is maximumMore than 1, so such candidate's phrase will enter selecting of next round.
(4) number of words of these candidate's phrases of comparison. If candidate's phrase that number of words is maximum only has 1, it just becomes soSubhead, and needn't use rule below. If the maximum candidate's phrase of number of words is more than 1, so such candidate's phraseSelecting that last takes turns will be entered.
(5) order that relatively these candidate's phrases occur, what first occur just becomes subhead.
In the present invention, each chapters and sections that the automatic generation phase of subhead separates for the automatic merogenesis stage of content automatically successivelyGenerate subhead.
The 4th chapters and sections that now separate taking the automatic merogenesis of the content stage, as example, illustrate what this chapters and sections subhead generated automaticallyProcess. For the caption character content of these chapters and sections, can call automatic word segmentation instrument, as Chinese Academy of Sciences's calculating is developedICTCLAS is partitioned into word and marks its part of speech from character string. The result of participle below:
Life/n the/p earth/n upper/f is /v greatness/a/u adventure/n
Every kind/r biology/n is each/have/v of r its/r position/q, Each performs its own functions for/w/v
Do not have/v is unnecessary/a or/c is harmful/a
They/r mutually/d balance/a
You/r these/r is clever/a /the u mankind/n
/ p this/become/v leading role/n of r
You/r from/4,000,000,000 years/m of the p earth/n /receive benefits/v of u legacy/n
You/r only/obtain/v of d 20/m ten thousand/m/q history/n
But/c you/r change/v/the u world/n /u looks/n
Though have/v of/c weakness/n
But/c you/occupy/v of r/u is all/b habitat/n
Own/b soil/n of conquer/v
Any/r biology/n all/d not /do/v of d mistake/v
Process/p 18/m ten thousand/m/q /u is nomadic/v years/n
Obtain/u of weather/n change/v gentleness/a
Stabilize/get off/v of a of start/v of the mankind/n
They/r no longer/d is with/make a living/v of go hunting/v of p
They/settle down/v of r in/be full of/v of p fishing and hunting/n and/c is wild/b plant/n /u wetland/n
Here/r soil/n ,/w water/n and/c life/n mixing/v
Even if/c is today/t
Major part/m the mankind/n all/d by// u of v seashore/n river bank/s and/c lakeside/n inhabitation/v
They/start/v of r evolution/v
But/c they/r can/v use/v /u energy/n just/c both arms/n and/give/v of c the Nature/n/ u thing/n
This/r is /come/u/u story/n of the thousands of year/m of the v mankind/n
Also/d is /v now/tetra-points of t, the one/m mankind/n /v 1,500,000,000/m people/n/u story/n
This/r number/n is /v affluence/a country/n population/n/u summation/n
The implication of part of speech label symbol is in table 3.
Part of speech label symbol and the implication of word in the above-mentioned word segmentation result of table 3.
According to the automatic product process of subhead, from front to back, scan one by one each word, if find it self and withTwo words of rear appearance, the part of speech of three words is followed successively by "/a " altogether, "/u " and "/n " or "/n ", "/u " and "/n ", i.e. symbolClose the nominal endocentric phrase of " adjective+()+noun " or " noun+()+noun " this form, just set it as the time of subheadSelect phrase. Specifically, the subhead candidate phrase of the 4th chapters and sections is as shown in table 4.
The subhead candidate phrase (before being suitable for selecting rule) of table 4 the 4th chapters and sections
| Candidate's phrase | Rule (1): the number of times that head occurs in these chapters and sections |
| Great adventure | Adventure: 1 |
| The clever mankind | The mankind: 5 |
| The looks in the world | Looks: 1 |
| The wetland of plant | Wetland: 1 |
| People's story | Story: 2 |
| The summation of population | Summation: 1 |
Table 4 also shows above-mentioned candidate's phrase is applied to the result after rule (1), according to this rule, and obviously " clever peopleClass " become the subhead of the 4th chapter.
The rest may be inferred, and whole chapters and sections subheads of this documentary film are as shown in table 5 below.
10 chapter titles of 90 minutes versions of table 5. documentary film Homeland
| Numbering of section | Chapters and sections subhead |
| 1 | The miracle of life |
| 2 | The water of the earth |
| 3 | Indispensable soil |
| 4 | The clever mankind |
| 5 | The hungry mankind 6 --> |
| 6 | Many population |
| 7 | Canadian pitch |
| 8 | The weather of the earth |
| 9 | The ice in Greenland |
| 10 | Poor country |
Claims (1)
1. the automatic merogenesis of documentary film content based on subtitle file and an automatic generation method for subhead thereof, its feature existsIn: the method was made up of the automatic merogenesis of content and subhead two stages of automatic generation,
(1) the automatic merogenesis stage of content, between the time occurring according to two the adjacent captions that record in documentary film subtitle fileEvery size, select the longest several positions, interval as cut-point, whole documentary film is divided into some chapters and sections in time;
The automatic merogenesis stage of described content comprises the following steps:
The first step, every a line of traversal subtitle file, always total N bar captions, if find that its initial character is numeral, illustrates thisBe the first row of new captions, record its numbering i and extraction and start accordingly T nowbegin(i) and disappear the momentTend(i)；
Second step, calculates current captions and starts now and upper captions and disappear time difference in moment,, i is greater than 1 natural number;
The 3rd step, sorted from big to small to (N-1) individual such time difference, obtained several i that come above, these iBe the position of content merogenesis;
(2) the automatic generation phase of subhead, according to the chapters and sections temporal information that the automatic merogenesis of the content stage divides, extracts between thisCaption character content between the chapters and sections start-stop moment, is used have " attribute+head " that in this caption character content, occur to tieThe phrase of structure, generates this joint subhead automatically;
The automatic generation phase of described subhead comprises the following steps:
The first step, the chapters and sections temporal information of dividing according to the automatic merogenesis of the content stage, extract between this chapters and sections start-stop moment itBetween caption character content;
Second step, the caption character content that uses automatic word segmentation instrument to extract the first step is carried out participle, and to the word mark separatingNote part of speech;
The 3rd step, from first to last scans the word that second step separates in order one by one, if find that this word is noun or adjective,And next word is noun, extract with regard to the phrase that sets it as " nominal endocentric phrase ", these phrases are as this chapters and sections small tenonCandidate's phrase of topic;
The 4th step through certain selecting rule, is selected a subhead as these chapters and sections from multiple candidate's phrases, described inSelecting rule is listed as follows, successively:
(A), if candidate's phrase only has 1, it just directly becomes subhead, and needn't use rule below;
(B) if candidate's phrase more than 1, that is its head just relatively, " noun+()+noun " or " adjective+()+Noun " second notional word in structure or the word that finally that part of speech is noun, the number of times occurring in these chapters and sections, ifCandidate's phrase that head occurrence number is maximum only has 1, and it just becomes subhead so, and needn't use rule below,If the maximum candidate's phrase of generative center language occurrence number is more than 1, so such candidate's phrase will enter next ruleSelect;
(C) attribute of these candidate's phrases relatively, in " noun+()+noun " or " adjective+()+noun " structureFirst notional word, namely first noun in " noun+()+noun " structure or the shape of " adjective+()+noun "Hold word, the number of times of the appearance in these chapters and sections, if the maximum candidate's phrase of attribute head occurrence number only has 1, soIt just becomes subhead, and needn't use rule below, if the maximum candidate's phrase of attribute head occurrence number is more than 1Individual, so such candidate's phrase will enter selecting of next rule;
(D) number of words of these candidate's phrases of comparison, if the maximum candidate's phrase of number of words only has 1, it just becomes small tenon soTopic, and needn't use rule below, if the maximum candidate's phrase of number of words more than 1, so such candidate's phrase will enterEnter selecting of last rule;
(E) order that relatively these candidate's phrases occur, what first occur just becomes subhead.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310417087.8A CN103473220B (en) | 2013-09-13 | 2013-09-13 | The automatic merogenesis of documentary film content based on subtitle file and the automatic generation method of subhead thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310417087.8A CN103473220B (en) | 2013-09-13 | 2013-09-13 | The automatic merogenesis of documentary film content based on subtitle file and the automatic generation method of subhead thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103473220A CN103473220A (en) | 2013-12-25 |
| CN103473220B true CN103473220B (en) | 2016-05-18 |
Family
ID=49798079
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310417087.8A Active CN103473220B (en) | 2013-09-13 | 2013-09-13 | The automatic merogenesis of documentary film content based on subtitle file and the automatic generation method of subhead thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103473220B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108600512A (en) * | 2018-03-27 | 2018-09-28 | 平安科技（深圳）有限公司 | Information cuing method, electronic device, terminal device and storage medium |
| CN110532391B (en) * | 2019-08-30 | 2022-07-05 | 网宿科技股份有限公司 | Text part-of-speech tagging method and device |
| CN113660432A (en) * | 2021-08-17 | 2021-11-16 | 安徽听见科技有限公司 | Translation subtitle production method and device, electronic equipment and storage medium |
| CN114697760B (en) | 2022-04-07 | 2023-12-19 | 脸萌有限公司 | Processing method, processing device, electronic equipment and medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101127870A (en) * | 2007-09-13 | 2008-02-20 | 深圳市融合视讯科技有限公司 | A creation and use method for video stream media bookmark |
| CN101539929A (en) * | 2009-04-17 | 2009-09-23 | 无锡天脉聚源传媒科技有限公司 | Method for indexing TV news by utilizing computer system |
| CN101650958A (en) * | 2009-07-23 | 2010-02-17 | 中国科学院声学研究所 | Extraction method and index establishment method of movie video scene clip |
| CN102724598A (en) * | 2011-12-05 | 2012-10-10 | 新奥特（北京）视频技术有限公司 | Method for splitting news items |
| CN102833638A (en) * | 2012-07-26 | 2012-12-19 | 北京数视宇通技术有限公司 | Automatic video segmentation and annotation method and system based on caption information |
-
2013
- 2013-09-13 CN CN201310417087.8A patent/CN103473220B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101127870A (en) * | 2007-09-13 | 2008-02-20 | 深圳市融合视讯科技有限公司 | A creation and use method for video stream media bookmark |
| CN101539929A (en) * | 2009-04-17 | 2009-09-23 | 无锡天脉聚源传媒科技有限公司 | Method for indexing TV news by utilizing computer system |
| CN101650958A (en) * | 2009-07-23 | 2010-02-17 | 中国科学院声学研究所 | Extraction method and index establishment method of movie video scene clip |
| CN102724598A (en) * | 2011-12-05 | 2012-10-10 | 新奥特（北京）视频技术有限公司 | Method for splitting news items |
| CN102833638A (en) * | 2012-07-26 | 2012-12-19 | 北京数视宇通技术有限公司 | Automatic video segmentation and annotation method and system based on caption information |
Non-Patent Citations (2)
| Title |
|---|
| 公文标题结构范例赏析(中)——《公文标题结构规范系列谈》之八;栾照钧;《办公室业务》;20090331(第3期);第27页 * |
| 新闻视频的场景分段索引及摘要生成;姜帆,章毓晋;《计算机学报》;20030731;第26卷(第7期);第859-865页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103473220A (en) | 2013-12-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101650958B (en) | Extraction method and index establishment method of movie video scene fragment | |
| Franco et al. | Voice-over translation: An overview | |
| CN103559214B (en) | Method and device for automatically generating video | |
| CN111538851B (en) | Method, system, equipment and storage medium for automatically generating demonstration video | |
| CN104915433A (en) | Method for searching for film and television video | |
| CN103473220B (en) | The automatic merogenesis of documentary film content based on subtitle file and the automatic generation method of subhead thereof | |
| CN105096932A (en) | Voice synthesis method and apparatus of talking book | |
| CN106340291A (en) | Bilingual subtitle production method and system | |
| CN101105943A (en) | Language aided expression system and its method | |
| CN103544978A (en) | Multimedia file manufacturing and playing method and intelligent terminal | |
| CN109672832A (en) | The processing method of digital movie interlude lyrics subtitle realization dynamic Special display effect | |
| Toynbee et al. | World music at the BBC World Service, 1942-2008: public diplomacy, cosmopolitanism, contradiction | |
| CN101777270A (en) | Language learning method based on scenes of common film and television works | |
| TW200411433A (en) | Method and recording medium for creating film search database | |
| Luo | Cultural Tourism and Cantonese Opera | |
| CN114218413A (en) | Background system for video playing and video editing | |
| CN201229777Y (en) | DVD playing device simultaneously displaying subtitles of multiple languages | |
| Zähres | Broadcasting your variety | |
| Rolston | Inscribing Jingju/Peking Opera: Textualization and Performance, Authorship and Censorship of the “National Drama” of China from the Late Qing to the Present | |
| Hung | Tan Dun Through the Lens of Western Media (Part I) | |
| Okyayuz et al. | The Turkish take on audiovisual translation: A story of politics and evolution | |
| Lacković | Audiovisual Translation: Challenges of Traslating Humor from English into Croatian in the American Sitcom" The Office" | |
| Sarma | Interrogating the Politics of Folk | |
| GUNN | GRANT LEIGH SAUNDERS | |
| Selvik et al. | Relevance and Marginalisation in Scandinavian and European Performing Arts 1770–1860: Questioning Canons |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |