CN102955858B - Search video file sorting method, system and server - Google Patents

Search video file sorting method, system and server Download PDF

Info

Publication number
CN102955858B
CN102955858B CN 201210448634 CN201210448634A CN102955858B CN 102955858 B CN102955858 B CN 102955858B CN 201210448634 CN201210448634 CN 201210448634 CN 201210448634 A CN201210448634 A CN 201210448634A CN 102955858 B CN102955858 B CN 102955858B
Authority
CN
Grant status
Grant
Patent type
Prior art keywords
video
object
server
plurality
file
Prior art date
Application number
CN 201210448634
Other languages
Chinese (zh)
Other versions
CN102955858A (en )
Inventor
邓雄书
Original Assignee
北京百度网讯科技有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Grant date

Links

Abstract

本发明提出一种视频文件的搜索排序方法、系统和服务器。 To provide a video file of the present invention is to sort the search methods, systems, and servers. 其中,所述方法包括:服务器对多个视频文件分别进行视频解码以获取每个视频文件的解码后的图像;服务器对解码后的图像进行图像识别以获取每个视频文件的多个物体对象的信息;服务器对每个视频文件中的每个物体对象的出现次数进行统计以生成每个视频文件对应的统计信息;服务器接收用户通过客户端发送的视频搜索请求;服务器根据视频搜索请求进行搜索以获得多个搜索结果;以及服务器根据统计信息对多个搜索结果进行排序。 Wherein said method comprises: a plurality of video file server each video decoding to obtain decoded image of each video file; server decoded image an image recognition subject to obtain a plurality of objects of each video file information; server number of occurrences of each object in each video object file to generate statistical information for each of the video files corresponding to the statistics; a video server receives a search request sent by a client user; video search server according to the search request obtaining a plurality of search results; and multiple server sort search results according to statistics. 根据本发明实施例的方法,可按用户兴趣对视频文件进行排序供用户选择,能够满足用户需求,便于用户根据统计信息对视频文件进行管理,方便,直观,用户体验好。 The method according to embodiments of the present invention, a user may be interested in the video file for the user to select the sort, to meet the needs of users, convenience of users based on the statistical information of the video file management, convenient, intuitive, user experience is good.

Description

视频文件的搜索排序方法、系统及服务器 Search video file sorting method, system and server

技术领域 FIELD

[0001]本发明涉及互联网技术领域,特别涉及一种视频文件的搜索排序方法、系统及服务器。 [0001] The present invention relates to the field of Internet technologies, and particularly relates to a video searching method for sorting documents, and server systems.

背景技术 Background technique

[0002]目前,通过文件浏览器、网络浏览器或视频播放应用搜索视频文件的方法,播放视频时只是对视频文件进行解码播放,并未对视频文件进行分析提取其内容信息,用户不能按照自己的兴趣对视频文件搜索结果进行选择,体验不佳。 [0002] Currently, the web browser or video playback application method of searching video files, playing video decode video files just by playing the file browser, not the video file is analyzed to extract its content information, the user can not follow their own interest in the search results to select video files, poor experience. 此外,由于目前的互联网提供了海量的视频数据,因此对于用户来讲,如何能够检索到自己感兴趣的视频文件成为比较困难的事情,往往需要下载多个视频文件,才能够找到。 In addition, due to the current Internet offers vast amounts of video data, so for users, how can retrieve video files you're interested in things become more difficult, often need to download multiple video files to be able to find. 因此,不仅浪费了用户的时间,也浪费了大量宝贵的流量资源。 Therefore, not only waste the user's time, but also waste a lot of valuable traffic resources. 特别是对于移动终端的用户来说,由于移动终端下载速度的限制,该问题更为明显。 Especially for the user of the mobile terminal, since the mobile terminal download speed limit, the problem is even more evident.

发明内容 SUMMARY

[0003]本发明的目的旨在至少解决所述技术缺陷之一。 [0003] The object of the present invention to solve at least one of the technical defects.

[0004]为此,本发明的第一个目的在于提出一种视频文件的搜索排序方法,能够满足用户需求,便于用户根据统计信息对视频文件进行管理,方便,直观,用户体验好。 [0004] For this purpose, a first object of the present invention is to propose a method of ordering video file searching, to meet the needs of users, convenience of users based on the statistical information of the video file management, convenient, intuitive, user experience is good.

[0005]本发明的第二个目的在于提出一种视频文件的搜索排序系统。 [0005] A second object of the present invention is to propose a system for ordering search video file.

[0006]本发明的第三个目的在于提出一种服务器。 [0006] A third object of the present invention is to provide a server.

[0007]为达到所述目的,本发明第一方面的实施例公开了一种视频文件的搜索排序方法,包括以下步骤:服务器对多个视频文件分别进行视频解码以获取每个视频文件的解码后的图像;所述服务器对所述解码后的图像进行图像识别以获取每个视频文件的多个物体对象的信息;所述服务器对每个视频文件中的每个物体对象的出现次数进行统计以生成每个视频文件对应的统计信息;所述服务器接收用户通过客户端发送的视频搜索请求;所述服务器根据所述视频搜索请求进行搜索以获得多个搜索结果,其中,所述多个搜索结果中至少部分搜索结果具有所述统计信息;以及所述服务器根据所述统计信息对所述多个搜索结果进行排序。 [0007] In order to achieve the above object, a first aspect of the embodiments of the present invention discloses a method of sorting the search video files, comprising the steps of: a plurality of video file server of the video decoder for each decoding of each video file image; and the image server for the decoded image to acquire information identifying a plurality of objects of each video object file; the server counting how often each object in each video object file to generate statistics corresponding to each video file; a video search server receiving a user via a request sent by a client; the server according to the search request to obtain a plurality of search video search results, wherein the plurality of search the results having search result at least in part the statistical information; and the server sorting the plurality of search results based on the statistical information.

[0008]根据本发明实施例的视频文件的搜索排序方法,通过对视频文件进行解码并对解码后的图像进行图像识别,获取视频文件中物体对象的统计信息,并将统计信息呈现给用户,从而可以按用户感兴趣的物体对象的统计信息对视频文件进行排序供用户选择,不仅能够满足用户需求,更快速且更准确地查找到其希望获得的视频文件,而且便于用户根据统计信息对视频文件进行管理,方便,直观,用户体验好。 [0008] The sorting method of searching video files embodiment of the present invention, the image file by decoding video image decoding and recognition, retrieve statistics objects in the video file object, and statistical information to the user, thus the video file according to the statistics of interest to the user object object sorting options for users, not only to meet customer needs, more quickly and accurately find the video files they wish to obtain, and easy for users based on statistical information on video file management, easy to use, intuitive, user experience is good.

[0009]为达到所述目的,本发明第二方面的实施例公开了一种视频文件的搜索排序系统,包括:客户端,用于接收用户输入的视频搜索请求,并将所述视频搜索请求发送至所述服务器;服务器,用于对多个视频文件分别进行视频解码以获取每个视频文件的解码后的图像,对所述解码后的图像进行图像识别以获取每个视频文件的多个物体对象的信息,并对每个视频文件中的每个物体对象的出现次数进行统计以生成每个视频文件对应的统计信息,以及根据所述客户端发送的搜索请求进行搜索以获取多个搜索结果,其中,所述多个搜索结果中至少部分搜索结果具有所述统计信息,并根据所述统计对所述多个搜索结果进tx排序。 [0009] In order to achieve the above object, an embodiment of the second aspect of the present invention discloses a system for ordering search video file, comprising: a client, a search request for video receiving user input, and the video search request send to the server; a plurality of video file for a plurality of video decoding to obtain respectively decoded images of each video file, an image of the decoded image recognition for each video file the object information of the object, and the number of occurrences of each object in each video object file to generate statistics for each video file corresponding to the statistical information, and searches based on the search request sent by the client to obtain a plurality of search the results, wherein the plurality of search results in the search result at least in part with the statistical information, and according to the statistics of the plurality of sorted search results into tx.

[0010]根据本发明实施例的视频文件的搜索排序系统,通过对视频文件进行解码并对解码后的图像进行图像识别,获取视频文件中物体对象的统计信息,并将统计信息呈现给用户,从而可以按用户感兴趣的物体对象的统计信息对视频文件进行排序供用户选择,不仅能够满足用户需求,更快速且更准确地查找到其希望获得的视频文件,而且便于用户根据统计信息对视频文件进行管理,方便,直观,用户体验好。 [0010] The search order of the video file system of the embodiment of the present invention, the image file by decoding video image decoding and recognition, retrieve statistics objects in the video file object, and statistical information to the user, thus the video file according to the statistics of interest to the user object object sorting options for users, not only to meet customer needs, more quickly and accurately find the video files they wish to obtain, and easy for users based on statistical information on video file management, easy to use, intuitive, user experience is good.

[0011]为达到所述目的,本发明第一方面的实施例公开了一种服务器,包括:视频解码模块,用于对多个视频文件分别进行视频解码以获取每个视频文件的解码后的图像;图像识别模块,用于对所述解码后的图像进行图像识别以获取每个视频文件的多个物体对象的信息;统计模块,用于对每个视频文件中的每个物体对象的出现次数进行统计以生成每个视频文件对应的统计信息;接收模块,用于接收用户通过客户端发送的视频搜索请求;搜索模块,用于根据视频搜索请求进行搜索以获得多个搜索结果,其中,所述多个搜索结果中至少部分搜索结果具有所述统计信息;以及排序模块,用于根据所述统计信息对所述多个搜索结果进行排序。 [0011] In order to achieve the above object, a first aspect of the embodiments of the present invention discloses a server, comprising: a video decoder module, a video file, respectively, a plurality of video decoder for decoding each of the video file image; an image recognition module, configured to image the decoded image to acquire information identifying a plurality of objects of each video file objects; statistics module for each object that appears for each object in the video file count the number of times to generate statistics corresponding to each video file; receiving means for receiving a video search request sent by the user via the client; search module, a search request for video searching according to obtain a plurality of search results, wherein the plurality of search results in the search result at least in part with the statistical information; and a sorting means for sorting the plurality of search results based on the statistical information.

[0012]根据本发明实施例的服务器,通过对视频文件进行解码并对解码后的图像进行图像识别,获取视频文件中物体对象的统计信息,并将统计信息呈现给用户,从而可以按用户感兴趣的物体对象的统计信息对视频文件进行排序供用户选择,不仅能够满足用户需求,更快速且更准确地查找到其希望获得的视频文件,而且便于用户根据统计信息对视频文件进行管理,方便,直观,用户体验好。 [0012] server according to an embodiment of the present invention, an image decoded by the video decoding and image identification file, get statistics object video object file, and statistical information to the user, whereby the user can sense by statistical information of the object of interest in the subject video file sorting options for users, not only to meet customer needs, more quickly and accurately find the video files they wish to obtain, and easy for users to manage video files according to statistics, convenient intuitive, user experience is good.

[0013]本发明附加的方面和优点将在下面的描述中部分给出,部分将从下面的描述中变得明显,或通过本发明的实践了解到。 [0013] This additional aspects and advantages of the invention will be set forth in part in the description which follows, from the following description in part be apparent from, or learned by practice of the present invention.

附图说明 BRIEF DESCRIPTION

[0014]本发明所述的和/或附加的方面和优点从下面结合附图对实施例的描述中将变得明显和容易理解,其中: [0014] according to the present invention and / or additional aspects and advantages from the following description of the embodiments will become apparent and more readily appreciated in conjunction with the accompanying drawings, wherein:

[0015]图1为根据本发明一个实施例的视频文件的搜索排序方法的流程图; [0015] FIG. 1 is a flowchart of a method of ordering video file searching according to one embodiment of the present invention;

[0016]图2为根据本发明又一个实施例的视频文件的搜索排序方法的流程图; [0016] FIG 2 is a flowchart of a method of searching video files ordered yet another embodiment of the present invention;

[0017]图3为根据本发明一个实施例的视频文件的搜索排序系统的结构示意图;以及 [0017] FIG. 3 is a schematic structural diagram of a video file the search sorting system according to one embodiment of the present invention; and

[0018]图4为根据本发明一个实施例的服务器的结构示意图。 [0018] FIG. 4 is a schematic structure of a server according to an embodiment of the present invention.

具体实施方式 detailed description

[0019]下面详细描述本发明的实施例,所述实施例的示例在附图中示出,其中自始至终相同或类似的标号表示相同或类似的元件或具有相同或类似功能的元件。 [0019] Example embodiments of the present invention is described in detail below, exemplary embodiments of the embodiment shown in the accompanying drawings, wherein same or similar reference numerals designate the same or similar elements or elements having the same or similar functions. 下面通过参考附图描述的实施例是示例性的,仅用于解释本发明,而不能解释为对本发明的限制。 By following with reference to the embodiments described are exemplary only for explaining the present invention and should not be construed as limiting the present invention. 相反,本发明的实施例包括落入所附加权利要求书的精神和内涵范围内的所有变化、修改和等同物。 In contrast, embodiments of the present invention includes all variations that fall within the appended claims the spirit and terms, modifications and equivalents thereof.

[0020]在本发明的描述中,除非另有说明,“多个”的含义是两个或两个以上。 [0020] In the present description, unless otherwise specified, the meaning of "more" is two or more. 流程图中或在此以其他方式描述的任何过程或方法描述可以被理解为,表示包括一个或更多个用于实现特定逻辑功能或过程的步骤的可执行指令的代码的模块、片段或部分,并且本发明的优选实施方式的范围包括另外的实现,其中可以不按所示出或讨论的顺序,包括根据所涉及的功能按基本同时的方式或按相反的顺序,来执行功能,这应被本发明的实施例所属技术领域的技术人员所理解。 In the flowchart in any process or method or otherwise described in this description may be understood as representing the code comprises one or more steps for implementing specific logical functions or processes executable instructions module, segment, or portion , and the scope of the preferred embodiment of the present invention includes other implementations, which may not be according to the sequence shown or discussed, including substantially simultaneous manner or in reverse order, depending upon the functionality to perform the functions involved, it should be Example embodiments are understood skilled in the art of the present invention.

[0021]下面参考附图描述根据本发明实施例的视频文件的搜索排序方法、系统和服务器。 [0021] The following figures described sorting method of searching video files embodiment of the present invention, the reference system and the server.

[0022]本发明实施例提出的一种视频文件的搜索排序方法,包括以下步骤:服务器对多个视频文件分别进行视频解码以获取每个视频文件的解码后的图像;服务器对解码后的图像进行图像识别以获取每个视频文件的多个物体对象的信息;服务器对每个视频文件中的每个物体对象的出现次数进行统计以生成每个视频文件对应的统计信息;服务器接收用户通过客户端发送的视频搜索请求;服务器根据视频搜索请求进行搜索以获得多个搜索结果,其中,多个搜索结果中至少部分搜索结果具有统计信息;以及服务器根据统计信息对多个搜索结果进行排序。 [0022] A video file searching sorting method according to the embodiment of the present invention is presented, comprising the steps of: a plurality of video file server each video decoding to obtain decoded image of each video file; decoded image server identification information of an image to obtain a plurality of objects of each video object file; server number of times each object in each video object file to generate statistics statistics for each corresponding video file; by the client server receiving a user sends video search request; the video search server search request to obtain a plurality of search results, wherein the plurality of search results, at least portion of search results with statistical information; and a server sorts the search results according to a plurality of statistical information.

[0023]图1为根据本发明一个实施例的视频文件的搜索排序方法的流程图。 [0023] FIG. 1 is a flowchart of a method of ordering video file searching according to one embodiment of the present invention.

[0024]如图1所示,根据本发明实施例的视频文件的搜索排序方法包括下述步骤。 [0024] As shown in FIG 1, comprising the steps of a method of ordering video file searching an embodiment of the present invention.

[0025]步骤SlOl,服务器对多个视频文件分别进行视频解码以获取每个视频文件的解码后的图像。 [0025] Step SlOl, the video file server of the plurality of decoded videos to obtain respectively decoded images of each video file.

[0026]具体地,服务器对多个视频文件分别进行解码,获取每个视频文件解码后的多帧原始图像,以供服务器进行图像识别,其中原始图像一般为YUV颜色格式或者RGB颜色格式。 [0026] Specifically, the video file server of the plurality of decoded separately to obtain multiple images of each frame of the original video file decoding, for the image recognition server, wherein the original image is typically a YUV color format or RGB color format. 在本发明的实施例中,该步骤为线下执行,服务器对其提供的视频文件进行分析。 In an embodiment of the present invention, this step is performed offline, a video file server to provide analyzed. 在本发明的其他实施例中,也可以是其他服务器对多个视频文件进行分析,之后将分析结果反馈给该服务器。 In other embodiments of the present invention, it may be other servers for analysis of a plurality of video files, then the analysis results back to the server.

[0027]步骤S102,服务器对解码后的图像进行图像识别以获取每个视频文件的多个物体对象的信息。 [0027] step S102, the decoded image server the image to acquire information identifying a plurality of objects of each video object file. 其中,多个物体对象可以包括人物,动物和/或建筑物等人或物。 Wherein the plurality of objects may include objects or things people, animals and / or other buildings.

[0028]具体地,服务器可通过人脸识别技术对获取的多帧原始图像进行识别,也可以通过其他能够识别出图像中物体对象信息的图像识别技术对获取的多帧原始图像进行识别。 [0028] Specifically, the server may identify a plurality of frames of the original image acquired by the face recognition technology, recognition may be performed on the plurality of frames of the original image obtained by other image recognition technology can recognize the target object in the image information. 当然在本发明的其他实施例中,还可对动物或者建筑物等进行识别。 Of course, in other embodiments of the present invention, an animal or a building and the like may also be identified.

[0029] 具体的识别算法可为SURF算法、harr及adboost算法等。 [0029] Specific recognition algorithm may be as the SURF algorithm, harr and adboost algorithm.

[0030]步骤S103,服务器对每个视频文件中的每个物体对象的出现次数进行统计以生成每个视频文件对应的统计信息,并将统计信息保存在数据库之中。 [0030] step S103, the server counting how often each object in each video object file to generate statistical information corresponding to the video file, and the statistical information stored in the database.

[0031]具体地,在本发明的实施例中,用户可通过多种方式输入其希望获得视频文件,例如通过文字输入或者图形输入,甚至还可在已有的视频文件中对该视频文件中的物体对象进行标注。 [0031] Specifically, in the embodiment of the present invention, the user can input a variety of ways to obtain their desired video file, such as by text input or drawing input, and even video files in the video file in the existing the object label object. 例如,服务器可提示用户对识别出的物体对象进行标注,用户接收到提示后为其标注相关信息后,服务器对识别出的物体对象的出现次数进行统计。 For example, the server may prompt the user for the identified target object tagging, the user receives a prompt after the label information, the server of the number of occurrences of the identified target object for statistics.

[0032]例如,若识别出的物体对象为刘德华或张曼玉等人物,服务器提示用户标注后用户可为识别出的物体对象标注上刘德华或张曼玉等人物信息,若识别出的物体对象为猫、狗、大楼或汽车等,服务器提示用户标注后用户可为识别出的物体对象标注上猫、狗、大楼或汽车等信息。 [0032] For example, if the recognized object object like Andy or Maggie figures, the server prompts the user denoted by a user may be shown as identifiable object target label on the like Andy or Maggie character information, if the identified object target is a cat, a dog and so on, building or car, the server prompts the user to identify a user can mark objects object Notation information on cats, dogs, buildings or other car. 在对下一帧图像进行识别时,若识别出与已标注的物体对象相同的物体对象,服务器自动为其添加标注,该物体对象的出现次数+1;若识别出与已标注的物体对象不同的物体对象,服务器则提示用户对其进行标注,该物体对象的出现次数记为I。 At the time of recognition of the next frame, and the object if the object has the same object is denoted by the recognized object, the server automatically to add annotation +1 the number of occurrences of the target object; if the object identified with the object is different from that marked objects object server then prompts the user to be marked, the number of occurrences of the object object is recorded as I.

[0033]步骤S104,服务器接收用户通过客户端发送的视频搜索请求。 [0033] step S104, the video server receives a search request transmitted by the client user.

[0034]其中,客户端可以为个人电脑、手机、平板电脑等。 [0034] in which the client may be a personal computer, phone, tablet, and so on.

[0035]具体地,用户可通过客户端100中的文件浏览器或者网络浏览器输入视频搜索请求。 [0035] In particular, the user can through the client file browser 100 or web browser, a video input terminal search request. 例如,若用户需要搜索刘德华主演的影片,可通过客户端向服务器发送“刘德华主演的影片”的搜索请求。 For example, if the user needs to search movie starring Andy Lau, you can send "starring Andy Lau" search request by the client to the server.

[0036]步骤S105,服务器根据视频搜索请求进行搜索以获得多个搜索结果,其中,多个搜索结果中至少部分搜索结果具有统计信息。 [0036] step S105, the server searches based on the search request to obtain a plurality of video search results, wherein the plurality of search results, at least portion of search results with statistical information.

[0037]步骤S106,服务器根据统计信息对多个搜索结果进行排序。 [0037] step S106, the server sorts the search results based on the plurality of statistics.

[0038]例如,若用户需要搜索刘德华主演的影片,服务器可根据“刘德华出现次数”对搜索结果进行排序,刘德华出现次数最多的影片即为刘德华主演的影片。 [0038] For example, if the user needs to search movie starring Andy Lau, the server can be to sort search results based on "the number of times Andy Lau appeared," the film is the highest number of films starring Andy Lau appeared.

[0039]根据本发明实施例的视频文件的搜索排序方法,通过对视频文件进行解码并对解码后的图像进行图像识别,获取视频文件中物体对象的统计信息,并将统计信息呈现给用户,从而可以按用户感兴趣的物体对象的统计信息对视频文件进行排序供用户选择,不仅能够满足用户需求,更快速且更准确地查找到其希望获得的视频文件,而且便于用户根据统计信息对视频文件进行管理,方便,直观,用户体验好。 [0039] The sorting method of searching video files embodiment of the present invention, the image file by decoding video image decoding and recognition, retrieve statistics objects in the video file object, and statistical information to the user, thus the video file according to the statistics of interest to the user object object sorting options for users, not only to meet customer needs, more quickly and accurately find the video files they wish to obtain, and easy for users based on statistical information on video file management, easy to use, intuitive, user experience is good.

[0040]在本发明的一个实施例中,当用户播放视频文件时,还可以在播放前或播放后将视频文件中的物体对象的统计信息呈现给用户,例如,呈现出现哪些演员、演员出现次数以及出现哪些动物、建筑物等,可以吸引用户注意并且方便用户选择所需资源,进一步提高用户体验。 [0040] In one embodiment of the present invention, when a user plays a video file, may also be present in the statistics before playing object or objects after playing video file to a user, e.g., presenting what appears actor, actress appears times and what animals, buildings, etc. appear, to attract the user's attention and easy to access the resources needed to further improve the user experience.

[0041]图2为根据本发明又一个实施例的视频文件的搜索排序方法的流程图。 [0041] FIG 2 is a flowchart of a method of searching video files ordered yet another embodiment of the present invention. 在以上实施例中以物体对象的出现次数进行统计,在本发明的实施例中还可对物体对象的出现时间进行统计。 In the above embodiment, the object to the number of occurrences statistics object, the object may be subject to the appearance time statistics in the embodiment of the present invention.

[0042]如图2所示,根据本发明实施例的视频文件的搜索排序方法包括下述步骤。 [0042] As shown in FIG 2, comprising the steps of a method of searching video files sorting embodiment of the present invention.

[0043]步骤S201,服务器对多个视频文件分别进行视频解码以获取每个视频文件的解码后的图像。 [0043] step S201, the video file server of the plurality of video decoding to obtain respectively decoded images of each video file.

[0044]具体地,服务器对多个视频文件分别进行解码,获取每个视频文件解码后的多帧原始图像,以供服务器进行图像识别,其中原始图像一般为YUV颜色格式或者RGB颜色格式。 [0044] Specifically, the video file server of the plurality of decoded separately to obtain multiple images of each frame of the original video file decoding, for the image recognition server, wherein the original image is typically a YUV color format or RGB color format. 在本发明的实施例中,该步骤为线下执行,服务器对其提供的视频文件进行分析。 In an embodiment of the present invention, this step is performed offline, a video file server to provide analyzed. 在本发明的其他实施例中,也可以是其他服务器对多个视频文件进行分析,之后将分析结果反馈给该服务器。 In other embodiments of the present invention, it may be other servers for analysis of a plurality of video files, then the analysis results back to the server.

[0045]步骤S202,服务器对解码后的图像进行图像识别以获取每个视频文件的多个物体对象的信息。 [0045] Step S202, the decoded image server the image to acquire information identifying a plurality of objects of each video object file. 其中,多个物体对象可以包括人物,动物和/或建筑物等人或物。 Wherein the plurality of objects may include objects or things people, animals and / or other buildings.

[0046]具体地,服务器可通过人脸识别技术对获取的多帧原始图像进行识别,也可以通过其他能够识别出图像中物体对象信息的图像识别技术对获取的多帧原始图像进行识别。 [0046] Specifically, the server may identify a plurality of frames of the original image acquired by the face recognition technology, recognition may be performed on the plurality of frames of the original image obtained by other image recognition technology can recognize the target object in the image information. 当然在本发明的其他实施例中,还可对动物或者建筑物等进行识别。 Of course, in other embodiments of the present invention, an animal or a building and the like may also be identified. 具体的识别算法可为SURF算法、harr及adboost算法等。 Specific recognition algorithm may be as the SURF algorithm, harr and adboost algorithm.

[0047]步骤S203,服务器对每个视频文件中的每个物体对象的出现次数及出现时间进行统计以生成每个视频文件对应的统计信息,并将统计信息保存在数据库之中。 [0047] step S203, the server counting how often each object in each video object files and the time of occurrence to generate statistics corresponding to each video file, and the statistical information stored in the database.

[0048]其中,统计信息包括每个物体对象的出现次数信息和出现时间信息。 [0048] wherein the statistical information includes information on the number of occurrences of each target object and time information appears.

[0049]具体地,在本发明的实施例中,用户可通过多种方式输入其希望获得视频文件,例如通过文字输入或者图形输入,甚至还可在已有的视频文件中对该视频文件中的物体对象进行标注。 [0049] Specifically, in the embodiment of the present invention, the user can input a variety of ways to obtain their desired video file, such as by text input or drawing input, and even video files in the video file in the existing the object label object. 例如,服务器可提示用户对识别出的物体对象进行标注,用户接收到提示后为其标注相关信息后,服务器对识别出的物体对象的出现次数及出现时间进行统计。 After labeling for related information for example, the server may prompt the user for the identified target object tagging, the user receives a prompt, the number of occurrences of the identified server object and object occurrence time statistics.

[0050]例如,若识别出的物体对象为刘德华或张曼玉等人物,服务器提示用户标注后用户可为识别出的物体对象标注上刘德华或张曼玉等人物信息,若识别出的物体对象为猫、狗、大楼或汽车等,服务器提示用户标注后用户可为识别出的物体对象标注上猫、狗、大楼或汽车等信息。 [0050] For example, if the recognized object object like Andy or Maggie figures, the server prompts the user denoted by a user may be shown as identifiable object target label on the like Andy or Maggie character information, if the identified object target is a cat, a dog and so on, building or car, the server prompts the user to identify a user can mark objects object Notation information on cats, dogs, buildings or other car. 在对下一帧图像进行识别时,若识别出与已标注的物体对象相同的物体对象,服务器自动为其添加标注,该物体对象的出现次数+1,并记录该物体对象的出现时间;若识别出与已标注的物体对象不同的物体对象,服务器则提示用户对其进行标注,该物体对象的出现次数记为I,并记录该物体对象的出现时间。 At the time of recognition of the next frame, and the object if the object has the same object is denoted by the recognized object, the server automatically to add annotation +1 the number of occurrences of the target object, and records the time of occurrence of the target object; if recognized objects and the object denoted by a different target object, the server prompts the user to be marked, the number of occurrences of the target object is referred to as I, and records the occurrence time of the object's object.

[0051]步骤S204,服务器接收用户通过客户端发送的视频搜索请求。 [0051] step S204, the video server receives a search request transmitted by the client user.

[0052]其中,客户端可以为个人电脑、手机、平板电脑等。 [0052] in which the client may be a personal computer, phone, tablet, and so on.

[0053]具体地,用户可通过客户端100中的文件浏览器或者网络浏览器输入视频搜索请求。 [0053] In particular, the user can through the client file browser 100 or web browser, a video input terminal search request. 例如,若用户需要搜索刘德华主演的影片,可通过客户端向服务器发送“刘德华主演的影片”的搜索请求。 For example, if the user needs to search movie starring Andy Lau, you can send "starring Andy Lau" search request by the client to the server.

[0054]步骤S205,服务器根据视频搜索请求进行搜索以获得多个搜索结果,其中,多个搜索结果中至少部分搜索结果具有统计信息 [0054] step S205, the video server to obtain a search request to search the plurality of search results, wherein the plurality of search results, at least portion of search results with statistical information

[0055]具体地,服务器通过搜索可获得多个影片结果,其中至少部分具有刘德华出现次数信息。 [0055] Specifically, the server searches through a plurality of video results obtained, wherein at least a portion having a number of occurrences Andy information.

[0056]步骤S206,服务器根据统计信息对多个搜索结果进行排序。 [0056] Step S206, the server sorts the search results based on the plurality of statistics.

[0057]例如,若用户需要搜索刘德华主演的影片,服务器可根据“刘德华出现次数”对搜索结果进行排序,刘德华出现次数最多的影片即为刘德华主演的影片。 [0057] For example, if the user needs to search movie starring Andy Lau, the server can be to sort search results based on "the number of times Andy Lau appeared," the film is the highest number of films starring Andy Lau appeared.

[0058]根据本发明实施例的视频文件的搜索排序方法,通过对视频文件中物体对象的出现次数及出现时间进行统计生成统计信息并呈现给用户,可以使用户能直接定位自己感兴趣的人物、动物或其他物体,进一步满足用户需求,提升用户体验。 [0058] The sorting method of searching video files embodiment of the present invention, generates statistics and statistical information presented to the user, can enable the user to directly locate the character of interest by the number of objects in the video file object appears and occurrence time , animals or other objects, to further meet customer needs, improve the user experience.

[0059]在本发明的一个实施例中,当用户播放视频文件时,还可以在播放前或播放后将视频文件中的物体对象的统计信息呈现给用户,例如,呈现出现哪些演员、演员出现次数,出现时间以及出现哪些动物、建筑物等,可以吸引用户注意并且方便用户选择所需资源,进一步提尚用户体验。 [0059] In one embodiment of the present invention, when a user plays a video file, may also be present in the statistics before playing object or objects after playing video file to a user, e.g., presenting what appears actor, actress appears number, time of occurrence and which animals and buildings appear to attract the user's attention and easy to access the resources needed to further improve the user experience yet.

[0060]为了实现上述实施例,本发明还提出一种视频文件的搜索排序系统。 [0060] In order to achieve the above-described embodiments, the present invention also provides a system for ordering search video file.

[0061]本发明实施例提出了一种视频文件的搜索排序系统,包括:客户端,用于接收用户输入的视频搜索请求,并将视频搜索请求发送至服务器;服务器,用于对多个视频文件分别进行视频解码以获取每个视频文件的解码后的图像,对解码后的图像进行图像识别以获取每个视频文件的多个物体对象的信息,并对每个视频文件中的每个物体对象的出现次数进行统计以生成每个视频文件对应的统计信息,以及根据客户端发送的搜索请求进行搜索以获取多个搜索结果,其中,多个搜索结果中至少部分搜索结果具有统计信息,并根据统计对多个搜索结果进行排序。 [0061] Example embodiments of the present invention proposes a sorting system for searching video files, comprising: a client configured to receive user input video search request, and transmits the search request to the video server; server, for a plurality of video files are decoded to obtain video images of each video file is decoded, the decoded image of the image to acquire information identifying a plurality of objects of each video object file, and each file for each object in the video statistical number of occurrences of the object to generate statistical information corresponding to each video file, and searches based on the search request sent by a client to obtain a plurality of search results, wherein the plurality of search results, at least portion of search results with statistical information, and sort the search results according to multiple statistics.

[0062]图3为根据本发明一个实施例的视频文件的搜索排序系统的结构示意图。 [0062] FIG. 3 is a schematic structural diagram of a video file the search sorting system according to one embodiment of the present invention.

[0063]如图3所示,根据本发明实施例的视频文件的搜索排序系统包括:客户端100和服务器200。 [0063] As shown in FIG 3, a video file the search ranking system embodiment of the present invention comprises: the client 100 and the server 200.

[0064]具体地,客户端100用于接收用户输入的视频搜索请求,并将视频搜索请求发送至服务器。 [0064] Specifically, the video client 100 for receiving user input of a search request and send the search request to the video server. 其中,客户端100可以为个人电脑,手机、平板电脑等。 Among them, the client 100 can wait for personal computers, mobile phones, tablet PCs. 更具体地,用户可通过客户端100中的文件浏览器或者网络浏览器输入视频搜索请求。 More specifically, the user may file browser 100 or web browser input video search request via the client.

[0065]服务器200用于对多个视频文件分别进行视频解码以获取每个视频文件的解码后的图像,对解码后的图像进行图像识别以获取每个视频文件的多个物体对象的信息,并对每个视频文件中的每个物体对象的出现次数及出现时间进行统计以生成每个视频文件对应的统计信息,以及根据客户端发送的搜索请求进行搜索以获取多个搜索结果,其中,多个搜索结果中至少部分搜索结果具有统计信息,并根据统计信息对多个搜索结果进行排序。 [0065] server 200 for a plurality of video files are video decoding to obtain decoded image of each video file, an image of the decoded image to acquire information identifying a plurality of objects of each video object file, time of occurrence and the number of occurrences of each object and each video object file to generate statistics statistics for each file corresponding to the video, as well as searches search request sent by the client to obtain a plurality of search results according to which, the plurality of search results with search result at least in part statistics, and a plurality of search results are ordered statistics.

[0066]在本发明的实施例中,服务器200在线下执行对解码后的图像进行图像识别以获取每个视频文件的多个物体对象的信息。 [0066] In an embodiment of the present invention, the execution of the decoded image after the image recognition server 200 at a plurality of online information for each video object file objects. 服务器200可以对视频文件进行分析,也可以接收其他服务器对多个视频文件进行分析后反馈的结果。 Server 200 may analyze the video file, the server may also receive other feedback results after multiple video files to be analyzed.

[0067]具体地,服务器对多个视频文件分别进行解码,获取每个视频文件解码后的YUV颜色格式或者RGB颜色格式的多帧原始图像,并通过人脸识别技术或者其他能够识别出图像中物体对象信息的图像识别技术对获取的多帧原始图像进行识别,其中多个物体对象可以包括人物,动物和/或建筑物等人或物,以及根据客户端发送的搜索请求进行搜索以获取多个搜索结果,并根据统计信息对多个搜索结果进行排序。 [0067] Specifically, the plurality of the video file server separately decode each video file acquired after decoding YUV color format or a plurality of frames of the original image of the RGB color format, and through the other face recognition technology or the image can be recognized object image recognition technology object information acquiring a plurality of frames of the original image is identified, wherein the plurality of objects may include character objects, animals and / or the person or thing and other buildings, and a search request transmitted by the client, searching for multiple search results and sort the search results according to statistics more information. 例如,若客户端发送的“刘德华主演的影片”的搜索请求,则服务器按照此请求搜索具有刘德华出现次数信息的结果。 For example, "starring Andy Lau" If a client sends a search request, the server in accordance with this request a search results occurrences Andy information. 并根据刘德华出现次数由多到少进行排序,供用户选择。 According to Andy and the number appears to be sorted by a little more choice for users.

[0068]更具体地,在本发明的实施例中,用户可通过多种方式输入其希望获得视频文件,例如通过文字输入或者图形输入,甚至还可在已有的视频文件中对该视频文件中的物体对象进行标注。 [0068] More specifically, in the embodiment of the present invention, the user can input a variety of ways to obtain their desired video file, such as by text input or drawing input, and even video files in the video file already labeling objects in the object. 例如,服务器可提示用户对识别出的物体对象进行标注,用户接收到提示后为其标注相关信息后,服务器对识别出的物体对象的出现次数进行统计。 For example, the server may prompt the user for the identified target object tagging, the user receives a prompt after the label information, the server of the number of occurrences of the identified target object for statistics.

[0069]例如,若识别出的物体对象为刘德华或张曼玉等人物,服务器提示用户标注后用户可为识别出的物体对象标注上刘德华或张曼玉等人物信息,若识别出的物体对象为猫、狗、大楼或汽车等,服务器提示用户标注后用户可为识别出的物体对象标注上猫、狗、大楼或汽车等信息。 [0069] For example, if the recognized object object like Andy or Maggie figures, the server prompts the user denoted by a user may be shown as identifiable object target label on the like Andy or Maggie character information, if the identified object target is a cat, a dog and so on, building or car, the server prompts the user to identify a user can mark objects object Notation information on cats, dogs, buildings or other car. 在对下一帧图像进行识别时,若识别出与已标注的物体对象相同的物体对象,服务器自动为其添加标注,该物体对象的出现次数+1,并记录该物体对象的出现时间;若识别出与已标注的物体对象不同的物体对象,服务器则提示用户对其进行标注,该物体对象的出现次数记为I,并记录该物体对象的出现时间。 At the time of recognition of the next frame, and the object if the object has the same object is denoted by the recognized object, the server automatically to add annotation +1 the number of occurrences of the target object, and records the time of occurrence of the target object; if recognized objects and the object denoted by a different target object, the server prompts the user to be marked, the number of occurrences of the target object is referred to as I, and records the occurrence time of the object's object.

[0070]根据本发明实施例的视频文件的搜索排序系统,通过对视频文件进行解码并对解码后的图像进行图像识别,获取视频文件中物体对象的统计信息,并将统计信息呈现给用户,从而可以按用户感兴趣的物体对象的统计信息对视频文件进行排序供用户选择,不仅能够满足用户需求,更快速且更准确地查找到其希望获得的视频文件,而且便于用户根据统计信息对视频文件进行管理,方便,直观,用户体验好。 [0070] The search order of the video file system of the embodiment of the present invention, the image file by decoding video image decoding and recognition, retrieve statistics objects in the video file object, and statistical information to the user, thus the video file according to the statistics of interest to the user object object sorting options for users, not only to meet customer needs, more quickly and accurately find the video files they wish to obtain, and easy for users based on statistical information on video file management, easy to use, intuitive, user experience is good.

[0071]在本发明的一个实施例中,当用户播放视频文件时,还可以在播放前或播放后将视频文件中的物体对象的统计信息呈现给用户,例如,呈现出现哪些演员、演员出现次数以及出现哪些动物、建筑物等,可以吸引用户注意并且方便用户选择所需资源,进一步提高用户体验。 [0071] In one embodiment of the present invention, when a user plays a video file, may also be present in the statistics before playing object or objects after playing video file to a user, e.g., presenting what appears actor, actress appears times and what animals, buildings, etc. appear, to attract the user's attention and easy to access the resources needed to further improve the user experience.

[0072] 为了实现上述实施例,本发明还提出一种服务器。 [0072] In order to achieve the above-described embodiments, the present invention further provides a server.

[0073]本发明提出了 一种服务器,包括:视频解码模块,用于对多个视频文件分别进行视频解码以获取每个视频文件的解码后的图像;图像识别模块,用于对解码后的图像进行图像识别以获取每个视频文件的多个物体对象的信息;统计模块,用于对每个视频文件中的每个物体对象的出现次数进行统计以生成每个视频文件对应的统计信息;接收模块,用于接收用户通过客户端发送的视频搜索请求;搜索模块,用于根据视频搜索请求进行搜索以获得多个搜索结果,其中,多个搜索结果中至少部分搜索结果具有统计信息;以及排序模块,用于根据统计信息对多个搜索结果进行排序。 [0073] The present invention provides a server comprising: a video decoder module, a video file, respectively, a plurality of video decoder for decoding each video image file; an image recognition module, for decoding after image information of image recognition to obtain a plurality of objects of each video file objects; statistics module for counting how often each object in each video object file to generate statistical information corresponding to the video files; receiving means for receiving video search user requests transmitted by the client; search module configured to search based on the search request to obtain a plurality of video search results, wherein the plurality of search results in the search result at least in part with the statistical information; sorting module for sorting the plurality of search results based on statistical information.

[0074]图4为根据本发明一个实施例的服务器的结构示意图。 [0074] FIG. 4 is a schematic structure of a server according to an embodiment of the present invention.

[0075]如图4所示,根据本发明实施例的服务器包括:视频解码模块100、图像识别模块200、统计模块300、接收模块400、搜索模块500和排序模块600。 [0075] As shown in FIG 4, a server according to embodiments of the present invention comprises: a video decoder module 100, the image recognition module 200, a statistics module 300, a receiving module 400, search module 500 and sorting module 600.

[0076]具体地,视频解码模块100用于对多个视频文件分别进行视频解码以获取每个视频文件的解码后的图像。 [0076] Specifically, the video decoder module 100 for each of the plurality of video decoding video file to obtain the decoded image of each video file. 更具体地,视频解码模块100对多个视频文件分别进行解码,获取每个视频文件解码后的多帧原始图像,以供图像识别模块200进行图像识别,其中原始图像一般为YUV颜色格式或者RGB颜色格式。 More specifically, a plurality of video decoding module 100 respectively decode the video file, acquiring a plurality of frames of the original image after decoding each video file, an image recognition module 200 for image recognition, in which the original image is typically a YUV format or RGB color color format.

[0077]图像识别模块200用于对解码后的图像进行图像识别以获取每个视频文件的多个物体对象的信息。 [0077] The image recognition module 200 is configured to image the decoded image to acquire information identifying a plurality of objects of each video object file. 其中,多个物体对象可以包括人物,动物和/或建筑物等人或物。 Wherein the plurality of objects may include objects or things people, animals and / or other buildings. 更具体地,图像识别模块200可通过人脸识别技术对获取的多帧原始图像进行识别,也可以通过其他能够识别出图像中物体对象信息的图像识别技术对获取的多帧原始图像进行识别。 More specifically, the image recognition module 200 may recognize a plurality of frames of the original image acquired by the face recognition technology, recognition may be performed on the plurality of frames of the original image obtained by other image recognition technology can recognize the target object in the image information. 当然在本发明的其他实施例中,还可对动物或者建筑物等进行识别。 Of course, in other embodiments of the present invention, an animal or a building and the like may also be identified. 具体的识别算法可为SURF算法、harr及adboost算法等。 Specific recognition algorithm may be as the SURF algorithm, harr and adboost algorithm.

[0078]统计模块300用于对每个视频文件中的每个物体对象的出现次数进行统计以生成每个视频文件对应的统计信息,并将统计信息保存在数据库之中。 [0078] Statistics module 300 for counting how often each object in each video object file to generate statistical information corresponding to the video file, and the statistical information stored in the database. 具体地,在本发明的实施例中,用户可通过多种方式输入其希望获得视频文件,例如通过文字输入或者图形输入,甚至还可在已有的视频文件中对该视频文件中的物体对象进行标注。 Specifically, in the embodiment of the present invention, the user can input a variety of ways to obtain their desired video file, such as by text input or drawing input, and even the target object in the prior video file in the video file be labeled. 例如,统计模块300可提示用户对识别出的物体对象进行标注,用户接收到提示为其标注相关信息后,统计模块300对识别出的物体对象的出现次数进行统计并生成每个视频文件对应的统计信息。 For example, statistics module 300 may prompt the user for the identified target object tagging, the user receives a prompt for label information, statistics module 300 to the number of occurrences of the identified object and generates statistics object file corresponding to each video Statistics.

[0079]在本发明的一个实施例中,统计模块300还用于对每个视频文件中的每个物体对象的出现时间进行统计,并生成包含出现次数信息和出现时间信息的统计信息。 [0079] In one embodiment of the present invention, the statistics module 300 for further statistical information about each object appearance time of each video object file, and generates an occurrence count information and statistical information of time information appear.

[0080]例如,若识别出的物体对象为刘德华或张曼玉等人物,服务器提示用户标注后用户可为识别出的物体对象标注上刘德华或张曼玉等人物信息,若识别出的物体对象为猫、狗、大楼或汽车等,服务器提示用户标注后用户可为识别出的物体对象标注上猫、狗、大楼或汽车等信息。 [0080] For example, if the recognized object object like Andy or Maggie figures, the server prompts the user denoted by a user may be shown as identifiable object target label on the like Andy or Maggie character information, if the identified object target is a cat, a dog and so on, building or car, the server prompts the user to identify a user can mark objects object Notation information on cats, dogs, buildings or other car. 在对下一帧图像进行识别时,若识别出与已标注的物体对象相同的物体对象,服务器自动为其添加标注,该物体对象的出现次数+1,并记录该物体对象的出现时间;若识别出与已标注的物体对象不同的物体对象,服务器则提示用户对其进行标注,该物体对象的出现次数记为I,并记录该物体对象的出现时间。 At the time of recognition of the next frame, and the object if the object has the same object is denoted by the recognized object, the server automatically to add annotation +1 the number of occurrences of the target object, and records the time of occurrence of the target object; if recognized objects and the object denoted by a different target object, the server prompts the user to be marked, the number of occurrences of the target object is referred to as I, and records the occurrence time of the object's object.

[0081]接收模块400用于接收用户通过客户端发送的视频搜索请求。 [0081] The search request receiving module 400 for receiving a video transmitted by the user client. 例如,若用户需要搜索刘德华主演的影片,可通过客户端向服务器发送“刘德华主演的影片”的搜索请求,接收模块400接收用户发送的视频搜索请求。 For example, if the user needs to search for the movie starring Andy, can send "Andy starring" search request to the server through the client, a video receiving module 400 receives a search request sent by the user.

[0082]搜索模块500用于根据视频搜索请求进行搜索以获得多个搜索结果,其中,多个搜索结果中至少部分搜索结果具有统计信息。 [0082] The search module 500 searches for a video according to the search request to obtain a plurality of search results, wherein the plurality of search results, at least portion of search results with statistical information.

[0083]排序模块600用于根据统计信息对多个搜索结果进行排序。 [0083] The ranking module 600 for sorting the plurality of search results based on statistical information. 例如,若用户需要搜索刘德华主演的影片,排序模块600可根据“刘德华出现次数”对搜索结果进行排序,刘德华出现次数最多的影片即为刘德华主演的影片。 For example, if the user needs to search starring Andy Lau, ordering module 600 may search results sorted by "the number of times Andy Lau appeared," the film is the highest number of films starring Andy Lau appeared.

[0084]根据本发明实施例的服务器,通过对视频文件进行解码并对解码后的图像进行图像识别,获取视频文件中物体对象的统计信息,并将统计信息呈现给用户,从而可以按用户感兴趣的物体对象的统计信息对视频文件进行排序供用户选择,不仅能够满足用户需求,更快速且更准确地查找到其希望获得的视频文件,而且便于用户根据统计信息对视频文件进行管理,方便,直观,用户体验好。 [0084] server according to an embodiment of the present invention, an image decoded by the video decoding and image identification file, get statistics object video object file, and statistical information to the user, whereby the user can sense by statistical information of the object of interest in the subject video file sorting options for users, not only to meet customer needs, more quickly and accurately find the video files they wish to obtain, and easy for users to manage video files according to statistics, convenient intuitive, user experience is good.

[0085]在本发明的一个实施例中,当用户播放视频文件时,还可以在播放前或播放后将视频文件中的物体对象的统计信息呈现给用户,例如,呈现出现哪些演员、演员出现次数以及出现哪些动物、建筑物等,可以吸引用户注意并且方便用户选择所需资源,进一步提高用户体验。 [0085] In one embodiment of the present invention, when a user plays a video file, may also be present in the statistics before playing object or objects after playing video file to a user, e.g., presenting what appears actor, actress appears times and what animals, buildings, etc. appear, to attract the user's attention and easy to access the resources needed to further improve the user experience.

[0086]在本说明书的描述中,参考术语“一个实施例”、“一些实施例”、“示例”、“具体示例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本发明的至少一个实施例或示例中。 [0086] In the description of the present specification, reference to the term "one embodiment," "some embodiments", "an example", "a specific example", or "some examples" means that a description of the exemplary embodiment or embodiments described a particular feature, structure, material, or characteristic is included in at least one embodiment of the present invention, embodiments or examples. 在本说明书中,对所述术语的示意性表述不一定指的是相同的实施例或示例。 In the present specification, a schematic representation of the term is not necessarily referring to the same embodiment or exemplary embodiments. 而且,描述的具体特征、结构、材料或者特点可以在任何的一个或多个实施例或示例中以合适的方式结合。 Furthermore, the particular features, structures, materials, or characteristics described embodiments or examples may be at any one or more in a proper manner.

[0087]尽管已经示出和描述了本发明的实施例,对于本领域的普通技术人员而言,可以理解在不脱离本发明的原理和精神的情况下可以对这些实施例进行多种变化、修改、替换和变型,本发明的范围由所附权利要求及其等同限定。 [0087] While there has been illustrated and described embodiments of the present invention, those of ordinary skill in the art, to be understood that various changes may be made to these embodiments without departing from the principles and spirit of the present invention, modifications, substitutions and modifications, the scope of the invention being indicated by the appended claims and their equivalents.

Claims (15)

  1. 1.一种视频文件的搜索排序方法,其特征在于,包括以下步骤: 服务器对多个视频文件分别进行视频解码以获取每个视频文件的解码后的图像; 所述服务器对所述解码后的图像进行图像识别以获取每个视频文件的多个物体对象的信息; 所述服务器对每个视频文件中的每个物体对象的出现次数进行统计以生成每个视频文件对应的统计信息; 所述服务器接收用户通过客户端发送的视频搜索请求; 所述服务器根据所述视频搜索请求进行搜索以获得多个搜索结果,其中,所述多个搜索结果中至少部分搜索结果具有所述统计信息;以及所述服务器根据所述统计信息对所述多个搜索结果进行排序; 其中,当用户播放视频文件时,在播放前将视频文件中的物体对象的统计信息呈现给用户。 A sorting method for searching video files, characterized by comprising the steps of: a plurality of video file server each video decoding to obtain decoded image of each video file; the server after the decoding image information of image recognition to obtain a plurality of objects of each video file objects; the number of occurrences of each server object file for each video object in the statistics to generate statistical information corresponding to the video file; the server sent by a user through the client search request video; the video search server according to the search request to obtain a plurality of search results, wherein the plurality of search results in the search result at least in part with the statistical information; the server ordering the plurality of search results based on the statistical information; wherein, when a user plays a video file, the statistical information before playing the video file object object presented to the user.
  2. 2.根据权利要求1所述的方法,其特征在于,所述多个物体对象包括人物。 2. The method according to claim 1, wherein said object comprises a plurality of character objects.
  3. 3.根据权利要求1或2所述的方法,其特征在于,所述服务器通过人脸识别对所述图像进行识别。 3. The method of claim 1 or claim 2, wherein said server identification by the image recognition.
  4. 4.根据权利要求1或2所述的方法,其特征在于,所述多个物体对象还包括动物和/或建筑物。 4. The method of claim 1 or claim 2, wherein said plurality of target objects further comprises animal and / or buildings.
  5. 5.根据权利要求1所述的方法,其特征在于,还包括: 所述服务器对所述每个物体对象的出现时间进行统计,其中,所述统计信息包括出现次数信息和出现时间信息。 5. The method according to claim 1, characterized in that, further comprising: a server time of the occurrence statistics object for each object, wherein the statistical information includes the number of occurrences occurrence information and time information.
  6. 6.一种视频文件的搜索排序系统,其特征在于,包括: 客户端,用于接收用户输入的视频搜索请求,并将所述视频搜索请求发送至服务器; 服务器,用于对多个视频文件分别进行视频解码以获取每个视频文件的解码后的图像,对所述解码后的图像进行图像识别以获取每个视频文件的多个物体对象的信息,并对每个视频文件中的每个物体对象的出现次数进行统计以生成每个视频文件对应的统计信息,以及根据所述客户端发送的搜索请求进行搜索以获取多个搜索结果,其中,所述多个搜索结果中至少部分搜索结果具有所述统计信息,并根据所述统计信息对所述多个搜索结果进行排序; 所述客户端还用于:当用户播放视频文件时,在播放前将视频文件中的物体对象的统计信息呈现给用户。 A system for ordering search video file, characterized by comprising: a client, configured to receive a video search request input by the user, and transmits the search request to the video server; for a plurality of video files are decoded to obtain video images of each video file is decoded, the decoded image of an image to obtain information identifying a plurality of objects each recorded video objects, each video file and each the number of occurrences of the object to generate statistics object statistical information corresponding to the video file, and performing searches search request sent by the client according to obtain a plurality of search results, wherein the plurality of search results in the search result at least in part with the statistical information, the statistical information according to the plurality of search results are sorted; the client further configured to: when the user is playing a video file, the statistical information before playing video object file object presented to the user.
  7. 7.根据权利要求6所述的系统,其特征在于,所述多个物体对象包括人物。 7. The system according to claim 6, wherein said object comprises a plurality of character objects.
  8. 8.根据权利要求6所述的系统,其特征在于,所述服务器通过人脸识别对所述图像进行识别。 8. The system according to claim 6, characterized in that the server identified by the image recognition.
  9. 9.根据权利要求6-8任一项所述的系统,其特征在于,所述多个物体对象还包括动物和/或建筑物。 9. The system of any one of claims 6-8, wherein said plurality of target objects further comprises animal and / or buildings.
  10. 10.根据权利要求6所述的系统,其特征在于,所述服务器还用于对所述每个物体对象的出现时间进行统计,其中,所述统计信息包括出现次数信息和出现时间信息。 10. The system according to claim 6, wherein said server is further configured to time the occurrence statistics object for each object, wherein the statistical information includes the number of occurrences occurrence information and time information.
  11. 11.一种服务器,其特征在于,包括: 视频解码模块,用于对多个视频文件分别进行视频解码以获取每个视频文件的解码后的图像; 图像识别模块,用于对所述解码后的图像进行图像识别以获取每个视频文件的多个物体对象的信息; 统计模块,用于对每个视频文件中的每个物体对象的出现次数进行统计以生成每个视频文件对应的统计信息; 接收模块,用于接收用户通过客户端发送的视频搜索请求; 搜索模块,用于根据视频搜索请求进行搜索以获得多个搜索结果,其中,所述多个搜索结果中至少部分搜索结果具有所述统计信息;以及排序模块,用于根据所述统计信息对所述多个搜索结果进行排序; 其中,当用户播放视频文件时,在播放前将视频文件中的物体对象的统计信息呈现给用户。 11. A server, comprising: a video decoder module, a video file, respectively, a plurality of video decoder for decoding each video image file; an image recognition module, for decoding said rear the image information of image recognition to obtain a plurality of objects of each video object file; statistics module for statistics for each video file to generate statistics corresponding to the number of occurrences for each object in each video object file ; receiving module, configured to receive a video search through the client user sends a request; searching module configured to search according to the search request to obtain a plurality of video search results, wherein the plurality of search results in the search result at least in part having said statistical information; and a sorting means for sorting the plurality of search results based on the statistical information; wherein, when a user plays a video file, the statistical information before playing the video file object object presented to the user .
  12. 12.根据权利要求11所述的服务器,其特征在于,所述多个物体对象包括人物。 12. A server as claimed in claim 11, wherein said object comprises a plurality of character objects.
  13. 13.根据权利要求11所述的服务器,其特征在于,所述图像识别模块通过人脸识别对所述解码后的图像进行识别。 13. A server as claimed in claim 11, wherein said image recognition module for recognition of the decoded image by the face recognition.
  14. 14.根据权利要求11-13任一项所述的服务器,其特征在于,所述多个物体对象还包括动物和/或建筑物。 14. A server according to any one of claims 11-13, wherein said plurality of target objects further comprises animal and / or buildings.
  15. 15.根据权利要求11所述的服务器,其特征在于,所述统计模块还用于对所述每个物体对象的出现时间进行统计,其中,所述统计信息包括出现次数信息和出现时间信息。 15. The server according to claim 11, wherein said module is further configured to count time of the occurrence statistics object for each object, wherein the statistical information includes the number of occurrences occurrence information and time information.
CN 201210448634 2012-11-09 2012-11-09 Search video file sorting method, system and server CN102955858B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 201210448634 CN102955858B (en) 2012-11-09 2012-11-09 Search video file sorting method, system and server

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 201210448634 CN102955858B (en) 2012-11-09 2012-11-09 Search video file sorting method, system and server

Publications (2)

Publication Number Publication Date
CN102955858A true CN102955858A (en) 2013-03-06
CN102955858B true CN102955858B (en) 2016-07-27

Family

ID=47764664

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201210448634 CN102955858B (en) 2012-11-09 2012-11-09 Search video file sorting method, system and server

Country Status (1)

Country Link
CN (1) CN102955858B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104754373A (en) * 2013-12-27 2015-07-01 联想(北京)有限公司 Video acquisition method and electronic device
CN105744140A (en) * 2014-12-10 2016-07-06 信泰光学(深圳)有限公司 Device and method for recording occurrence number of objects to be detected
CN104866565A (en) * 2015-05-21 2015-08-26 百度在线网络技术(北京)有限公司 Picture searching method and apparatus

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101021857A (en) * 2006-10-20 2007-08-22 鲍东山 Video searching system based on content analysis
CN101493813A (en) * 2008-01-25 2009-07-29 北京新岸线网络技术有限公司 Video search method based on content and implementing method thereof
CN102117331A (en) * 2011-03-07 2011-07-06 北京百度网讯科技有限公司 Video search method and system

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7881505B2 (en) * 2006-09-29 2011-02-01 Pittsburgh Pattern Recognition, Inc. Video retrieval system for human face content
CN102419945A (en) * 2011-12-09 2012-04-18 上海聚力传媒技术有限公司 Method, device, equipment and system for presenting display information in video

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101021857A (en) * 2006-10-20 2007-08-22 鲍东山 Video searching system based on content analysis
CN101493813A (en) * 2008-01-25 2009-07-29 北京新岸线网络技术有限公司 Video search method based on content and implementing method thereof
CN102117331A (en) * 2011-03-07 2011-07-06 北京百度网讯科技有限公司 Video search method and system

Also Published As

Publication number Publication date Type
CN102955858A (en) 2013-03-06 application

Similar Documents

Publication Publication Date Title
Furini et al. STIMO: STIll and MOving video storyboard for the web scenario
US20080306995A1 (en) Automatic story creation using semantic classifiers for images and associated meta data
US20080109369A1 (en) Content Management System
US20070124282A1 (en) Video data directory
US20090144772A1 (en) Video object tag creation and processing
US20080313146A1 (en) Content search service, finding content, and prefetching for thin client
US8315423B1 (en) Providing information in an image-based information retrieval system
US20130071087A1 (en) Video management system
US8132200B1 (en) Intra-video ratings
US20080209010A1 (en) Information sharing between images
US20110246560A1 (en) Social context for inter-media objects
US20140188997A1 (en) Creating and Sharing Inline Media Commentary Within a Network
US20090094189A1 (en) Methods, systems, and computer program products for managing tags added by users engaged in social tagging of content
US20150046537A1 (en) Retrieving video annotation metadata using a p2p network and copyright free indexes
US20120078899A1 (en) Systems and methods for defining objects of interest in multimedia content
US20100082653A1 (en) Event media search
US8856121B1 (en) Event based metadata synthesis
US20120027256A1 (en) Automatic Media Sharing Via Shutter Click
US20130051772A1 (en) Methods and apparatus to access media
US20100095345A1 (en) System and method for acquiring and distributing keyframe timelines
US20110208722A1 (en) Method and apparatus for segmenting and summarizing media content
US20120209841A1 (en) Bookmarking segments of content
US20120078626A1 (en) Systems and methods for converting speech in multimedia content to text
US8099407B2 (en) Methods and systems for processing media files
US8566315B1 (en) Sequenced video segment mix

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
C14 Grant of patent or utility model