CN101646050B - Text annotation method and system, playing method and system of video files - Google Patents
Text annotation method and system, playing method and system of video files Download PDFInfo
- Publication number
- CN101646050B CN101646050B CN2009101707358A CN200910170735A CN101646050B CN 101646050 B CN101646050 B CN 101646050B CN 2009101707358 A CN2009101707358 A CN 2009101707358A CN 200910170735 A CN200910170735 A CN 200910170735A CN 101646050 B CN101646050 B CN 101646050B
- Authority
- CN
- China
- Prior art keywords
- mark
- frame
- video
- timestamp
- video file
- 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
Abstract
The invention discloses a text annotation method and a playing method of video files and corresponding systems thereof, wherein, the text annotation method of the video files comprises the step of carrying out annotation by lens on the video files by utilizing the information of the script mark number and the script mark sheet and the like in film and television pieces. The quick search of the video contents can be realized by searching for the self defining keywords in the text annotation information. Based on the search of text annotation information, the playing method of video files realizes that users can acquire corresponding contents in the video according to individual preference and requirements; specific contents in the video files are played by reorganizing corresponding contents, thus providing viewing experience of 'self defining' playing video for users.
Description
Technical field
The present invention relates to video processing technique, relate in particular to a kind of text annotation method and system of video file, and the player method of video file and system.
Background technology
Along with the number of present internet video data and the growth of total duration, people begin concern video file content and semanteme are described, and describe and definition as the standardization that all relates to content of multimedia in mpeg-4 and the mpeg-7 standard.Such description can not only help user search to arrive the desired video file, also can strengthen the interactive and flexibility of multimedia system.Yet, video playback mode in the past, the content of accepting video file that the user can only be passive if drag with the player slide block, not only wastes time and energy, and misses easily and omit and oneself want the content of watching.Present video frequency searching is unit mostly with the video file, is that 45 minutes video file is an example with duration, and the user may not be ready that the time cost of spending 45 minutes obtains 5 minutes the content of oneself wanting may had only.
The note of video file is helped retrieval to video file, the main at present manual annotation mode that adopts usually.Yet, pure artificial mode consumption human capital not only consuming time, but also the difference of understanding owing to the individual is easily misled.
In the prior art, the broadcast mode of accepting video file content that the user usually can only be passive, and if drag with the player slide block, not only waste time and energy, and miss easily and omit and oneself want the content of watching.
Therefore, need a kind of effectively to the text annotation method of video file to substitute existing manual annotation mode.In addition, thus how to realize that it is the target that industry is constantly pursued that quick retrieval to video file makes the user obtain more friendly video tastes always.
Summary of the invention
A kind of method of effectively video file being carried out text annotation that provides is provided technical problem to be solved by this invention, and realize " self-defined " broadcast mode of video file, thereby provide better viewing experience for the user based on retrieval to text annotated information.The invention provides following technical scheme:
One aspect of the present invention provides a kind of text annotation method of video file, and this method comprises: the frame of retrieve video file is also determined script holder's frame; From script holder's frame, extract a mark, obtain the annotation information of the corresponding video segment of a mark according to the field mark; Obtain the timestamp of the corresponding video segment of a mark; The timestamp and the annotation information of the corresponding video segment of a record mark.
Among the embodiment of the text annotation method of video file provided by the invention, it further comprises according to the step that the field mark obtains the annotation information of the corresponding video segment of a mark: obtain the single and/or lens-dividing script information of the script holder corresponding with a mark according to the field mark, as the annotation information of the corresponding video segment of field mark.
Among the embodiment of the text annotation method of video file provided by the invention, the frame of its retrieve video file determines that the step of script holder's frame comprises: retrieve video file frame by frame, determine script holder's frame according to the grey level histogram information of each frame; The step of extracting a mark from script holder's frame comprises: extract a mark based on bar-code identification method or digit recognition method from script holder's frame.
Among the embodiment of the text annotation method of video file provided by the invention, its step that obtains the timestamp of the corresponding video segment of a mark comprises: the timestamp that obtains the corresponding video segment of a mark according to the quantity of FPS and frame by retrieval; The timestamp of the corresponding video segment of script holder's frame back court mark is removed in the number of times correction that has occurred according to script holder's frame.
Among the embodiment of the text annotation method of video file provided by the invention, this method also comprises step: after video file is sheared, and the corresponding timestamp of revising the video segment that is not sheared.
Another aspect of the present invention provides a kind of text annotation system of video file, and text annotation system comprises: script holder's frame judgment means is used for the frame of retrieve video file and determines script holder's frame; The field mark extracts and the annotation information deriving means, is used for extracting a mark from script holder's frame that script holder's frame judgment means is determined, according to the annotation information of an acquisition of field mark and a corresponding video segment of mark; The timestamp deriving means is used to obtain to extract with a mark and the timestamp of the corresponding video segment of field mark that the annotation information deriving means extracts; Tape deck is used for the timestamp of logging timestamp deriving means acquisition and the annotation information of being extracted by extraction of field mark and annotation information deriving means.
Among the embodiment of the text annotation system of video file provided by the invention, its mark extracts and the annotation information deriving means obtains the single and/or lens-dividing script information of the script holder corresponding with a mark according to the field mark, as the annotation information of the corresponding video segment of field mark.
Among the embodiment of the text annotation system of video file provided by the invention, its script holder's frame judgment means is the retrieve video file frame by frame, determines script holder's frame according to the grey level histogram information of each frame; The field mark extracts and the annotation information deriving means comprises: a mark extraction unit is used for extracting a mark based on bar-code identification method or digit recognition method from script holder's frame a transmission mark; And the annotation information deriving means, be used to receive the field mark that extracts from field mark extraction unit, and obtain and an annotation information of the corresponding video segment of mark according to the field mark.
Among the embodiment of the text annotation system of video file provided by the invention, its timestamp deriving means comprises: the timestamp acquiring unit is used for obtaining a timestamp of mark correspondence video segment according to the quantity of FPS and frame by retrieval; Send the timestamp of the corresponding video segment of a mark; And the timestamp amending unit, be used to receive the timestamp of the corresponding video segment of field mark that obtains from the timestamp acquiring unit, and remove the timestamp of the corresponding video segment of field mark behind script holder's frame according to the number of times correction that script holder's frame has occurred.
Among the embodiment of the text annotation system of video file provided by the invention, text annotation system comprises the update of time stamp device, after being used for video file sheared, and the corresponding timestamp of revising the video segment that is not sheared.
Another aspect of the present invention provides a kind of player method of video file, wherein video file comprises a plurality of video segments with annotation information and timestamp, this player method comprises: receive the keyword input, annotation information is retrieved the timestamp of the video segment of acquisition and keyword coupling according to keyword; Video content according to the corresponding time period in the timestamp playing video file of the video segment that mates.
Among the embodiment of the player method of video file provided by the invention, this method is finished according to the retrieval of keyword to annotation information at user side; The video content of the timestamp time corresponding section of the video segment that user side mates with keyword from server end foradownloaded video file.
Among the embodiment of the player method of video file provided by the invention, its video segment is corresponding to the scene/camera lens of video file, and the annotation information of video segment comprises the single and/or lens-dividing script information of the script holder of scene/camera lens.Wherein, annotation information comprises role, performer, action, scene and/or plot.
Another aspect of the present invention provides a kind of Play System of video file, wherein, video file has a plurality of video segments of annotation information and timestamp, this Play System comprises: key search and timestamp deriving means, be used to receive the keyword input, annotation information is retrieved the timestamp of the video segment of acquisition and keyword coupling according to keyword; Media playing apparatus is used for the video content of basis by the corresponding time period of timestamp playing video file of the video segment of the coupling of key search and the acquisition of timestamp deriving means.
Among the embodiment of the Play System of video file provided by the invention, this Play System further comprises: the annotation information download apparatus is used for the annotation information of video file is downloaded to user side; Key search and timestamp deriving means are finished according to the retrieval of keyword to annotation information at user side; The video segment download apparatus is used at user side from the video content of server end foradownloaded video file with the timestamp time corresponding section of the video segment of keyword coupling.
Among the embodiment of the Play System of video file provided by the invention, its video segment is corresponding to the scene/camera lens of video file, and the annotation information of video segment comprises the single and/or lens-dividing script information of the script holder of scene/camera lens.Wherein, annotation information comprises role, performer, action, scene and/or plot.
The present invention is based on contents such as script holder's frame, mark, script holder's list, script video file is pursued the text annotation of camera lens, provide a kind of and effectively video content has been carried out note ground method.Retrieve according to user-defined keyword, only play the associated scenario that keyword hits, thereby the user retrieving the video broadcasting method of view content by the personal like, is that the user provides better viewing experience with the broadcast mode of this personalization.
Description of drawings
Fig. 1 is the schematic diagram of the information of clap-stick and correspondence thereof;
Each video segment broadcast start time stabbed T (i) and corresponding relation thereof after Fig. 2 illustrated the broadcast start time stamp t (i) of each script holder's frame and removes script holder's frame;
Fig. 3 illustrates the flow chart according to an embodiment of the text annotation method of video file of the present invention;
Fig. 4 illustrates the flow chart according to another embodiment of the text annotation method of video file of the present invention;
Fig. 5 illustrates the structural representation according to an embodiment of the text annotation system of video file of the present invention;
Fig. 6 illustrates the structural representation according to another embodiment of the text annotation system of video file of the present invention;
Fig. 7 illustrates the structural representation according to another embodiment of the text annotation system of video file of the present invention;
Fig. 8 illustrates the schematic flow sheet according to a specific embodiment of the text annotation method of video file of the present invention;
Fig. 9 illustrates the flow chart according to an embodiment of the player method of video file of the present invention;
Figure 10 illustrates the flow chart according to another embodiment of the player method of video file of the present invention;
Figure 11 illustrates the structural representation according to an embodiment of the Play System of video file of the present invention;
Figure 12 illustrates the structural representation according to another embodiment of the Play System of video file of the present invention;
Figure 13 illustrates a specific embodiment according to the player method of video file of the present invention;
Figure 14 (a) is common stream transmission process schematic diagram;
Figure 14 (b) illustrates the stream transmission realization flow schematic diagram of the player method of video file of the present invention.
The term explanation:
Script holder's frame: the frame that will comprise clap-stick information is called script holder's frame.
Mark: for for simplicity, with reel number, number, mirror number three field informations are collectively referred to as a mark; I+1 mark is expressed as x (i).
Timestamp: before removing script holder's frame, the broadcast start time stamp of script holder's frame is t (i); Remove after script holder's frame, each video segment broadcast start time stamp is T (i).For the purpose of accurately, can select SMPTE (the Society of Motion Picture andTelevision Engineers) time for use.
Keyword d (i): the annotation information that from store database of information such as script holder's list, lens-dividing script, checks in according to field mark x (i).
Counter: be used to add up the number of the key frame that retrieves, the number of script holder's frame for example, quantity is represented with N (i).
Number of pictures per second (Frame Per Second, FPS): calculate the influence of script holder's frame of extracting respective numbers according to FPS to time parameter.
Head/run-out: the part before first script holder's frame is called head; Part after the content that last frame guided is called run-out.
Embodiment
With reference to the accompanying drawings the present invention is described more fully, the various embodiment that the present invention is exemplary are described.
Films and television programs can write down the sequence number of the camera lens of being clapped by clap-stick when taking, and clap-stick are taken into picture before starting shooting at every turn.Fig. 1 is the schematic diagram of clap-stick and corresponding information thereof, shown in clap-stick comprise following information: reel number (ROLL), number (SCENE), mirror number (SHOT), number of times (TAKE), director (DIRETOR), photographer (CAMERAMAN), production unit (PRODUCTION), date (DATE) and other additional informations etc.Wherein, according to reel number, number, numbers three fields of mirror can orient the essential information of current shooting on script holder's list, as the performer of participation, role, scene, incident etc.In post-production, the frame that includes clap-stick information can be sheared at all and just be removed after editing is finished.Every partial video fragment broadcast start time stabbed T (i) after Fig. 2 showed the broadcast start time stamp t (i) of each script holder's frame and removes script holder's frame, and corresponding relation; The relative time that is if having time is 0 point with the video playback starting point, is unit with the second.After removing script holder's frame, leader is added in first video segment usually; The sheet portion is added in last video segment.Suppose to have from 0 to n mark, video file will have n+1 video segment so.
It is semantic to consider that note to video file inside is based on, and prior art mainly adopts the manual annotation mode at present.Yet, pure artificial mode consumption human capital not only consuming time, but also because the difference that the individual understands is easy to generate misleading.Therefore, the present invention is directed to films and television programs, proposed a kind of new annotate method; Promptly, can conveniently finish note work, not need the note personnel to watch and understand the content of whole video file by materials such as script, a mark, script holder's lists.
Fig. 3 illustrates the flow chart according to an embodiment of the text annotation method of video file of the present invention.
As shown in Figure 3, step 302, the frame of retrieve video file is also determined script holder's frame.For example, the retrieve video file needs to judge whether the frame in the video file is script holder's frame in the process of retrieval frame by frame.Wherein " judge whether frame ", can carry out with manual type into the script holder; Also can take automated manner to carry out.Script holder's frame has obvious characteristics (for example Du Te texture, color) with respect to captured picture, can judge by the histogram that compares each frame.Specifically, the automatic judging method that " judges whether the frame for the script holder " comprises: the grey level histogram of getting each frame picture, because the captured content of script holder's frame is black and white clap-stick, its grey level histogram can have double-peak feature clearly, exists with gray scale to locate two peaks into the center near locating for 0 (deceive) with gray scale close 255 (in vain).Picture with this feature can be differentiated the frame for the script holder.
The invention provides among the embodiment of text annotation method of video file, the annotation information that keyword d (i) comprises can be subdivided into again: role, performer, action, scene, plot or other similar information.
Following mask body is introduced two kinds of field marks implementation method of identification automatically:
A, bar-code identification method.The basic thought of this method is that the information that keeps on original clap-stick is constant, but the ad-hoc location on clap-stick is stamped and a corresponding bar code of mark.When reading this frame picture, at first find the position of this bar code; Secondly, according to the width and the spacing of black line in the bar code that is read and white line bar, correspond to corresponding digital (mark).If in the process of taking, bar code and video camera and out of plumb, deflect, then can be solved (referring to one dimension and the research of two-dimensional barcode image recognition system by the geometric transformation of digital picture, Fan Yongfa, He Hongzhuan, " Chinese journal of scientific instrument ", S2 phase in 2003, Electromechanical Engineering College, Hehai Univ.).
Adopt the bar-code identification method, from the angle of field mark identification, technology realize going up comparative maturity, easy, error rate is lower.
B, digit recognition method.The basic thought of this method is that the information and the using method that keep on original clap-stick are constant, by hand-written Arabic numerals recognition technology the field mark is discerned.Concrete steps are as follows:
At first, source images is carried out the black and white binary conversion treatment.
Secondly, the image after the first step is handled carries out the opening operation (earlier operation that image carried out erosion operation and then expand) of morphological images in handling, and then carry out closed operation (earlier image being carried out the operation of dilation operation and then erosion operation), reach the effect of removing the background noise and filling up the crack in the handwritten numeral.
Three, utilizing (FloodFill) algorithm that floods, is that partly fill the black that is communicated with on the border with white.Because Arabic numerals 0-9 is the character of connection, do not exist as
The disconnected character of this class can be distinguished single Arabic numerals with connectedness.When end-of-fill, resulting fill area is exactly the image-region of certain hand-written Arabic numerals among the figure, at this moment, obtains the right set of pixel coordinate point in this Arabic numerals image-region.
Four, the right set of certain digital pixel coordinate point that obtains by the 3rd step, by the skeletal extraction algorithm (referring to work such as R.C.Gonzales, Ruan Qiuqi etc. translate, " Digital Image Processing ", the Electronic Industry Press, 2002) extract the skeleton of handwritten numeral, again its skeleton image is mapped to certain 0/1 matrix of 8 * 8, the matrix intermediate value is that 1 point is represented the point in the matrix of handwritten numeral skeleton process, 0 numeral background, requiring each row of digital picture or each row, all to have a value at least be 1.
Five, [0/1] matrix of 8 * 8 is carried out content analysis.The method that can take has: template matches, rule judgement etc.
Adopt the digit recognition method, do not require that the using method to original image pickup method and clap-stick has any modification.
Fig. 4 illustrates the flow chart according to another embodiment of the text annotation method of video file of the present invention.As shown in Figure 4, wherein step 402 can be identical with step 302 shown in Figure 3 and step 304 with step 404, for for purpose of brevity, repeats no more here.
Next execution in step 406, obtain the timestamp t (i) of the corresponding video segment of a mark according to the quantity of FPS and frame by retrieval.This step can be identical with step 306 shown in Figure 3.
Be with the obvious difference of text annotation method flow shown in Figure 3: as shown in Figure 4, after the execution in step 406, with execution in step 408, the timestamp of the corresponding video segment of script holder's frame back court mark is removed in the number of times correction that has occurred according to script holder's frame.When forming the video file that can play continuously, script holder's frame of this video file can in the end be removed, and therefore need revise with respect to the timestamp of file starting point present frame.In view of the FPS of video fixes, with script holder's frame number of times that counters count retrieves, before can calculating script holder's frame and being removed and the time difference after removing.Promptly utilize the timestamp amending unit to converse to remove time stamp T (i)=t (i)-(N/FPS)-t (0), wherein N=i+1 that the video segment of correspondence begins to play behind i+1 script holder's frame.Among the embodiment of the text annotation method of video file provided by the invention, time stamp T (0) zero setting that corresponding video segment begins to play behind the 1st script holder's frame will be removed.
Execution in step 410 in regular turn, write down the timestamp and the annotation information of the corresponding video segment of this mark.For example, note respectively according to abovementioned steps 404 and obtain obtaining time stamp T (i) with the corresponding video segment of this mark with the keyword d (i) of the corresponding video segment of this mark with according to abovementioned steps 408; Thereby generate the annotation information table (seeing table 1) with time stamp T (i) pairing of this video segment by keyword d (i).
Table 1
Among the embodiment of the text annotation method of video file provided by the invention, this video annotation method also comprises step: after video file is sheared, and the corresponding timestamp of revising the video segment that is not sheared.The division of video segment is foundation with time, if video content can influence the validity of annotation information through shearing.The shearing that video content may pass through has following several: head are sheared, (in general the middle part content need not editor after post-production is finished, and is ordered shearing owing to comprise responsive camera lens if do shearing again) sheared at the middle part, run-out is sheared.Wherein, run-out is sheared can not influence annotation information, does not therefore do consideration.Head are sheared (comprising that also head length increases) and shear at the middle part, if operated by the side of makinging, only the time parameter in the annotation information need be done synchronous upgrading and get final product.
Fig. 5 illustrates the structural representation according to an embodiment of the text annotation system of video file of the present invention.As shown in Figure 5, the text annotation system 50 of video file comprises that script holder's frame judgment means 502, a mark extract and annotation information deriving means 504, timestamp deriving means 506 and tape deck 508.Wherein, 502 pairs of video files of script holder's frame judgment means are retrieved frame by frame, and whether judgment frame is script holder's frame; The method of determining script holder's frame adopts as preceding to utilize the captured content of script holder's frame be black and white clap-stick, has tangible grey level histogram feature, determines script holder's frame according to the grey level histogram information of each frame.After definite certain frame was script holder's frame, a mark extracted and annotation information deriving means 504 is carried out bar-code identification method or digit recognition method, extracts a mark from script holder's frame, obtains the annotation information of the corresponding video segment of a mark according to the field mark.For example, obtain the single and/or lens-dividing script information of the script holder corresponding according to the field mark, as the annotation information of the corresponding video segment of field mark with a mark.Timestamp deriving means 506 obtains the timestamp of the corresponding video segment of a mark subsequently, for example obtains the timestamp t (i) of the corresponding video segment of a mark according to the quantity of FPS and frame by retrieval.At last, tape deck 508 writes down the timestamp and the annotation information of the corresponding video segments of a mark, thereby generates the annotation information table with the timestamp pairing of corresponding video segment by annotation information keyword d (i).
Fig. 6 illustrates the structural representation according to another embodiment of the text annotation system of video file of the present invention.As shown in Figure 6, script holder's frame judgment means 602, timestamp deriving means 606 and tape deck 608 in the text annotation system 60 of video file can be identical with corresponding module shown in Figure 5, for for purpose of brevity, repeat no more here.Be with text annotation system difference shown in Figure 5: as shown in Figure 6, a mark extracts and annotation information deriving means 604 comprises a mark extraction unit 6041 and annotation information acquiring unit 6042; Wherein a mark extraction unit 6041 is carried out bar-code identification method or digit recognition method, extracts a mark from script holder's frame; Annotation information acquiring unit 6042 obtains and an annotation information of the corresponding video segment of mark according to the field mark.For example, obtain the single and/or lens-dividing script information of the script holder corresponding according to the field mark, as the annotation information of the corresponding video segment of field mark with a mark.
Fig. 7 illustrates the structural representation according to another embodiment of the text annotation system of video file of the present invention.As shown in Figure 7, the script holder's frame judgment means 702 in the text annotation system 70 of video file, a mark extract and annotation information deriving means 704 can be identical with corresponding module shown in Figure 5 with tape deck 708, for for purpose of brevity, repeat no more here.Be with text annotation system difference shown in Figure 5: as shown in Figure 7, timestamp deriving means 706 specifically comprises timestamp acquiring unit 7061 and timestamp amending unit 7062; Wherein the timestamp acquiring unit 7061, are used for obtaining according to the quantity of FPS and frame by retrieval the timestamp of the corresponding video segment of a mark; Timestamp amending unit 7062, the timestamp of the corresponding video segment of script holder's frame back court mark is removed in the number of times correction that is used for having occurred according to script holder's frame.In view of the FPS of video fixes, with script holder's frame number of times that counters count retrieves, before can calculating script holder's frame and removing and the time difference after removing.Promptly utilize the timestamp amending unit to converse to remove time stamp T (i)=t (i)-(N/FPS)-t (0), wherein N=i+1 that the video segment of correspondence begins to play behind i+1 script holder's frame.Among the embodiment of the text annotation system of video file provided by the invention, time stamp T (0) zero setting that corresponding video segment behind the 1st script holder's frame of removal can be begun to play.
Among another embodiment of the text annotation system of video file of the present invention, the text annotation system 50,60 of video file, 70 except comprising that script holder's frame judgment means 502,602,702, a mark extract and annotation information deriving means 504,604,704, timestamp deriving means 506,606,706 and tape deck 508, outside 608,708, also comprise the update of time stamp device, be used for after video file is sheared the corresponding timestamp of revising the video segment that is not sheared.
Introduce a specific embodiment of the text annotation method of video file of the present invention below.
Fig. 8 illustrates the schematic flow sheet according to a specific embodiment of the text annotation method of video file of the present invention.As shown in Figure 8, step 802 is at first carried out the initialization of the text annotation system of video file, comprises counter O reset, each variable zero setting, and FPS (as per second 24 frames, per second 30 frames etc.) is set according to video format.
Next step 804 begins retrieve video file frame by frame.
Step 806 adopts aforesaid automatic identification method, judges by script holder's frame judgment means 502,602,702 whether the frame in the video file is script holder's frame.In step 806, if be confirmed to be the 1st script holder's frame, then execution in step 808.
In step 808, field mark extraction unit 6041 is used for " an identification mark " to obtain the 1st field mark, and be x (0) with the 1st script holder's labelled notation, according to the field mark x (0) that is provided with, annotation information acquiring unit 6042 is searched in database to obtain the information that correspondent keyword d (0) describes this video segment; Timestamp acquiring unit 7061 obtains the timestamp of the corresponding video segment of this mark according to the quantity of FPS and frame by retrieval, i.e. the 1st script holder's frame broadcast start time t (0), timestamp amending unit 7062 will be removed time stamp T (0) zero setting that the video segment of the 1st script holder's frame back court mark mark x (0) correspondence begins to play simultaneously.Tape deck 508,608,708 note a time stamp T (0) of the corresponding video segment of mark x (0) and annotation information " keyword d (0) ".
Next,, carry out " counter+1 ", continue execution in step 812 with retrieve video file frame by frame in step 810.
In step 814, the frame that is continued to carry out in the video file by script holder's frame judgment means 502,602,702 is not script holder's frame.
If in step 814, be confirmed to be i+1 script holder's frame, and judge that in step 816 i+1 the mark that is obtained by field mark extraction unit 6041 is new; Execution in step 818 so, are x (i) with i+1 script holder's labelled notation in regular turn, and according to the field mark x (i) of mark, annotation information acquiring unit 6042 is searched in database to obtain the information that correspondent keyword d (i) describes this video segment; Timestamp acquiring unit 7061 obtains the timestamp of the corresponding video segment of a mark according to the quantity of FPS and frame by retrieval, i.e. i+1 script holder's frame broadcast start time t (i), time stamp T (i)=t (i)-(N/FPS)-t (0), the wherein N=i+1 of the corresponding video segment of script holder's frame back court mark removed in the number of times correction that occurred according to script holder's frame of timestamp amending unit 7062 subsequently; And by tape deck 508,608,708 note a time stamp T (i) of the corresponding video segment of mark x (i) and annotation information " keyword d (i) ".
Execution in step 820 " counter+1 " then, redirect execution in step again 812, the circulation abovementioned steps is with retrieve video file frame by frame.
If in step 816, judge that i+1 the mark that is obtained by field mark extraction unit 6041 is not new, then jump to step 820, carry out " counter+1 ".Continue execution in step 812, the circulation abovementioned steps is with retrieve video file frame by frame.
The circulation abovementioned steps is not script holder's frame with retrieve video file frame by frame up to the frame of judging video file in step 814, and judges that in step 822 present frame be the last frame of this video file, then to the text annotation flow process end of this video file.
The text annotated information that preceding method obtained can be used for the quick searching system to video file, and the user can import the keyword relevant with own desired video in search box, thereby realizes the quick retrieval to video file.Keyword can comprise different information description classifications, also can search in full in whole comment file.A plurality of keywords can allow certain logical relation, and for example " performer A+ performer B " expression search has all video segments of " performer A " or " performer B " appearance.Among the embodiment of the text annotation method of video file provided by the invention, for realizing the quick retrieval of video file, the user can adjust choosing of keyword according to the result of first retrieval, repeatedly retrieve, in the hope of reaching more comprehensively, obtain exactly needed video file.
Fig. 9 illustrates the flow chart according to an embodiment of the player method of video file of the present invention.The file annotation information of video file and video file association store are in media server.At this moment, the retrieval that comment file is carried out is to finish at the Web server end.Video file comprises a plurality of video segments with annotation information and timestamp among this embodiment, and wherein timestamp and annotation information can be the annotate method acquisitions according to aforementioned any one video file.
As shown in Figure 9, step 902 receives the keyword input, annotation information is retrieved the timestamp of the video segment of acquisition and keyword coupling according to keyword.For example the Web server specialized page that will contain search box is pushed to the user, and the user can import the keyword relevant with own desired video in search box.Web browser will be retrieved correspondent keyword according to keyword in the annotation information table, the timestamp of the video segment of acquisition and keyword coupling.
Step 904 is according to the video content of corresponding time period in the timestamp playing video file of the video segment that mates.For example Web browser with search hit, send to media player with the timestamp of the video segment of keyword coupling; The result that media player will hit is converted into the RTSP request, thereby reproduction time stabs the video content of corresponding time period successively.
Figure 10 illustrates the flow chart according to another embodiment of the player method of video file of the present invention.The annotation information that the present invention adopts the text annotation method of aforementioned video file to obtain can be stored on the Web server by the XML form, and related with the video file itself on having Web server.Annotation information is stored in Web server with the XML form, and key search can be finished at user side, and need not to carry out key search at server end.
As shown in figure 10, step 1002, Web browser downloads to user side with text annotated information, for example, Web browser downloads to local user side with the annotation information table from Web server earlier, the high speed processing performance of utilizing the local user to hold, thus convenient user subsequently finishes key search fast on the user side of this locality.
Step 1004 receives the keyword input, annotation information is retrieved the timestamp of the video segment of acquisition and keyword coupling according to keyword; For example the local user holds the specialized page that will contain search box to be pushed to the user, and the user can import the keyword relevant with own desired video in search box.User side will be retrieved correspondent keyword according to keyword in the text annotated information of having downloaded, the timestamp of the video segment of acquisition and keyword coupling.Thereby finish according to of the retrieval of key word annotation information at user side.
In step 1006, the video content of the timestamp time corresponding section of the video segment that user side mates with keyword from Web server foradownloaded video file earlier,
Execution in step 1008, media player are play-overed the video content of corresponding time period in the video file.
Among another embodiment of the player method of video file of the present invention, if key search is miss, then whether the user can select to watch according to the common video broadcast mode, if select "Yes", then Web server propelling movement ordinary pages is carried out the normal video broadcast.
Among another embodiment of the player method of video file of the present invention, if directly utilize the comment file of the plug-in subtitle file of SUB form as corresponding video, subtitle file and video file can be existed on the media server together, associate with same video numbering.With cut apart based on the video of field mark different, based on cutting apart of subtitle file video can be cut apart with too small granularity, be unfavorable for watching continuously.But for being the spectators of main purpose with the inquiry, this note also has certain meaning.
Figure 11 illustrates the structural representation according to an embodiment of the Play System of video file of the present invention.Video file comprises a plurality of video segments with annotation information and timestamp among this embodiment, and wherein timestamp and annotation information can be the annotate method acquisitions according to aforementioned any one video file.As shown in figure 11, the Play System 110 of video file comprises key search and timestamp deriving means 1102 and video play device 1104.Wherein, key search and timestamp deriving means 1102 are used to receive the keyword input, annotation information are retrieved the timestamp of the video segment of acquisition and keyword coupling according to keyword; For example the Web server specialized page that will contain search box is pushed to the user, and the user can import the keyword relevant with own desired video in search box.Web browser will be retrieved correspondent keyword according to keyword in the annotation information table, the timestamp of the video segment of acquisition and keyword coupling.Video play device 1104 is used for the video content according to the corresponding time period of timestamp playing video file of the video segment that mates.For example Web browser with search hit, send to media player with the timestamp of the video segment of keyword coupling; The result that media player will hit is converted into the RTSP request, thereby reproduction time stabs the video content of corresponding time period successively.
Figure 12 illustrates the structural representation according to another embodiment of the Play System of video file of the present invention.The annotation information that the present invention adopts the text annotation method of aforementioned video file to obtain can be stored on the Web server by the XML form, and related with the video file itself on having Web server.Annotation information is stored in Web server with the XML form, and key search can be finished at user side, and need not to carry out key search at server end.As shown in figure 12, the Play System 120 of video file comprises annotation information download apparatus 1202, key search and timestamp deriving means 1204, video segment download apparatus 1206 and media playing apparatus 1208.Wherein, annotation information download apparatus 1202 is used for the annotation information of video file is downloaded to user side.For example, Web browser earlier downloads to local user side with annotation information from Web server, the high speed processing performance of utilizing the local user to hold computer, thus convenient user subsequently finishes key search fast on the user side of this locality.Key search and timestamp deriving means 1204 are used to receive the keyword input, finish according to keyword at user side annotation information is retrieved, the timestamp of the video segment of acquisition and keyword coupling.For example the local computer user side specialized page that will contain search box is pushed to the user, and the user can import the keyword relevant with own desired video in search box.User side will be retrieved correspondent keyword according to keyword in the annotation information of having downloaded, the timestamp of the video segment of acquisition and keyword coupling.Thereby finish according to the retrieval of keyword annotation information at user side.Video segment download apparatus 1206 is used for the video content of timestamp time corresponding section of the video segment of foradownloaded video file and keyword coupling.For example, the timestamp of the video segment of and keyword coupling that obtained according to key search and timestamp deriving means 1204, the local user holds computer to download video content with the timestamp time corresponding section of the video segment of keyword coupling from Web server.Media playing apparatus 1208 is used for play-overing the video content of the video file time corresponding section that video segment download apparatus 1206 downloads.
Among another embodiment of the Play System of video file of the present invention, if key search and timestamp deriving means 1102,1204 do not obtain and the timestamp of the video segment of keyword coupling, then Play System 110,120 is play this video in normal way.
In the player method and Play System thereof of video file of the present invention, video segment is corresponding to the scene/camera lens of video file, and the annotation information of video segment comprises the single and/or lens-dividing script information of the script holder of scene/camera lens.Among the embodiment of the Play System of video file provided by the invention, annotation information comprises role, performer, action, scene and/or plot.
Introduce a specific embodiment of the player method of video file of the present invention below.As shown in figure 13, the user to carry out the concrete operation flow of " self-defined " playing video file as follows:
In the Play System of video file, at first execution in step 1302, a certain video file that user's selection will be watched.In step 1304, Web server will judge whether video file supports the broadcast mode of " self-defined " according to the object that the user selects.For video annotation information being arranged and supports the video file of self-defined broadcast, when the user by the Streaming Media website, when selecting to watch this video, comprise two options in the webpage that is occurred:
(1) watches this video in normal way;
(2) watch this video in self-defined mode.
Being stored on the Web server with the XML form with annotation information is that example describes: in step 1306, when the user selects " watching video in self-defined mode ", this Play System execution in step 1308, Web server will contain the page push of search box to the user, be downloaded by Web browser according to the flow process text that obtain, that contain annotation information of aforementioned annotation information simultaneously.
In step 1310, the user can import the keyword relevant with own desired video in search box.
In step 1312, a plurality of keywords can allow certain logical relation, if key search has hit results in step 1314, then execution in step 1316, and Web browser sends to media player with the result of search hit.
In step 1318, media player is converted into the RTSP request with the result who is hit, and downloads the corresponding video fragment by the video segment download apparatus from Web server, sends to media player, plays the video segment that is hit successively by it.
In step 1314, if key search is miss, then execution in step 1320, be whether the user can select to watch according to the common video broadcast mode, if the user selects "Yes", then the Web server execution in step 1322, push ordinary pages and carry out the normal video broadcast.
As shown in figure 13, do not support the broadcast mode of " self-defined " if in step 1304, judge video file, when perhaps if the user selects " watching this video in normal way " in step 1306, then also redirect execution in step of Web server 1322, push ordinary pages and carry out the normal video broadcast.
Among the embodiment of the player method of video file provided by the invention, for realizing " self-defined " playing video file, the user can adjust choosing of keyword according to the result of first retrieval, repeatedly retrieve, experience in the hope of obtaining the video-see of satisfying the demand, thereby make and watch video more flexible, convenient by retrieval comprehensively and accurately.
Introduce the stream transmission realization flow of the player method of video file of the present invention below.
Shown in Figure 14 (a), common stream transmission process is:
A) after the user selects a certain streaming media service, use the HTTP/TCP exchange of control information between Web browser 1402 and the Web server 1404, so that the real time data of needs transmission is found out from raw information;
B) Web browser on the user PC 1402 starts the video player program, uses HTTP from the meta file that Web server 1404 obtains about this video player to be carried out initialization.The parameter such as type of coding, URL that comprises directory information, video data among the Meta file.
C) operation real-time stream protocol (RTSP) between video player 1408 and the video server 1406 is with the required control information of exchange video Data Transmission.Video server 1406 uses the RTSP/UDP agreements that media data is transferred to user's player, in case media data arrives at user side, video player 1408 can be play output.
The stream transmission realization flow of the player method of video file of the present invention is shown in Figure 14 (b), if the user selects to watch video in " self-defined " mode, then at aforementioned flow process b) in, video player 1408 obtains from Web browser 1402, except meta file, also have one and comprise user's " search hit result " file.With the annotation information table shown in the table 1 is example, if the user is that keyword is retrieved with " performer A ", then the result of search hit is as shown in table 2 below.
| The number of hitting | from | to |
| 1 | T(0) | T(1) |
| 2 | T(2) | T(3) |
| 3 | T(n-1) | T(n) |
| 4 | T(n) | T_end |
Table 2
At aforementioned flow process c) in, after video server 1406 receives this hit results, be translated into the RTSP request, require video player 1408 to play the video segment that hits successively:
C-＞S：PLAY?rtsp://audio.example.com/twister.en?RTSP/1.0
CSeq：123
Session：12345678
Range：SMPTE＝T(0)-T(1)
C-＞S：PLAY?rtsp://audio.example.com/twister.en?RTSP/1.0
CSeq：124
Session：12345678
Range：SMPTE＝T(2)-T(3)
C-＞S：PLAY?rtsp://audio.example.com/twister.en?RTSP/1.0
CSeq：125
Session：12345678
Range：SMPTE＝T(n-1)-T(n)
C-＞S：PLAY?rtsp://audio.example.com/twister.en?RTSP/1.0
CSeq：126
Session：12345678
Range：SMPTE＝T(n)-T_end
The method of the annotate method of video file provided by the invention, video file method for quickly retrieving and self-defined playing video file has remedied and has lacked flexibility and interactive shortcoming on traditional Internet video broadcast mode, realized playing, for the user provides better viewing experience according to the corresponding contents that user's hobby and demand are extracted in the video.In addition, the method by the camera lens text annotation of the video files class video content that the present invention proposes not only can be used in category of the present invention, can also be applied in widely in the associated scenario of other video file retrievals.
Though reference example embodiment has described the present invention, should be appreciated that the present invention is not limited to disclosed exemplary embodiment.Many modifications and variations are conspicuous for the ordinary skill in the art.Selecting and describing embodiment is for better explanation principle of the present invention and practical application, thereby and makes those of ordinary skill in the art can understand the various embodiment that have various modifications that the present invention's design is suitable for special-purpose.
Claims (16)
1. the text annotation method of a video file is characterized in that, described method comprises:
Retrieve the frame of described video file and determine script holder's frame; Described script holder's frame is the frame that comprises clap-stick information;
From described script holder's frame, extract a mark, obtain the annotation information of described the corresponding video segment of mark according to described mark; Reel number, number, mirror number three field informations are collectively referred to as a mark;
Obtain the timestamp of described the corresponding video segment of mark;
Write down the timestamp and the annotation information of described the corresponding video segment of mark.
2. the text annotation method of video file according to claim 1, it is characterized in that, the described step that obtains the annotation information of described the corresponding video segment of mark according to described mark comprises: obtain and described the single and/or lens-dividing script information of script holder that mark is corresponding according to described mark, as the annotation information of described mark correspondence video segment.
3. the text annotation method of video file according to claim 1 is characterized in that,
The frame of the described video file of described retrieval determines that the step of script holder's frame comprises:
Retrieve described video file frame by frame, determine script holder's frame according to the grey level histogram information of each frame;
Described step from described script holder's frame extraction mark comprises:
Extract a mark based on bar-code identification method or digit recognition method from described script holder's frame.
4. the text annotation method of video file according to claim 1 is characterized in that, the step of the timestamp of described the corresponding video segment of mark of described acquisition comprises:
Obtain the timestamp of described the corresponding video segment of mark according to the quantity of FPS and frame by retrieval; And
The timestamp of described the corresponding video segment of mark behind script holder's frame is removed in the number of times correction that has occurred according to script holder's frame.
5. method according to claim 1 is characterized in that, described method also comprises step:
After described video file sheared, the corresponding timestamp of revising the video segment that is not sheared.
6. the text annotation system of a video file is characterized in that, described text annotation system (50,60,70) comprising:
Script holder's frame judgment means (502,602,702) is used to retrieve the frame of described video file and determines script holder's frame; Described script holder's frame is the frame that comprises clap-stick information;
The field mark extracts and annotation information deriving means (504,604,704), be used for from described script holder's frame judgment means (502,602,702) extract a mark in described script holder's frame of Que Dinging, according to the annotation information of described mark acquisition with described the corresponding video segment of mark; Reel number, number, mirror number three field informations are collectively referred to as a mark;
Timestamp deriving means (506,606,706) is used for obtaining and described field mark extracts and the timestamp of described mark correspondence video segment of annotation information deriving means (504,604,704) extraction;
Tape deck (508,608,708) is used to write down the described timestamp of described timestamp deriving means (506,606,706) acquisition and the described annotation information of being extracted by described mark extraction and annotation information deriving means (504,604,704).
7. the text annotation system of video file according to claim 6, it is characterized in that, described mark extracts and annotation information deriving means (504,604,704) obtain and described the single and/or lens-dividing script information of script holder that mark is corresponding according to described mark, as the annotation information of described the corresponding video segment of mark.
8. the text annotation system of video file according to claim 6 is characterized in that, described script holder's frame judgment means (502,602,702) is retrieved described video file frame by frame, determines script holder's frame according to the grey level histogram information of each frame;
Described field mark extracts and annotation information deriving means (504,604,704) comprising:
Field mark extraction unit (6041) is used for extracting a mark based on bar-code identification method or digit recognition method from described script holder's frame, sends described mark; And
Annotation information deriving means (6042) is used for the field mark that reception is extracted from described mark extraction unit (6041), and according to the annotation information of described mark acquisition with described the corresponding video segment of mark.
9. the text annotation system of video file according to claim 6 is characterized in that, described timestamp deriving means (506,606,706) comprising:
Timestamp acquiring unit (7061) is used for the timestamp that quantity according to FPS and frame by retrieval obtains described the corresponding video segment of mark; Send the timestamp of described the corresponding video segment of mark; And
Timestamp amending unit (7062), be used for receiving the timestamp of described the corresponding video segment of mark that obtains from described timestamp acquiring unit (7061), and remove the timestamp of described the corresponding video segment of mark behind script holder's frame according to the number of times correction that script holder's frame has occurred.
10. the text annotation system of video file according to claim 6 is characterized in that, described system comprises the update of time stamp device, after being used for described video file sheared, and the corresponding timestamp of revising the video segment that is not sheared.
11. the player method of a video file, it is characterized in that, described video file comprises having the annotation information that obtains as the text annotation method of any described video file in the claim 1 to 5 and a plurality of video segments of timestamp, and described player method comprises:
The input of reception keyword is retrieved described annotation information according to described keyword, the timestamp of the video segment of acquisition and described keyword coupling;
Video content according to the corresponding time period in the timestamp playing video file of the video segment of described coupling.
12. the player method of video file according to claim 11 is characterized in that, finishes according to the retrieval of described keyword to described annotation information at user side; Described user side is downloaded the described video file video content with the timestamp time corresponding section of the video segment of described keyword coupling from server end.
13. the player method of video file according to claim 11, it is characterized in that, described video segment is corresponding to the scene/camera lens of described video file, and the annotation information of described video segment comprises the single and/or lens-dividing script information of the script holder of described scene/camera lens;
Described annotation information comprises role, performer, action, scene and/or plot.
14. the Play System of a video file, it is characterized in that, described video file comprises having the annotation information that obtains as the text annotation method of any described video file in the claim 1 to 5 and a plurality of video segments of timestamp, and described Play System (110,120) comprising:
Key search and timestamp deriving means (1102,1204) are used to receive the keyword input, described annotation information are retrieved the timestamp of the video segment of acquisition and described keyword coupling according to described keyword;
Media playing apparatus (1104,1208) is used for the video content of basis by the corresponding time period in the timestamp playing video file of the video segment of the described coupling of described key search and timestamp deriving means (1102,1204) acquisition.
15. the Play System of video file according to claim 14 is characterized in that, described Play System (120) further comprises:
Annotation information download apparatus (1202) is used for the annotation information of described video file is downloaded to user side;
Described key search and timestamp deriving means (1204) are finished according to the retrieval of described keyword to described annotation information at described user side;
Video segment download apparatus (1206) is used for downloading from server end at described user side the video content of described video file and the timestamp time corresponding section of the video segment of described keyword coupling.
16. the Play System of video file according to claim 14, it is characterized in that, described video segment is corresponding to the scene/camera lens of described video file, and the annotation information of described video segment comprises the single and/or lens-dividing script information of the script holder of described scene/camera lens;
Described annotation information comprises role, performer, action, scene and/or plot.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009101707358A CN101646050B (en) | 2009-09-09 | 2009-09-09 | Text annotation method and system, playing method and system of video files |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009101707358A CN101646050B (en) | 2009-09-09 | 2009-09-09 | Text annotation method and system, playing method and system of video files |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101646050A CN101646050A (en) | 2010-02-10 |
| CN101646050B true CN101646050B (en) | 2011-04-06 |
Family
ID=41657736
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009101707358A Active CN101646050B (en) | 2009-09-09 | 2009-09-09 | Text annotation method and system, playing method and system of video files |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101646050B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110795597A (en) * | 2018-07-17 | 2020-02-14 | 上海智臻智能网络科技股份有限公司 | Video keyword determination method, video retrieval method, video keyword determination device, video retrieval device, storage medium and terminal |
Families Citing this family (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102572293A (en) * | 2010-12-16 | 2012-07-11 | 新奥特（北京）视频技术有限公司 | Field recording-based retrieval system |
| CN102572294A (en) * | 2010-12-16 | 2012-07-11 | 新奥特（北京）视频技术有限公司 | Field recoding system with ranking function |
| US8755430B2 (en) | 2010-12-22 | 2014-06-17 | Disney Enterprises, Inc. | System and method for audio video pattern sequence diagnostic tool |
| CN102129474B (en) * | 2011-04-20 | 2015-02-11 | 浙江宇视科技有限公司 | Method, device and system for retrieving video data |
| CN102685554B (en) * | 2012-05-24 | 2015-09-30 | 北京国双科技有限公司 | The processing method of video playback and device |
| CN102737689B (en) * | 2012-06-08 | 2016-01-20 | Tcl集团股份有限公司 | The localization method of a kind of head/run-out position, device and TV |
| CN103517092B (en) | 2012-06-29 | 2018-01-30 | 腾讯科技（深圳）有限公司 | A kind of method and device of video display |
| CN102903377B (en) * | 2012-11-01 | 2015-07-22 | 北京奇艺世纪科技有限公司 | Method and device for playing video with multiple video segments |
| WO2015024249A1 (en) * | 2013-08-23 | 2015-02-26 | Telefonaktiebolaget L M Ericsson(Publ) | On demand information for video |
| CN104469508B (en) * | 2013-09-13 | 2018-07-20 | 中国电信股份有限公司 | Method, server and the system of video location are carried out based on the barrage information content |
| CN103686200A (en) * | 2013-12-27 | 2014-03-26 | 乐视致新电子科技（天津）有限公司 | Intelligent television video resource searching method and system |
| CN103839562A (en) * | 2014-03-17 | 2014-06-04 | 杨雅 | Video creation system |
| CN105022738A (en) * | 2014-04-21 | 2015-11-04 | 上海京知信息科技有限公司 | Extracting and mapping method of network picture format file on the basis of histograms |
| CN105893387B (en) * | 2015-01-04 | 2021-03-23 | 伊姆西Ip控股有限责任公司 | Intelligent multimedia processing method and system |
| CN106205656B (en) * | 2015-05-08 | 2020-02-21 | 阿里巴巴集团控股有限公司 | Method and device for positioning file playing position |
| CN105187795B (en) * | 2015-09-14 | 2018-11-09 | 博康云信科技有限公司 | A kind of video tab localization method and device based on view library |
| CN105430536B (en) * | 2015-10-30 | 2018-09-11 | 北京奇艺世纪科技有限公司 | A kind of video pushing method and device |
| CN105898362A (en) * | 2015-11-25 | 2016-08-24 | 乐视网信息技术（北京）股份有限公司 | Video content retrieval method and device |
| CN108024148B (en) * | 2016-10-31 | 2020-02-28 | 腾讯科技（深圳）有限公司 | Behavior feature-based multimedia file identification method, processing method and device |
| CN106557586B (en) * | 2016-11-30 | 2020-06-19 | 北京星选科技有限公司 | Database access processing method and device |
| CN107704525A (en) * | 2017-09-04 | 2018-02-16 | 优酷网络技术（北京）有限公司 | Video searching method and device |
| CN110019880A (en) * | 2017-09-04 | 2019-07-16 | 优酷网络技术（北京）有限公司 | Video clipping method and device |
| CN107835397B (en) * | 2017-12-22 | 2019-12-24 | 成都华栖云科技有限公司 | Multi-lens video synchronization method |
| CN109101558B (en) * | 2018-07-12 | 2022-07-01 | 北京猫眼文化传媒有限公司 | Video retrieval method and device |
-
2009
- 2009-09-09 CN CN2009101707358A patent/CN101646050B/en active Active
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110795597A (en) * | 2018-07-17 | 2020-02-14 | 上海智臻智能网络科技股份有限公司 | Video keyword determination method, video retrieval method, video keyword determination device, video retrieval device, storage medium and terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101646050A (en) | 2010-02-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101646050B (en) | Text annotation method and system, playing method and system of video files | |
| US10714145B2 (en) | Systems and methods to associate multimedia tags with user comments and generate user modifiable snippets around a tag time for efficient storage and sharing of tagged items | |
| US11956516B2 (en) | System and method for creating and distributing multimedia content | |
| CN102290082B (en) | Method and device for processing brilliant video replay clip | |
| CN107707931B (en) | Method and device for generating interpretation data according to video data, method and device for synthesizing data and electronic equipment | |
| Saba et al. | Analysis of vision based systems to detect real time goal events in soccer videos | |
| CN109889882B (en) | Video clip synthesis method and system | |
| US8677399B2 (en) | Preprocessing video to insert visual elements and applications thereof | |
| CN1816809A (en) | Enhanced organization and retrieval of digital images | |
| CN106060578A (en) | Producing video data | |
| WO2011050280A2 (en) | Method and apparatus for video search and delivery | |
| JP2001216726A (en) | Device and method for generating audio signal and/or video signal | |
| CN104915433A (en) | Method for searching for film and television video | |
| US20150195626A1 (en) | Augmented media service providing method, apparatus thereof, and system thereof | |
| JP2006155384A (en) | Video comment input/display method and device, program, and storage medium with program stored | |
| CN106454493A (en) | Method for inquiring currently played television program information and smart TV | |
| CN105340278A (en) | Automated pre and post roll prodution | |
| CN111368141A (en) | Video tag expansion method and device, computer equipment and storage medium | |
| EP3735778B1 (en) | Coordinates as ancillary data | |
| CN114117120A (en) | Video file intelligent index generation system and method based on content analysis | |
| KR102261221B1 (en) | System And Method For Obtaining Image Information, And Method For Displaying Image Acquisition Information | |
| US20230164369A1 (en) | Event progress detection in media items | |
| JP2009129039A (en) | Content storage device and content storage method | |
| AU734202B2 (en) | Linking metadata with a time-sequential digital signal | |
| NO20140958A1 (en) | Digital content search method and system |
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 |