CN108388583A

CN108388583A - A kind of video searching method and video searching apparatus based on video content

Info

Publication number: CN108388583A
Application number: CN201810077785.0A
Authority: CN
Inventors: 罗江春; 陈锡岩
Original assignee: Beijing Survey Technology Co Ltd
Current assignee: Beijing Survey Technology Co Ltd
Priority date: 2018-01-26
Filing date: 2018-01-26
Publication date: 2018-08-10
Also published as: WO2019144850A1

Abstract

The present invention provides a kind of video searching method and video searching apparatus based on video content, receive search sequence input by user, according to video frame index determine including at least one video frame corresponding with search sequence matching and its corresponding at least one video, wherein, the video frame index is establishd or updated according to the identification of the corresponding information of each video frame in video, according to the corresponding timestamp of at least one video frame, position the play position of its corresponding at least one video, at least one video is supplied to the user with the play position.The present invention can search for video corresponding with search sequence matching based on the information of each video frame in video, its video to be watched accurately is found for user, and after searching out corresponding video, the play position of video can be positioned to playing out at video frame corresponding with search sequence, reduce the operating time of user, the accuracy rate of the video search of raising.

Description

A kind of video searching method and video searching apparatus based on video content

Technical field

The present invention relates to video search field more particularly to a kind of video search techniques based on video content.

Background technology

With the continuous development of multimedia service, the number of videos stored in database is more and more, and user searches video The demand of rope is also increasing.

The prior art is mainly by the external label of video labeling, for example, video title, publisher's title, video frequency abstract, Or profile information come realize from the data of magnanimity search for user need video.But user can only sometimes remember some The picture either picture exhibition or the content of introduction, can not often search according to the general external label of video labeling Video including the picture.

Therefore, the method that the content of a kind of content based on video or video frame carries out video search how is proposed, at For this field urgent need to resolve the technical issues of one of.

Invention content

The object of the present invention is to provide a kind of video searching apparatus and video searching method based on video content.

According to an aspect of the invention, there is provided a kind of video searching apparatus based on video content, wherein the video Searcher includes：

Reception device, for receiving search sequence input by user；

Determining device, for according to video frame index determine including it is corresponding with search sequence matching at least One video frame and its corresponding at least one video, wherein the video frame index is according to each video frame in video The identification of corresponding information is establishd or updated；

Positioning device, for according to the corresponding timestamp of at least one video frame, it is corresponding at least one to position its At least one video is supplied to the user by the play position of video with the play position.

Preferably, which further includes：

First identification device, for the corresponding information of each video frame in each video to be identified, acquisition pair The markup information answered；

Updating device, for according to the markup information, establising or updating the video frame index；

Wherein, in determining device according to the search sequence match corresponding at least one video frame and its it is corresponding extremely The mode of a video includes less：

According to the search sequence, matches in the video frame index and to determine to have and match with the search sequence At least one video frame of markup information and its corresponding at least one video.

Preferably, which further includes：

Second identification device, for the corresponding information of each video frame in each video to be identified, acquisition pair The markup information answered；

Device is established, for according to the markup information, a subindex to be established for each video；

According to the search sequence, is matched in the corresponding each subindex of each video determine have and institute successively State at least one video frame for the markup information that search sequence matches.

Preferably, the markup information is obtained by following at least any one：

The characteristics of image for identifying a video frame images, one video frame images pair are determined according to described image feature The markup information answered；

Identify the characteristics of image of a video frame images, and the preposition image of one video frame images according to identification With the characteristics of image of postposition image, corresponding markup information is determined；

It identifies the audio-frequency information corresponding at least one continuous video frame, the audio-frequency information is converted into word letter Breath determines the corresponding markup information of at least one continuous video frame according to the text information；

Extract the caption information corresponding at least one continuous video frame, according to the caption information determine described at least One corresponding markup information of continuous video frame.

Preferably, which further includes：

Collator, for being ranked up at least one video, at least one video after being sorted；

Wherein, positioning device is used for：

According at least one video frame, the broadcasting position of at least one video after its corresponding described sequence is positioned It sets, at least one video after the sequence is supplied to by the user with the play position.

Preferably, at least one video is ranked up according to following at least any one：

The included quantity with the matched video frame of the search sequence in video；

The corresponding video distribution person's information of video；

The source-information of video；

The clarity of video；

The subject information of video；

Feedback information of the user to video.

Preferably, the positioning device is additionally operable to：

Obtain a target video of the user from selected at least one video；

If the target video includes multiple and the matched video frame of the search sequence, determine by any one of following The play position of the position target video：

Acquiescence is determined according to first in the target video with the timestamp of the matched video frame of the search sequence；

Acquiescence positioning to the highest video frame of search sequence matching degree；

It is selected by the user.

According to another aspect of the present invention, a kind of video searching method based on video content is additionally provided, wherein should Video searching method includes：

A. search sequence input by user is received；

B. according to video frame index determine including at least one video frame corresponding with search sequence matching And its corresponding at least one video, wherein the video frame index is according to the corresponding information of each video frame in video Identification establishd or updated；

C. according to the corresponding timestamp of at least one video frame, the broadcasting position of its corresponding at least one video is positioned It sets, at least one video is supplied to by the user with the play position.

Preferably, which further includes：

The corresponding information of each video frame in each video is identified, corresponding markup information is obtained；

According to the markup information, the video frame index is establishd or updated；

Wherein, corresponding at least one video frame and its corresponding at least one are matched according to the search sequence in step b The mode of a video includes：

Preferably, which further includes：

According to the markup information, a subindex is established for each video；

Preferably, the markup information is obtained by following at least any one：

Identify the characteristics of image of a video frame images, and according to the preposition image of the video frame images of identification with after The characteristics of image for setting image determines corresponding markup information；

It identifies the audio-frequency information corresponding to a video frame, the audio-frequency information is converted into text information, according to described Text information determines the corresponding markup information of the video frame；

The caption information corresponding to a video frame is extracted, the corresponding mark of the video frame is determined according to the caption information Note information.

Preferably, which further includes：

At least one video is ranked up, at least one video after being sorted；

Wherein, step c includes：

The corresponding video distribution person's information of video；

The source-information of video；

The clarity of video；

The subject information of video；

Feedback information of the user to video.

Preferably, it is supplied to the user to further include at least one video with the play position in step c：

Obtain a target video of the user from selected at least one video；

It is selected by the user.

According to a further aspect of the invention, a kind of computer readable storage medium is additionally provided, it is described computer-readable Storage medium is stored with computer code, when the computer code is performed, any one of them video search side as above Method is performed.

According to a further aspect of the invention, a kind of computer program product is additionally provided, when the computer program produces When product are executed by computer equipment, any one of them video searching method as above is performed.

According to a further aspect of the invention, a kind of computer equipment is additionally provided, the computer equipment includes storage Device and processor, computer code is stored in the memory, and the processor is configured to by executing the computer Code is to execute any one of them method as above.

Compared with prior art, the present invention has the following advantages：

The present invention according to video frame index determine including corresponding with search sequence matching at least one regard Frequency frame and its corresponding at least one video, and the video frame index is according to the corresponding letter of each video frame in video The identification of breath is establishd or updated, and then according to the corresponding timestamp of at least one video frame, it is corresponding at least to position its At least one video is supplied to the user by the play position of one video with the play position, not only can be with base The information of each video frame determines video corresponding with search sequence matching in video, accurately finds it for user and wants The video of viewing, and after searching out corresponding video, the play position of video can be positioned to corresponding with search sequence Video frame at play out so that user can quickly watch video clip corresponding with the search sequence, reduce The operating time of user, and can accurately be that user position to play position corresponding with search sequence.

Description of the drawings

By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, of the invention other Feature, objects and advantages will become more apparent upon：

Fig. 1 shows a kind of structural representation of video searching apparatus based on video content according to one aspect of the invention Figure；

Fig. 2 shows a kind of signals of the flow of video searching method based on video content according to a further aspect of the present invention Figure；

Fig. 3 shows the block diagram of the exemplary computer system/server suitable for being used for realizing embodiment of the present invention.

Same or analogous reference numeral represents same or analogous component in attached drawing.

Specific implementation mode

It should be mentioned that some exemplary embodiments are described as before exemplary embodiment is discussed in greater detail The processing described as flow chart or method.Although operations are described as the processing of sequence by flow chart, therein to be permitted Multioperation can be implemented concurrently, concomitantly or simultaneously.In addition, the sequence of operations can be rearranged.When it The processing can be terminated when operation completion, it is also possible to the additional step being not included in attached drawing.The processing It can correspond to method, function, regulation, subroutine, subprogram etc..

The computer equipment includes user equipment and the network equipment.Wherein, the user equipment includes but not limited to electricity Brain, smart mobile phone, PDA etc.；The network equipment includes but not limited to that single network server, multiple network servers form Server group or the cloud being made of a large amount of computers or network server for being based on cloud computing (Cloud Computing), wherein Cloud computing is one kind of Distributed Calculation, a super virtual computer being made of the computer collection of a group loose couplings.Its In, the computer equipment can isolated operation realize the present invention, also can access network and by with other calculating in network The present invention is realized in the interactive operation of machine equipment.Wherein, the network residing for the computer equipment include but not limited to internet, Wide area network, Metropolitan Area Network (MAN), LAN, VPN network etc..

It should be noted that the user equipment, the network equipment and network etc. are only for example, other are existing or from now on may be used The computer equipment or network that can occur such as are applicable to the present invention, should also be included within the scope of the present invention, and to draw It is incorporated herein with mode.

Method (some of them are illustrated by flow) discussed hereafter can be by hardware, software, firmware, centre Part, microcode, hardware description language or its arbitrary combination are implemented.Implement when with software, firmware, middleware or microcode When, program code or code segment to implement necessary task can be stored in machine or computer-readable medium and (for example deposit Storage media) in.(one or more) processor can implement necessary task.

Specific structure and function details disclosed herein are only representative, and are for describing the present invention show The purpose of example property embodiment.But the present invention can be implemented by many alternative forms, and be not interpreted as It is limited only by the embodiments set forth herein.

Although it should be understood that may have been used term " first ", " second " etc. herein to describe each unit, But these units should not be limited by these terms.The use of these items is only for by a unit and another unit It distinguishes.For example, without departing substantially from the range of exemplary embodiment, it is single that first unit can be referred to as second Member, and similarly second unit can be referred to as first unit.Term "and/or" used herein above include one of them or The arbitrary and all combination of more listed associated items.

It should be understood that when a unit is referred to as " connecting " or when " coupled " to another unit, can directly connect Another unit is connect or be coupled to, or may exist temporary location.In contrast, when a unit is referred to as " directly connecting Connect " or " direct-coupling " arrive another unit when, then be not present temporary location.It should explain in a comparable manner and be used to retouch State the relationship between unit other words (such as " between being in ... " compared to " between being directly in ... ", " and with ... it is adjacent Closely " compared to " with ... be directly adjacent to " etc.).

Term used herein above is not intended to limit exemplary embodiment just for the sake of description specific embodiment.Unless Context clearly refers else, otherwise singulative used herein above "one", " one " also attempt to include plural number.Also answer When understanding, term " include " and or " include " used herein above provide stated feature, integer, step, operation, The presence of unit and/or component, and do not preclude the presence or addition of other one or more features, integer, step, operation, unit, Component and/or a combination thereof.

It should further be mentioned that in some replace implementations, the function action being previously mentioned can be according to different from attached The sequence indicated in figure occurs.For example, involved function action is depended on, the two width figures shown in succession actually may be used Substantially simultaneously to execute or can execute in a reverse order sometimes.

Present invention is further described in detail below in conjunction with the accompanying drawings.

Fig. 1 shows a kind of structural representation of video searching apparatus based on video content according to one aspect of the invention Figure.Video searching apparatus 1 includes：Reception device 101, determining device 102 and positioning device 103.

Here, video searching apparatus 1, for example in computer equipment, the computer equipment includes user equipment and net Network equipment.It is described in detail so that the video searching apparatus 1 is located in the network equipment as an example below.

Wherein, the network equipment includes but not limited to the service of single network server, multiple network servers composition Device group or the cloud being made of a large amount of computers or network server for being based on cloud computing (Cloud Computing), wherein cloud meter It is one kind of Distributed Calculation, a super virtual computer being made of the computer collection of a group loose couplings.Wherein, institute State computer equipment can isolated operation realize the present invention, also can access network and by with other computer equipments in network Interactive operation realize the present invention.Wherein, the network residing for the computer equipment includes but not limited to internet, wide area Net, Metropolitan Area Network (MAN), LAN, VPN network etc..

Reception device 101 receives search sequence input by user.Specifically, if user wants to search for some and search sequence phase The video clip of pass, or when wanting that it includes the video of aforementioned video segment to search for some, by input interface input inquiry sequence, And search button is clicked, reception device 101 receives search sequence input by user, so that follow up device is searched in video database Rope includes the video of the video clip including video clip corresponding with search sequence, or further.For example, user Think search including the video clip of " Huang know bright eat at home " or including the video of the video clip, user regarding Frequency search input interface input：" Huang knows bright eat at home " this search sequence, reception device 101 is for example by primary or more The mode of secondary calls application interface receives " Huang knows bright eat at home " this search sequence input by user.

Determining device 102 according to video frame index determine including corresponding with search sequence matching at least one A video frame and its corresponding at least one video, wherein the video frame index is according to each video frame pair in video The identification for the information answered is establishd or updated.

Here, the video frame index can be the general index for including video frame in all videos, it is every for can also be The subindex that a video is established respectively.By the way that each video frame in each video to be identified respectively, identify Go out the information included by each video frame, wherein the same video frame may include different information, then be regarded according to each The information that frequency frame and each video frame respectively include establishes a total video frame index for corresponding to all videos, or A video frame index is established respectively for each video.The video frame index includes：Each video frame respectively with its Including at least one information correspondence.Further, each described video frame can also respectively include an attribute Label, the attribute are marked for marking which video is the video frame belong to.

Specifically, after video searching apparatus 1 is to the corresponding information of each video frame is identified in each video, from Each video frame obtains the synopsis of the content of at least one correspondence video frame respectively, and the synopsis is for example distinguished According to the keyword that comes out of different information summaries included in the video frame, referred to herein as markup information, therefore, depending on Frequency frame index includes at least one：The correspondence of video frame different markup informations from the video frame respectively.Determine dress 102 are set according to the search sequence information that receives, accurately matching, by way of fuzzy matching or both is combined, regarding Matching inquiry is carried out in frequency frame index, as judged whether the search sequence hits each video frame institute in the video frame index successively Corresponding markup information, if hitting some markup information, the corresponding video frame of the markup information is that this input by user is looked into The video frame corresponding to sequence is ask, and can determine the corresponding video of the video frame in turn.For example, determining device 102 is according to institute The attribute label for stating at least one video frame can determine corresponding at least one video.Wherein, with the search sequence It can belong to the video frame of the same video to match corresponding at least one video frame, can also be to belong to different video Video frame；At least one video frame can be continuous video frame, can also be discontinuous video frame.Wherein, right The identification of the corresponding information of each video frame includes but not limited in video：Scene Recognition, base based on independent frame image In the scene Recognition of continuous multiple frames image, the identification based on the corresponding audio-frequency information of image, it is based on the corresponding caption information of image Identification.

For example, the 1000th video frame of video V1 is the yellow picture for knowing bright eat at home, which also illustrates Huang Know the bright content for eating fried bean sauce, or the yellow picture for knowing bright parent, then after information in the video frame being identified, video frame rope May include in drawing：Attribute labeled as video V1 the 1000th video frame respectively with " Huang know bright eat at home ", " Huang dawn is bright to eat The correspondence of fried bean sauce " and " Huang knows bright parent ".If it includes that Huang Xiaoming exists that the 1000-1049 video frame of video V1, which is all, Family has a meal, and Huang dawn is bright, and eat is fried bean sauce, and the yellow picture for knowing bright parent three, but in the 1025-1049 video frame There is not yellow dawn bright parent, then in video frame index, each video in the 1000-1024 video frame of video V1 Frame is all respectively corresponding " Huang knows bright eat at home ", " Huang dawn bright eat fried bean sauce " and " yellow dawn bright parent "；The 1025- of video V1 Each video frame in 1049 video frame corresponding respectively " Huang knows bright eat at home " and " yellow know bright eat fried bean sauce ".If searching The video database of rope video includes video V1, and video V2, video V3 ... wait multiple videos, can establish including the data Total video frame index of video frame in library in all videos, or it only includes its video that each video, which establishes one, The video frame index of frame, when for example, video V1 establishes a corresponding video frame index, determining device 102 is regarded each successively Frequently it is that the search sequence matches corresponding at least one video frame in corresponding video frame index.

If the search sequence that reception device 101 receives is " Huang knows bright eat at home ", mould may be used in determining device 102 Matched mode is pasted, is inquired in the video frame index by " Huang Xiaoming ", " Huang knows bright have a meal ", " Huang knows bright be in ", " Huang dawn The mark letter of the keywords such as bright eat at home ", " star's eat at home ", " Huang Xiaoming and angelababy eat at homes " hit Breath, the corresponding video frame of the markup information of hit be with the video frame corresponding to " Huang knows bright eat at home ", and in turn It can determine the corresponding video of the video frame；Or accurate matching way is used, it searches for and corresponds to only in the video frame index Video frame in " Huang knows bright eat at home " and its corresponding video；Or match using fuzzy matching and accurately the side being combined Formula is that user matches corresponding video frame and video.If by " Huang knows bright eat at home " this search sequence in video frame index The markup information of hit is 54, and determining device 102 determines that 54 markup informations of hit correspond to respectively in video frame index In：The 1000-1049 video frame of video V1, the 225th video frame of video V3 and the 25th frame of video V8, the 126th Frame, the 127th frame include then the 1000-1049 video frame of video V1 with " Huang knows bright eat at home " corresponding video frame, depending on The 225th video frame of frequency V3 and the 25th, 126,127 video frame of video V8, it is corresponding with " Huang knows bright eat at home " Video is respectively video V1, V3 and V8.

Where it determines that device 102 determines video corresponding with search sequence according to video frame index, wherein the video Frame index is according to establishing the identification of the corresponding information of each video frame in video or newer, accordingly, it is determined that device 102 can carry out retrieval matching, determination is wherein queried according to video frame index to the information of each video frame in video The information of sequence hits determines video frame corresponding with search sequence, so that it is determined that being looked into described according to the described information of hit The corresponding video of sequence is ask, user is supplied to.Content based on video, i.e., the information of each video frame is searched in video Rope so that user can more accurately search the video that oneself wants viewing, improve the search experience of user.

Those skilled in the art will be understood that the knowledge to the corresponding information of each video frame in video is only otherwise Citing, the existing or knowledge to the corresponding information of each video frame in video that is likely to occur from now on otherwise, such as may be used It should also be included in the scope of the present invention suitable for the present invention, and be incorporated herein by reference herein.

Positioning device 103 positions its and corresponding at least one regards according to the corresponding timestamp of at least one video frame At least one video is supplied to the user by the play position of frequency with the play position.Specifically, every in video A video frame has its corresponding timestamp in the video.Described in the determination of determining device 102 is corresponding with search sequence at least After one video frame, positioning device 103 obtains timestamp of each video frame in corresponding video, then By the play position of video corresponding at least one video frame position to the matched video frame of search sequence Position, when choosing the video to watch so as to user, the video since at video frame corresponding with search sequence play.It is fixed Position device 103 is so that user it can want the video clip seen with direct viewing.

Preferably, the positioning device 103 obtains a mesh of the user from selected at least one video Video is marked, if the target video includes multiple and the matched video frame of the search sequence, is positioned by any one of following The play position of the target video：Acquiescence is according to first in the target video and the matched video frame of the search sequence Timestamp determine；It is selected by the user.Specifically, it is selected at least one video that user's self locating device 103 provides In target video when being watched, positioning device 103 obtains the target video that user chooses.If in the target video Including the multiple and matched video frame of the search sequence, then：It can give tacit consent to and position the play position of the video to timestamp The position of that minimum video frame, the i.e. position of the earliest video frame of reproduction time can also give tacit consent to positioning to matching degree Highest video frame；By the play position random position of the video to any one matched video frame；Or it is selected by user Which video frame is play position position to, and the method for selection has tabular form and playing progress bar labeling form, wherein the row Sheet form includes and includes not only pop-up list, for example, after a video is chosen by user, positioning device 103 is with pop-up list Form provide a choice box to the user, by user select since which corresponding video frame play.Then positioning device At least one video that located play position is supplied to user by 103 so which video no matter user therefrom select Watched, the video all can directly since with from the matched video frame of the search sequence play, thus user can be straight Connect viewing picture corresponding with the search sequence.

It connects example, is with " Huang dawn bright have a meal " matched video frame in video V1：1000-1049 in video V1 regard The the 25th, 126,127 video frame in the 225th video frame, video V8 in frequency frame, video V3, then can give tacit consent to video V1 Play position position to the 1000th video frame, acquiescence positions the play position of V3 to the 225th video frame, and acquiescence is by V8 Play position position to the 25th video frame；Or V1 is positioned to any one video in the 1000-1049 video frame Any one video frame in frame, the 25th of V8 the, 126,127 video frame；Or it can select to determine play position by user Position is to which matched video frame.

Here, sequence not only can be searched for and be inquired to video searching apparatus 1 based on the information of each video frame in video Row match corresponding video, are accurately found for user and meet the video that user watches condition, and are determined in determining device 102 After corresponding video, positioning device 103 can accurately position the play position of the video to position corresponding with search sequence It sets, so that the video is played out from video frame corresponding with search sequence, so that user can more quickly watch To video clip corresponding with the search sequence, the search efficiency of user is improved, reduces the operating time of user.Cause This, video searching apparatus 1 can not only accurately search for user to be wanted to watch video, moreover it is possible to which the play position of video is accurate The position for the video-frequency band for thinking viewing is positioned to user.

Preferably, video searching apparatus 1 further includes that 104 (not shown) of the first identification device and updating device 105 (do not show Go out).

The corresponding information of each video frame in each video is identified in first identification device 104, is corresponded to Markup information.Specifically, the first identification device 104 carries out image knowledge to each video frame in each video respectively Not, and to the corresponding pertinent audio information of each video frame or caption information it is identified, obtains corresponding at least one A markup information, wherein the markup information is the abstract of the content included by corresponding video frame.For example, the of video V1 1000 video frame are the yellow pictures for knowing bright eat at home, which further includes the content that Huang Xiaoming eats fried bean sauce, Huo Zhehuang Know bright parent, therefore, the 1000th video frame of video V1 respectively with markup information：The bright eat at homes of Huang Xiao, Huang know bright eat Fried bean sauce, yellow dawn bright parent correspond to.

Preferably, the mode of the corresponding at least one markup information of acquisition includes but not limited to：

1) characteristics of image for identifying a video frame images, one video frame images are determined according to described image feature Corresponding markup information.Specifically, by way of image recognition, the characteristics of image of the video frame images is identified, according to institute It states characteristics of image and determines at least one markup information that the video frame images include, at least one markup information difference For the abstract of the different content included by the video frame.

2) characteristics of image of a video frame images, and the preposition figure of one video frame images according to identification are identified The characteristics of image of picture and postposition image determines corresponding markup information.It specifically, if can not be true according only to the video frame images Determine markup information, it can be according to the image of the preposition at least one video frame and at least one video frame of postposition of the video frame images Feature, comprehensive descision determine at least one synopsis that the video frame images include, for example, according to front and back multiple video frame Middle object position movement determine video frame in the object state, then using at least one synopsis as this At least one markup information of frame video, and the video frame correspondence updates at least one markup information of the video frame respectively To in the video frame index.For example, the content that a video frame images are shown is an automobile, it can not according only to a video frame Identify the state that the state that the automobile occurred in the video frame is off still moves, it is therefore desirable to according to the video frame Preposition image or postposition image determine the operating status of the automobile jointly, if according to the automobile occurred in the video frame before it The change in location in image and postposition image is set, judges the automobile for operating status, and is judged according to the size of change in location Automobile is high-speed cruising state, then using automobilism or high-speed running automobile as the markup information of the video frame, regarding It is corresponding with the video frame in frequency frame index.

3) it identifies the audio-frequency information corresponding to a video frame, the audio-frequency information is converted into text information, according to institute It states text information and determines the corresponding markup information of the video frame.Specifically, the corresponding audio-frequency information of the video frame is identified, it will The audio-frequency information carries out speech recognition, is converted into text information, determines that the video frame is corresponding according to the text information Markup information.

4) caption information corresponding to a video frame is extracted, determines that the video frame is corresponding according to the caption information Markup information.

5) by way of artificial intelligence, the characteristics of image of one video frame images of automatic identification, the video frame images The content and the video frame of audio-frequency information corresponding to the characteristics of image of preposition image and postposition image, the video frame images The content of the corresponding caption information of image, and determine that the video frame images are corresponding based on the above- mentioned information by people's intelligent recognition Markup information.

Those skilled in the art will be understood that the mode of the corresponding at least one markup information of above-mentioned acquisition is only to lift Example, it is existing or be likely to occur from now on other obtain the mode of corresponding at least one markup information, it is such as applicable It should also be included in the scope of the present invention in the present invention, and be incorporated herein by reference herein.

Updating device 105 establishs or updates the video frame index, wherein in determining device according to the markup information The middle mode that corresponding at least one video frame and its corresponding at least one video are matched according to the search sequence includes：Root According to the search sequence, is matched in the video frame index and determine that there is the markup information to match with the search sequence At least one video frame and its corresponding at least one video.Specifically, updating device 105 is according to each video frame in video Correspondence between corresponding markup information establishes video frame index.Wherein, video frame index includes：Each is regarded Frequency frame distinguishes the correspondence of corresponding at least one markup information, so as to video searching apparatus 1 by search sequence with regard Markup information in frequency frame index is matched, if successful match, the corresponding video frame of the markup information is and search sequence Matched video frame, then video searching apparatus 1 is marked according to the attribute of each video frame just can determine that video frame place Video.Wherein, the video frame index can be the general index established according to each video frame of each video, also may be used To be the subindex established respectively according to each video frame in any one video.When the video frame index is packet When including the general index of each video frame in each video, after the search sequence that reception device 101 receives user, determine Device 102 is matched according to the search sequence in the video frame index, then according to matched at least one video frame point It Que Ding not its corresponding video；If the video frame is at least one subindex corresponding with each video, determine Device 102 is matched in the corresponding subindex of each video successively according to the search sequence of reception, is corresponded in matching Video frame when, determining device 102 can directly know video corresponding with the video frame.

For example, the 48th video frame of video V112 is pictures of the Zhang Jie in singing, by the audio-frequency information of the video frame, The poster information occurred on caption information and the video frame, the first identification device 104 may recognize that the content of the video frame also Including：Zhang Jie sings in Chinese new song, and Zhang Jie sings the information such as inverse war, thus will " Zhang Jie is singing ", " Zhang Jie is newly sung Chinese Sound singing ", the markup information of " Zhang Jie sings inverse war " as the video frame.The markup information of other video frame in obtaining the video And after the markup information of the video frame of other videos, the foundation of updating device 105, which includes that video frame is corresponding with markup information, closes The video frame index of system.If the video frame index is total video frame index, including：Attribute is labeled as video V112 The 48th video frame respectively with markup information " Zhang Jie is singing ", " Zhang Jie sings in Chinese new song ", " Zhang Jie sings inverse war " It is corresponding；If the video frame is the subindex corresponding to each video, is wrapped in the corresponding video frame index of video V112 It includes：112nd video frame respectively with markup information " Zhang Jie is singing ", " Zhang Jie sings in Chinese new song ", " Zhang Jie sings inverse War " is corresponding.Wherein, the markup information of each video frame can have multiple, be carried out comprehensively in face of the video frame with never Tongfang Mark, so that the video frame is retrieved or is matched to by easier.

Preferably, video searching apparatus 1 further includes that 106 (not shown) of the second identification device and updating device 107 (do not show Go out).

The corresponding information of each video frame in each video is identified in second identification device 106, is corresponded to Markup information；Device 107 is established according to the markup information, a subindex is established for each video；Wherein, in determination The mode packet of corresponding at least one video frame and its corresponding at least one video is matched in device according to the search sequence It includes：According to the search sequence, successively in the corresponding each subindex of each video match determine have looked into described Ask at least one video frame for the markup information that sequence matches.Here, the second identification device 106 and aforementioned first identification device 104 realization method is identical.Specifically, the second identification device 106 obtains markup information corresponding with video frame, establishes device 107 establish a subindex according to the markup information, for each video, and reception device 101 receives the search sequence of user Afterwards, the markup information during determining device 102 successively indexes the corresponding video frame of each video and the search sequence Carry out matching correspondence, and according at least one markup information corresponding with search sequence matching determine with it is described at least one The corresponding video frame of markup information institute.

Preferably, it can be provided to the user and be searched in video in the video of arbitrary video-frequency band based on video searching apparatus 1 Function of search so that user when watching video by the search of the keyword inputted to it, can be directly by play position It positions to it and thinks the video frame position of viewing.If selection one regards at least one video that user's self locating device 103 provides Frequency plays out viewing or user arbitrarily clicks a video and plays out, and user can be according to the corresponding video of the video Frame index, i.e. subindex, input user thinks the arbitrary video frame or piece of video of viewing in function of search input frame in video Markup information corresponding to section, so that video searching apparatus 1 scans in the video.Obtaining, the markup information is corresponding After video frame, video searching apparatus 1 positions play position to video frame corresponding with the markup information.Wherein, if with When the corresponding video frame of the markup information is multiple, the acquiescence of positioning device 103 positions play position to its timestamp minimum At video frame, that is, it is at the video frame of earliest play position.For example, video V27 is film video《Ice and snow unusual relationship》Segment, In the presence of video frame index corresponding with video V27, wherein video frame index includes the 285th video frame and markup information " ice The Snow Queen Quest, which enchants, builds ice and snow palace " correspondence.User inputs in the corresponding function of search input frames of video V27 " ice and snow palace " this search sequence, determining device 102 can be user match " ice and snow queen, which enchants, builds ice and snow palace " this Then markup information determines that its corresponding video frame is the 285th video frame, 103 basis of positioning device according to the markup information Play position is positioned to playing the position of the 285th video frame, and requries the users and be by the timestamp of the 285th video frame It is no to play since the position.If user agrees to that this is regarded the video directly from " ice and snow queen, which enchants, builds ice and snow palace " Frequency frame starts to play.It is pulled manually without user here, being realized by video searching apparatus 1, it just can be just directly fixed by searching for Position to think viewing position play out.

Preferably, video searching apparatus 1 further includes 108 (not shown) of collator.

Collator is ranked up at least one video, at least one video after being sorted；Wherein, it positions Device 103 is used for：According at least one video frame, the broadcasting of at least one video after its corresponding described sequence is positioned At least one video after the sequence is supplied to the user by position with the play position.Specifically, to it is described at least The mode that one video is ranked up includes but not limited to：

1) quantity with the matched video frame of the search sequence included in video.Specifically, for example, being looked into described Sort Priority height or the weight order for asking the video more than the quantity of the video frame of sequences match are higher.

2) the corresponding video distribution person's information of video.Specifically, if the historical record that video distribution person issues video compares It is more, and have higher rating, then the Sort Priority of the video is higher or weight order is higher.

3) source-information of video.Specifically, if source video sequence is in more well-known big website, for example, iqiyi.com, excellent Extremely, the well-known website such as Sohu, then the Sort Priority of the video is higher or weight order is higher.

4) clarity or fluency of video.Specifically, if the clarity of video is higher, or broadcasting is more smooth, then should The Sort Priority of video is higher or weight order is higher.

5) subject information of video.Specifically, if the theme of video is more popular within the most a recent period of time, the video Sort Priority it is higher or weight order is higher.

6) feedback information of the user to video.It specifically, should if user is relatively high to the feedback of the video or scoring The Sort Priority of video is higher or weight order is higher.

7) which kind of mode matching inquiry is carried out by, for example, fuzzy matching or accurate matched mode, if the video is By accurately matching acquisition, then the Sort Priority of the video is higher or weight order is higher；If the video is by fuzzy Matching obtains, then the Sort Priority of the video is higher or weight order is higher.

Those skilled in the art will be understood that the above-mentioned mode being ranked up to the video is only for example, it is existing or Other modes being ranked up to video that person will be likely to occur from now on, are such as applicable to the present invention and should also be included in protection of the present invention In range, and it is incorporated herein by reference herein.

Fig. 2 shows a kind of signals of the flow of video searching method based on video content according to a further aspect of the present invention Figure.

In step s 201, video searching apparatus 1 receives search sequence input by user.Specifically, if user wants to search for Some and the relevant video clip of search sequence, or when wanting that it includes the video of aforementioned video segment to search for some, pass through input Interface input inquiry sequence, and search button is clicked, video searching apparatus 1 receives search sequence input by user, so as to follow-up Device is searched in video database including video clip corresponding with search sequence, or further, including the video The video of segment.For example, user thinks that search is regarded including the video clip of " Huang knows bright eat at home " or including this The video of frequency segment, user input in video search input interface：" Huang knows bright eat at home " this search sequence, in step In S201, video searching apparatus 1 is for example received by way of primary or multiple calls application interface input by user " Huang knows bright eat at home " this search sequence.

In step S202, video searching apparatus 1 according to video frame index determine including with the search sequence Match corresponding at least one video frame and its corresponding at least one video, wherein the video frame index is according to video In the identification of the corresponding information of each video frame establishd or updated.

Specifically, after video searching apparatus 1 is to the corresponding information of each video frame is identified in each video, from Each video frame obtains the synopsis of the content of at least one correspondence video frame respectively, and the synopsis is for example distinguished According to the keyword that comes out of different information summaries included in the video frame, referred to herein as markup information, therefore, depending on Frequency frame index includes at least one：The correspondence of video frame different markup informations from the video frame respectively.In step In S202, video searching apparatus 1 is according to the search sequence information received, by accurately matching, fuzzy matching or both phase In conjunction with mode, matching inquiry is carried out in video frame index, such as judge whether the search sequence hits the video frame rope successively Draw the markup information corresponding to each video frame, if hitting some markup information, the corresponding video frame of the markup information is It is the video frame corresponding to the search sequence input by user, and can determines the corresponding video of the video frame in turn.For example, In step S202, video searching apparatus 1 according to the attribute of at least one video frame label i.e. can determine it is corresponding extremely A few video.Wherein, at least one video frame corresponding with search sequence matching can belong to same to regard The video frame of frequency can also be the video frame for belonging to different video；At least one video frame can be continuous video frame, It can also be discontinuous video frame.Wherein, include but not limited to the identification of the corresponding information of each video frame in video： Scene Recognition based on independent frame image, is based on the corresponding audio-frequency information of image at the scene Recognition based on continuous multiple frames image Identification, the identification based on the corresponding caption information of image.

For example, the 1000th video frame of video V1 is the yellow picture for knowing bright eat at home, which also illustrates Huang Know the bright content for eating fried bean sauce, or the yellow picture for knowing bright parent, then after information in the video frame being identified, video frame rope May include in drawing：Attribute labeled as video V1 the 1000th video frame respectively with " Huang know bright eat at home ", " Huang dawn is bright to eat The correspondence of fried bean sauce " and " Huang knows bright parent ".If it includes that Huang Xiaoming exists that the 1000-1049 video frame of video V1, which is all, Family has a meal, and Huang dawn is bright, and eat is fried bean sauce, and the yellow picture for knowing bright parent three, but in the 1025-1049 video frame There is not yellow dawn bright parent, then in video frame index, each video in the 1000-1024 video frame of video V1 Frame is all respectively corresponding " Huang knows bright eat at home ", " Huang dawn bright eat fried bean sauce " and " yellow dawn bright parent "；The 1025- of video V1 Each video frame in 1049 video frame corresponding respectively " Huang knows bright eat at home " and " yellow know bright eat fried bean sauce ".If searching The video database of rope video includes video V1, and video V2, video V3 ... wait multiple videos, can establish including the data Total video frame index of video frame in library in all videos, or it only includes its video that each video, which establishes one, The video frame index of frame, when for example, video V1 establishes a corresponding video frame index, in step S202, video search dress It is that the search sequence matches corresponding at least one video frame in the corresponding video frame index of each video successively to set 1.

If the search sequence that in step s 201, video searching apparatus 1 receives is " Huang knows bright eat at home ", in step In S202, the mode of fuzzy matching may be used in video searching apparatus 1, in the video frame index inquiry by " Huang Xiaoming ", " Huang knows bright have a meal ", " Huang knows bright be in ", " Huang knows bright eat at home ", " star's eat at home ", " Huang Xiaoming and angelababy The markup information of the keywords such as eat at home " hit, the corresponding video frame of the markup information of hit is and " Huang Xiaoming exists Family has a meal " corresponding to video frame, and can determine the corresponding video of the video frame in turn；Or accurate matching way is used, Search corresponds to the video frame of " Huang knows bright eat at home " and its corresponding video only in the video frame index；Or it uses The mode that fuzzy matching and accurately matching is combined is that user matches corresponding video frame and video.If the quilt in video frame index The markup information of " Huang knows bright eat at home " this search sequence hit is 54, and in step S202, video searching apparatus 1 is true Surely 54 markup informations hit correspond respectively in video frame index：The 1000-1049 video frame of video V1, video The 225th video frame of V3 and the 25th frame, the 126th frame, the 127th frame of video V8, then it is corresponding with " Huang knows bright eat at home " Video frame include video V1 the 1000-1049 video frame, the 225th video frame of video V3 and the of video V8 25,126,127 video frame, video corresponding with " Huang knows bright eat at home " is respectively video V1, V3 and V8.

Here, in step S202, video searching apparatus 1 determines regard corresponding with search sequence according to video frame index Frequently, wherein the video frame index is established or updated according to the identification of the corresponding information of each video frame in video , therefore, in step S202, video searching apparatus 1 can be according to video frame index, to each video frame in video Information carries out retrieval matching, determines the information for being wherein queried sequence hits, is determined according to the described information of hit and inquires sequence Corresponding video frame is arranged, so that it is determined that video corresponding with the search sequence, is supplied to user.Content based on video, i.e., The information of each video frame scans in video so that user can more accurately search oneself and want regarding for viewing Frequently, the search experience of user is improved.

In step S203, it is right to position its according to the corresponding timestamp of at least one video frame for video searching apparatus 1 At least one video is supplied to the user by the play position at least one video answered with the play position.Tool Body, each video frame in video has its corresponding timestamp in the video.In step S202, video searching apparatus After 1 determines at least one video frame corresponding with search sequence, in step S203, video searching apparatus 1 obtains wherein Timestamp of each video frame in corresponding video, then will be corresponding at least one video frame The play position of video is positioned to the position with the matched video frame of search sequence, so that user chooses the video to watch When, the video since at video frame corresponding with search sequence play.Video searching apparatus 1 is so that user can be with direct viewing It wants the video clip seen.

Preferably, it is selected from least one video to obtain the user in step S203 for video searching apparatus 1 The target video selected, if the target video includes multiple and the matched video frame of the search sequence, by following Any one positions the play position of the target video：Acquiescence is according to first in the target video and the search sequence The timestamp of the video frame matched determines；It is selected by the user.Specifically, user provides at least from video searching apparatus 1 When choosing a target video to be watched in one video, in step S203, video searching apparatus 1 obtains the institute that user chooses State target video.If the target video includes multiple and the matched video frame of the search sequence,：It can give tacit consent to this The play position of video positions the position of that video frame to timestamp minimum, the i.e. position of the earliest video frame of reproduction time It sets, positioning can also be given tacit consent to the highest video frame of matching degree；By the play position random position of the video to any one Matched video frame；Or selecting play position to position to which video frame by user, the method for selection has tabular form and broadcasts Progress bar labeling form is put, wherein the tabular form includes and includes not only pop-up list, for example, a video is selected by user In after, in step S203, video searching apparatus 1 provides a choice box to the user in the form of pop-up list, is selected by user It selects and plays since which corresponding video frame.Then in step S203, video searching apparatus 1 will located play position At least one video is supplied to user so which video no matter user therefrom select watch, which all can be straight Connect since with from the matched video frame of the search sequence play, thus user can be with direct viewing and the search sequence pair The picture answered.

Here, sequence not only can be searched for and be inquired to video searching apparatus 1 based on the information of each video frame in video Row match corresponding video, are accurately found for user and meet the video that user watches condition, and true in video searching apparatus 1 After fixed corresponding video, in step S203, video searching apparatus 1 can accurately by the play position of the video position to The corresponding position of search sequence makes the video be played out from video frame corresponding with search sequence, so that user can More quickly to watch video clip corresponding with the search sequence, the search efficiency of user is improved, use is reduced The operating time at family.Therefore, video searching apparatus 1 can not only accurately search for user wants to watch video, moreover it is possible to by video Play position accurately positioning to user think viewing video-frequency band position.

Preferably, video searching apparatus 1 further includes step S204 (not shown) and step S205 (not shown).

In step S204, video searching apparatus 1 knows the corresponding information of each video frame in each video Not, corresponding markup information is obtained.Specifically, in step S204, video searching apparatus 1 is respectively to every in each video One video frame carries out image recognition, and knows to the corresponding pertinent audio information of each video frame or caption information Not, corresponding at least one markup information is obtained, wherein the markup information is plucking for the content included by corresponding video frame It wants.For example, the 1000th video frame of video V1 is the yellow picture for knowing bright eat at home, the video frame further include Huang Xiaoming eat it is fried The content in sauce face or the parent of Huang Xiaoming, therefore, the 1000th video frame of video V1 respectively with markup information：Huang Xiaoming Eat at home, yellow dawn be bright to eat fried bean sauce, yellow know bright parent and correspond to.

In step S205, video searching apparatus 1 establishs or updates the video frame index according to the markup information, Wherein, corresponding at least one video frame is matched according to the search sequence in determining device and its corresponding at least one regarded The mode of frequency includes：According to the search sequence, is matched in the video frame index and determine have and the search sequence phase At least one video frame of matched markup information and its corresponding at least one video.Specifically, in step S205, video Searcher 1 establishes video frame rope according to the correspondence between the corresponding markup information of each video frame in video Draw.Wherein, video frame index includes：Each video frame distinguishes the corresponding pass of corresponding at least one markup information System, so that video searching apparatus 1 matches search sequence with the markup information in video frame index, if successful match, The corresponding video frame of the markup information be with the matched video frame of search sequence, then video searching apparatus 1 regarded according to each The attribute label of frequency frame just can determine that the video where the video frame.Wherein, the video frame index can be according to each The general index that each video frame of video is established, can also be to be built respectively according to each video frame in any one video A vertical subindex.When it includes the general index of each video frame in each video that the video frame index, which is, After video searching apparatus 1 receives the search sequence of user, video searching apparatus 1 is according to the search sequence in the video frame index It is matched, its corresponding video is then determined according to matched at least one video frame respectively；If the video frame is At least one subindex corresponding with each video, video searching apparatus 1 is according to the search sequence of reception successively each It is matched in the corresponding subindex of video, when matching obtains corresponding video frame, video searching apparatus 1 can directly be known Video corresponding with the video frame.

For example, the 48th video frame of video V112 is pictures of the Zhang Jie in singing, by the audio-frequency information of the video frame, The poster information occurred on caption information and the video frame, in step S204, video searching apparatus 1 may recognize that this is regarded The content of frequency frame further includes：Zhang Jie sings in Chinese new song, and Zhang Jie sings the information such as inverse war, thus will " Zhang Jie is singing ", " open Outstanding person sings in Chinese new song ", the markup information of " Zhang Jie sings inverse war " as the video frame.Other videos in obtaining the video After the markup information of the video frame of the markup information of frame and other videos, in step S205, video searching apparatus 1 establishes packet Include the video frame index of the correspondence of video frame and markup information.If the video frame index is total video frame index, Including：Attribute labeled as video V112 the 48th video frame respectively with markup information " Zhang Jie is singing ", " Zhang Jie is in The new song singing of state ", " Zhang Jie sings inverse war " are corresponding；If the video frame be corresponding to each video subindex, depending on The corresponding video frame index of frequency V112 includes：112nd video frame respectively with markup information " Zhang Jie is singing ", " Zhang Jie exists Chinese new song singing ", " Zhang Jie sings inverse war " are corresponding.Wherein, the markup information of each video frame can have multiple, with never Tongfang is marked comprehensively in face of the video frame, so that the video frame is retrieved or is matched to by easier.

Preferably, video searching apparatus 1 further includes step S206 (not shown) and step S207 (not shown).

In step S206, video searching apparatus 1 knows the corresponding information of each video frame in each video Not, corresponding markup information is obtained；In step S207, video searching apparatus 1 is each video according to the markup information Establish a subindex；Wherein, corresponding at least one video frame and its corresponding at least one are matched according to the search sequence The mode of a video includes：According to the search sequence, matched in the corresponding each subindex of each video successively true Fixed at least one video frame with the markup information to match with the search sequence.Here, step S206 and aforementioned S204 Realization process it is identical.Specifically, in step S206, video searching apparatus 1 obtains markup information corresponding with video frame, In step S207, video searching apparatus 1 establishes a subindex according to the markup information, for each video, in step In S201, after video searching apparatus 1 receives the search sequence of user, in step S202, video searching apparatus 1 successively will be each It is corresponding that markup information in video corresponding video frame index and the search sequence carry out matching, and according to it is described The corresponding at least one markup information of search sequence matching determines and the corresponding video of at least one markup information institute Frame.

Preferably, it can be provided to the user and be searched in video in the video of arbitrary video-frequency band based on video searching apparatus 1 Function of search so that user when watching video by the search of the keyword inputted to it, can be directly by play position It positions to it and thinks the video frame position of viewing.If user selects one from least one video that video searching apparatus 1 provides Video plays out viewing or user arbitrarily clicks a video and plays out, and user can regard according to the video is corresponding Frequency frame index, i.e. subindex, input user thinks the arbitrary video frame or video of viewing in function of search input frame in video Markup information corresponding to segment, so that video searching apparatus 1 scans in the video.It is corresponded to obtaining the markup information Video frame after, video searching apparatus 1 positions play position to video frame corresponding with the markup information.Wherein, if When video frame corresponding with the markup information is multiple, in step S203, the acquiescence of video searching apparatus 1 positions play position To the video frame of its timestamp minimum, that is, it is at the video frame of earliest play position.For example, video V27 is film video 《Ice and snow unusual relationship》Segment, there is video frame index corresponding with video V27, wherein video frame index includes the 285th and regards The correspondence of frequency frame and markup information " ice and snow queen, which enchants, builds ice and snow palace ".User is in the corresponding search of video V27 " ice and snow palace " this search sequence is inputted in function input frame, in step S202, video searching apparatus 1 can be that user matches " ice and snow queen, which enchants, builds ice and snow palace " this markup information, then determines that its corresponding video frame is according to the markup information 285th video frame, in step S203, video searching apparatus 1 is according to the timestamp of the 285th video frame, by play position Whether positioning is requried the users and is played since the position to playing the position of the 285th video frame.If user agrees to, Directly from " ice and snow queen, which enchants, builds ice and snow palace ", this video frame starts to play the video.Here, being filled by video search 1 realization is set to pull manually without user, just can by search for just directly positioning to think viewing position play out.

Preferably, video searching apparatus 1 further includes step S208 (not shown).

In step S208, video searching apparatus 1 is ranked up at least one video, after being sorted at least One video；Wherein, in step S203, video searching apparatus 1 is used for：According at least one video frame, it is right to position its The play position of at least one video after the sequence answered is regarded at least one after the sequence with the play position Frequency is supplied to the user.Specifically, the mode being ranked up at least one video includes but not limited to：

The present invention also provides a kind of computer readable storage medium, the computer-readable recording medium storage has calculating Machine code, when the computer code is performed, any one of them video searching method as above is performed.

The present invention also provides a kind of computer program products, when the computer program product is executed by computer equipment When, any one of them video searching method as above is performed.

The present invention also provides a kind of computer equipment, the computer equipment includes memory and processor, described to deposit Computer code is stored in reservoir, the processor is configured to as above any to execute by executing the computer code Video searching method described in.

Fig. 3 shows the block diagram of the exemplary computer system/server 12 suitable for being used for realizing embodiment of the present invention. The computer system/server 12 that Fig. 3 is shown is only an example, should not be to the function and use scope of the embodiment of the present invention Bring any restrictions.

As shown in figure 3, computer system/server 12 is showed in the form of universal computing device.Computer system/service The component of device 12 can include but is not limited to：One or more processor or processing unit 16, system storage 28, connection The bus 18 of different system component (including system storage 28 and processing unit 16).

Bus 18 indicates one or more in a few class bus structures, including memory bus or Memory Controller, Peripheral bus, graphics acceleration port, processor or the local bus using the arbitrary bus structures in a variety of bus structures.It lifts For example, these architectures include but not limited to industry standard architecture (ISA) bus, microchannel architecture (MAC) Bus, enhanced isa bus, Video Electronics Standards Association (VESA) local bus and peripheral component interconnection (PCI) bus.

Computer system/server 12 typically comprises a variety of computer system readable media.These media can be appointed What usable medium that can be accessed by computer system/server 12, including volatile and non-volatile media, it is moveable and Immovable medium.

Memory 28 may include the computer system readable media of form of volatile memory, such as random access memory Device (RAM) 30 and/or cache memory 32.Computer system/server 12 may further include it is other it is removable/no Movably, volatile/non-volatile computer system storage medium.Only as an example, storage system 34 can be used for reading and writing Immovable, non-volatile magnetic media (Fig. 3 is not shown, commonly referred to as " hard disk drive ").It, can although being not shown in Fig. 3 To provide for the disc driver to moving non-volatile magnetic disk (such as " floppy disk ") read-write, and to removable non-volatile Property CD (such as CD-ROM, DVD-ROM or other optical mediums) read and write CD drive.In these cases, each to drive Dynamic device can be connected by one or more data media interfaces with bus 18.Memory 28 may include at least one program There is one group of (for example, at least one) program module, these program modules to be configured to perform the present invention for product, the program product The function of each embodiment.

Program/utility 40 with one group of (at least one) program module 42 can be stored in such as memory 28 In, such program module 42 includes --- but being not limited to --- operating system, one or more application program, other programs Module and program data may include the realization of network environment in each or certain combination in these examples.Program mould Block 42 usually executes function and/or method in embodiment described in the invention.

Computer system/server 12 can also be (such as keyboard, sensing equipment, aobvious with one or more external equipments 14 Show device 24 etc.) communication, it is logical that the equipment interacted with the computer system/server 12 can be also enabled a user to one or more Letter, and/or any set with so that the computer system/server 12 communicated with one or more of the other computing device Standby (such as network interface card, modem etc.) communicates.This communication can be carried out by input/output (I/O) interface 22.And And computer system/server 12 can also pass through network adapter 20 and one or more network (such as LAN (LAN), wide area network (WAN) and/or public network, such as internet) communication.As shown, network adapter 20 passes through bus 18 communicate with other modules of computer system/server 12.It should be understood that although being not shown in Fig. 3, computer can be combined Systems/servers 12 use other hardware and/or software module, including but not limited to：Microcode, device driver, at redundancy Manage unit, external disk drive array, RAID system, tape drive and data backup storage system etc..

Processing unit 16 is stored in the program in memory 28 by operation, to perform various functions application and data Processing.

For example, being stored with the computer program of the various functions and processing for executing the present invention, processing in memory 28 When unit 16 executes corresponding computer program, the identification that the present invention is intended to incoming call in network-side is implemented.

It should be noted that the present invention can be carried out in the assembly of software and/or software and hardware, for example, this hair Application-specific integrated circuit (ASIC) can be used in bright each device or any other is realized similar to hardware device.In one embodiment In, software program of the invention can be executed by processor to realize steps described above or function.Similarly, of the invention Software program (including relevant data structure) can be stored in computer readable recording medium storing program for performing, for example, RAM memory, Magnetic or optical driver or floppy disc and similar devices.In addition, hardware can be used to realize in some steps or function of the present invention, example Such as, coordinate to execute the circuit of each step or function as with processor.

It is obvious to a person skilled in the art that invention is not limited to the details of the above exemplary embodiments, Er Qie In the case of without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power Profit requires rather than above description limits, it is intended that all by what is fallen within the meaning and scope of the equivalent requirements of the claims Variation includes within the present invention.Any reference signs in the claims should not be construed as limiting the involved claims.This Outside, it is clear that one word of " comprising " is not excluded for other units or step, and odd number is not excluded for plural number.That is stated in system claims is multiple Unit or device can also be realized by a unit or device by software or hardware.The first, the second equal words are used for table Show title, and does not represent any particular order.

Claims

1. a kind of video searching method based on video content, wherein the video searching method includes：

A. search sequence input by user is received；

B. according to video frame index determine including at least one video frame corresponding with search sequence matching and its Corresponding at least one video, wherein the video frame index is according to the knowledge to the corresponding information of each video frame in video It is other to be establishd or updated；

C. according to the corresponding timestamp of at least one video frame, the play position of its corresponding at least one video is positioned, At least one video is supplied to the user with the play position.

2. video searching method according to claim 1, wherein the video searching method further includes：

Wherein, corresponding at least one video frame is matched according to the search sequence in step b and its corresponding at least one regarded The mode of frequency includes：

According to the search sequence, the mark for determining and having and matching with the search sequence is matched in the video frame index At least one video frame of information and its corresponding at least one video.

3. video searching method according to claim 1, wherein the video searching method further includes：

According to the search sequence, successively in the corresponding each subindex of each video match determine have looked into described Ask at least one video frame for the markup information that sequence matches.

4. video searching method according to claim 2 or 3, wherein the markup information passes through following at least any one It obtains：

The characteristics of image for identifying a video frame images, determines that one video frame images are corresponding according to described image feature Markup information；

Identify the characteristics of image of a video frame images, and according to the preposition image and postposition figure of the video frame images of identification The characteristics of image of picture determines corresponding markup information；

It identifies the audio-frequency information corresponding to a video frame, the audio-frequency information is converted into text information, according to the word Information determines the corresponding markup information of the video frame；

The caption information corresponding to a video frame is extracted, determines that the corresponding mark of the video frame is believed according to the caption information Breath.

5. video searching method according to any one of claim 1 to 4, wherein the video searching method further includes：

At least one video is ranked up, at least one video after being sorted；

Wherein, step c includes：

According at least one video frame, the play position of at least one video after its corresponding described sequence is positioned, with At least one video after the sequence is supplied to the user by the play position.

6. the video searching method according to any one of claim 5, wherein according to it is following any one of at least to it is described extremely A few video is ranked up：

The corresponding video distribution person's information of video；

The source-information of video；

The clarity of video；

The subject information of video；

Feedback information of the user to video.

7. video searching method according to any one of claim 1 to 6, wherein will with the play position in step c At least one video is supplied to the user to further include：

Obtain a target video of the user from selected at least one video；

If the target video includes multiple and the matched video frame of the search sequence, by any one of following to position State the play position of target video：

It is selected by the user.

8. a kind of video searching apparatus based on video content, wherein the video searching apparatus includes：

Reception device, for receiving search sequence input by user；

Determining device, for according to video frame index determine including it is corresponding at least one with search sequence matching Video frame and its corresponding at least one video, wherein the video frame index is corresponded to according to each video frame in video The identification of information establishd or updated；

Positioning device, for according to the corresponding timestamp of at least one video frame, positioning its corresponding at least one video Play position, at least one video is supplied to by the user with the play position.

9. video searching apparatus according to claim 8, wherein the video searching apparatus further includes：

First identification device obtains corresponding for the corresponding information of each video frame in each video to be identified Markup information；

Wherein, corresponding at least one video frame and its corresponding at least one are matched according to the search sequence in determining device The mode of a video includes：

10. video searching apparatus according to claim 8, wherein the video searching apparatus further includes：

Second identification device obtains corresponding for the corresponding information of each video frame in each video to be identified Markup information；

11. video searching apparatus according to claim 9 or 10, wherein the markup information passes through following at least any Item obtains：

Identify the characteristics of image of a video frame images, and according to the preposition image of one video frame images of identification with after The characteristics of image for setting image determines corresponding markup information；

12. the video searching apparatus according to any one of claim 8 to 11, wherein the video searching apparatus further includes：

Wherein, positioning device is used for：

13. the video searching apparatus according to any one of claim 12, wherein according to following at least any one to described At least one video is ranked up：

The corresponding video distribution person's information of video；

The source-information of video；

The clarity of video；

The subject information of video；

Feedback information of the user to video.

14. the video searching apparatus according to any one of claim 8 to 13, wherein the positioning device is additionally operable to：

Obtain a target video of the user from selected at least one video；

It is selected by the user.

15. a kind of computer readable storage medium, the computer-readable recording medium storage has computer code, when the meter Calculation machine code is performed, and the video searching method as described in any one of claim 1 to 7 is performed.

16. a kind of computer program product, when the computer program product is executed by computer equipment, such as claim 1 It is performed to the video searching method described in any one of 7.

17. a kind of computer equipment, the computer equipment include：

One or more processors；

Memory, for storing one or more computer programs；

When one or more of computer programs are executed by one or more of processors so that one or more of Processor realizes the video searching method as described in any one of claim 1 to 7.