WO2016184051A1 - 图片搜索方法、装置、设备及非易失性计算机存储介质 - Google Patents

图片搜索方法、装置、设备及非易失性计算机存储介质 Download PDF

Info

Publication number
WO2016184051A1
WO2016184051A1 PCT/CN2015/094339 CN2015094339W WO2016184051A1 WO 2016184051 A1 WO2016184051 A1 WO 2016184051A1 CN 2015094339 W CN2015094339 W CN 2015094339W WO 2016184051 A1 WO2016184051 A1 WO 2016184051A1
Authority
WO
WIPO (PCT)
Prior art keywords
search
information
searched
picture
user
Prior art date
Application number
PCT/CN2015/094339
Other languages
English (en)
French (fr)
Inventor
许笑天
侯思宇
姜岩
Original Assignee
百度在线网络技术(北京)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
Application filed by 百度在线网络技术(北京)有限公司 filed Critical 百度在线网络技术(北京)有限公司
Priority to US15/524,544 priority Critical patent/US20170337222A1/en
Publication of WO2016184051A1 publication Critical patent/WO2016184051A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/53Querying
    • G06F16/532Query formulation, e.g. graphical querying
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G06F16/5838Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using colour
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24Querying
    • G06F16/245Query processing
    • G06F16/2457Query processing with adaptation to user needs
    • G06F16/24578Query processing with adaptation to user needs using ranking
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/25Integrating or interfacing systems involving database management systems
    • G06F16/252Integrating or interfacing systems involving database management systems between a Database Management System and a front-end application
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/51Indexing; Data structures therefor; Storage structures
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/73Querying
    • G06F16/732Query formulation
    • G06F16/7328Query by example, e.g. a complete video frame or video sequence
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/951Indexing; Web crawling techniques

Definitions

  • the present disclosure relates to the field of information search technologies, and in particular, to a picture search method, apparatus, device, and non-volatile computer storage medium.
  • aspects of the present disclosure provide a picture search method, apparatus, device, and non-volatile computer storage medium for improving the efficiency of obtaining useful information from search results.
  • An aspect of the present disclosure provides a picture search method, including:
  • Search results are obtained according to the to-be-searched picture and the search intent information.
  • a picture search apparatus including:
  • An obtaining module configured to acquire a picture to be searched and search intent information for the picture to be searched
  • a searching module configured to obtain a search result according to the to-be-searched image and the search intent information.
  • an apparatus comprising:
  • One or more processors are One or more processors;
  • One or more programs the one or more programs being stored in the memory, when executed by the one or more processors:
  • Search results are obtained according to the to-be-searched picture and the search intent information.
  • a non-volatile computer storage medium storing one or more programs when the one or more programs are executed by one device causes The device:
  • Search results are obtained according to the to-be-searched picture and the search intent information.
  • the searched image and the search intention information for the image to be searched are obtained, and the search result is obtained according to the image to be searched and the search intention information. Since the search is performed according to the image to be searched and the search intention information at the same time, the obtained search result is the search result closest to the user's search intention, which greatly reduces the number of search results compared with the prior art. This can improve the efficiency of users' access to useful information in search results.
  • FIG. 1 is a schematic flowchart diagram of a picture search method according to an embodiment of the present disclosure
  • FIG. 2 is a schematic structural diagram of a picture search apparatus according to an embodiment of the present disclosure.
  • FIG. 1 is a schematic flowchart diagram of a picture search method according to an embodiment of the present disclosure. As shown in Figure 1, the method includes:
  • This embodiment provides a picture search method, specifically:
  • the image search device obtains the image to be searched, and obtains the search intention information for the image to be searched, and then searches according to the image to be searched and the search intention information, thereby obtaining and satisfying the image to be searched. Search for intent search results.
  • the above-mentioned search intent information for the picture to be searched mainly indicates the search intention or search requirement of the user to search for the picture.
  • the amount of information related to the picture to be searched is large, and by searching for the intent information, information that does not conform to the search intention can be filtered out from all the information related to the picture to be searched, and the amount of information is reduced.
  • the information related to the picture includes: calorie information of the vegetable, recipe information, purchase information, related news, comprehensive encyclopedia information, and the like.
  • the user's search intention is to search for the origin information
  • the amount of information related to the vegetable and related to the vegetable origin information is much smaller than the amount of information related to the vegetable, for example, at the same time
  • the information related to the vegetable and related to the origin information of the vegetable may include related news of the vegetable, comprehensive encyclopedia information, etc., and does not include calorie information, recipe information, etc. of the vegetable.
  • the search intention information for the picture to be searched is searched as a search condition, and the search intention can be directly obtained and the search intention is satisfied.
  • the search result of the information can greatly reduce the number of search results compared with the prior art only based on the image search, and the user can quickly obtain the desired information from the search result, thereby improving the user. The efficiency of getting useful information from it.
  • the object to be identified may be taken as a picture to be searched by the object to be identified, and the object to be identified may be a plant (such as a tree, a flower, etc.), an animal, or a clothing (such as a backpack). Tops, shoes), books, food, smart terminals (such as Mobile phones, tablets, printers, etc.), servers, network devices, etc.
  • a picture including the object to be identified or content may be selected from the local gallery as the picture to be searched.
  • a picture including the object to be identified or content may be acquired from the cloud as a picture to be searched.
  • the picture to be searched above may be one or more sheets.
  • the search intent information for the to-be-searched picture may include at least one of text information, voice information, and video information.
  • the user can express the search intent of the searched image by at least one of text, voice, and video.
  • the search intention information of the search picture includes: price information described in text form, for example, the specific text content may be: the price of the children's book, described in a voice form
  • the publisher information, such as the specific voice content may be: the publisher of the children's book, and the content information described in the form of a video, such as the specific video content may be: an illustration of the children's book continuously played.
  • the manner in which the search intent information is obtained may vary depending on the implementation form of the search intent information.
  • the text information in the search intention information from the user's point of view, the text information can be input through various text input tools, such as a keyboard, a mouse, a stylus, a touch screen, etc.; from the perspective of the image search device, the user can be accepted. Text information entered by various text input tools.
  • the voice information in the search intention information from the perspective of the user, the voice information can be input through a voice recording module, such as a microphone; from the perspective of the image search device, the voice information recorded by the voice recording module can be obtained.
  • the video shooting module such as a camera, can be used for shooting; from the perspective of the image search device, the video information captured by the video shooting module can be obtained.
  • step 101 the image to be searched and the search intent information for the image to be searched are obtained.
  • the image search device may be activated to enter the search page; then, the image search device obtains the image to be searched; and then the image search device receives the search intention information input by the user.
  • a shooting button (the button may be a camera icon) may be set on the search page for the user to issue a photographing instruction.
  • the user can initiate a search function (ie, launch the image search device), and the search function is initiated in a manner that is entered into the search page provided by the image search device.
  • the user issues a photographing instruction by searching the photographing button on the page;
  • the image searching device receives the photographing instruction issued by the user, starts the photographing module according to the instruction, and takes a photograph to the recognized object to obtain a to-be-searched image.
  • the user can use the camera module to align the object to be identified, and take a photo with the object to be recognized.
  • the picture search device receives search intent information input by the user.
  • the image search device receives the search intention information input by the user, after the image search device obtains the image to be searched, the image search device automatically starts the recording module to record the voice of the user to obtain search intention information in a voice form.
  • the image search device may acquire the audio information in the video stream as the search intention information by simultaneously recording a video stream through the camera module having the camera function.
  • the picture in the video stream is a picture to be searched, and the audio information in the video stream is search intent information in a voice form.
  • the picture search device receives the search intent information input by the user
  • at least one of a text input box, a voice recording button, and a video capture button may be set on the search page for the user to input the search intention information.
  • the user inputs the search intention information through the text output box on the search page; the image search device is After obtaining the picture to be searched, the search intent information input by the user is received.
  • the user issues a recording instruction by searching the recording button on the page, and after obtaining the image to be searched, the image searching device receives the recording instruction, starts the voice recording module, and records the voice expressing the user's search intention to obtain Search intent information in the form of speech.
  • the user sends a video shooting instruction by searching the video button on the page, and after obtaining the image to be searched, the image searching device receives the video shooting instruction, starts the video shooting module, and shoots the video expressing the user's search intention.
  • the original information input by the user is monitored in real time; whether the user has a search requirement according to the original information; When there is a search demand, the search page is entered, the image to be searched is obtained, and the original information is used as the search intention information for the image to be searched.
  • the image search device monitors the original information input by the user in real time, and determines whether the user has a search requirement according to the monitored original information.
  • the original information input by the user may include at least one of text information, voice information, and video information.
  • the image search device may preset a search demand word, and the search demand word is a word or sentence that reflects the user's search demand.
  • the search demand word may be similar to “go where”, “how much money”, “this Words or sentences such as "Where is the origin of the vegetables", "When is this movie released?", "Several number”.
  • the image search device may specifically determine whether the monitored original information belongs to a preset search demand word; when the determination result is yes, it is determined that the user has a search requirement; when the determination result is negative, it is determined that the user does not have a search requirement.
  • the image search device When it is determined that the user has a search demand, the image search device starts the search function and starts The search function is represented by entering the search page. After that, the image search device takes a picture to the recognized object to obtain a picture to be searched, and uses the monitored original information as the search intention information of the picture to be searched.
  • the image search module automatically starts to take a picture to obtain a picture to be searched.
  • the order in which the picture search device activates the photographing module and enters the search page is not limited.
  • a capture button (which may be a camera icon) may be set on the search page for the user to issue a photo instruction.
  • the image search device After the image search device determines that the user has a search request, the image search device first enters the search page. The user sends a photographing instruction through the photographing button on the search page; the image searching device receives the photographing instruction issued by the user, starts the photographing module according to the instruction, and takes a photo to be recognized to obtain the image to be searched.
  • the image search device may first perform a search according to the image to be searched to obtain an initial search result according to the search intention. Information, a secondary search in the initial search results to obtain the final search results.
  • the image search device extracts feature information of the image to be searched; and combines the feature information and the search intention information. To obtain search keywords; search directly according to the search keywords to obtain search results.
  • the search intention information includes voice information
  • the voice information can be converted into text information, and then the feature information and the text information are Line merge processing.
  • a plurality of general extraction algorithms may be used to extract features of the image to be searched.
  • a bag of word (BOW) algorithm may be used to extract an indefinite number of features in the image to be searched, each of which A feature corresponds to a feature vector so that multiple features can be extracted from a single image.
  • the image search device may obtain an inverted index corresponding to the feature information and an inverted index corresponding to the search intention information; and then perform weighting processing on the inverted index corresponding to the feature information and the inverted index corresponding to the search intention information, Obtain the above search keywords.
  • the image search device may retrieve the inverted index database by using the feature information and the search intention information, respectively, thereby obtaining an inverted index corresponding to the feature information and an inverted index corresponding to the search intention information.
  • the feature information and the search intent information may share an inverted index library, or a separate inverted index library may be used.
  • the foregoing feature information and the search intent information generally correspond to a plurality of inverted indexes, and the first N inverted indexes corresponding to the feature information and the search intention information may be respectively acquired, and the 2N inverted indexes are obtained. Weighting processing is performed; then the weighting processing results are sorted, and the top M weighting processing results are selected as search keywords. Where N and M are both natural numbers and M is less than or equal to N.
  • the user when the user sees a poster of a movie while riding the subway, if the user wants to know the screening and fare of the movie, the user can pick up the mobile phone and enter the search page to take the poster. And recording the voice information of the screening theater and the fare, then extracting the feature information of the poster by BOW or other extraction algorithm, and converting the voice information into text information, combining the feature information with the text information to generate a search keyword. Search by keyword, and get search results related to the movie's screening and fare. Present the obtained search results to the user. or
  • the user when the user sees a poster of a movie while riding the subway, if the user wants to know the screening and fare information of the movie, the user can turn on the recording function on the mobile phone, and record the inquiry screening theater. And the voice information of the fare, when it is judged according to the voice information that the user has a search demand, enter the search page and start the camera module (such as a camera), the user takes the poster through the camera module, and then extracts by using BOW or other extraction algorithm.
  • the camera module such as a camera
  • the feature information of the poster and converting the voice information into text information, combining the feature information with the text information, generating a search keyword, searching using the search keyword, and obtaining search results related to the screening and fare of the movie, Present the obtained search results to the user.
  • the present application simultaneously searches according to the image to be searched and the search intention information, and the obtained search result is the search result closest to the user's search intention, which greatly reduces the number of search results compared with the prior art. Therefore, it is possible to improve the efficiency of users in obtaining useful information in search results.
  • FIG. 2 is a schematic structural diagram of a picture search apparatus according to an embodiment of the present disclosure. As shown in FIG. 2, the method includes: an acquisition module 21 and a search module 22.
  • the obtaining module 21 is configured to obtain a picture to be searched and search intent information for the picture to be searched.
  • the search module 22 is configured to obtain a search result according to the to-be-searched image and the search intention information acquired by the obtaining module 21.
  • the obtaining module 21 is specifically configured to:
  • the search intent information input by the user is received.
  • the obtaining module 21 is specifically configured to:
  • the search page is entered, the picture to be searched is obtained, and the original information is used as search intent information for the picture to be searched.
  • the obtaining module 21 is configured to: when determining whether the user has a search requirement according to the original information, specifically:
  • the foregoing search intent information may include at least one of text information, voice information, and video information.
  • the searching module 42 is specifically configured to: extract feature information of the search image; perform a combination process on the feature information and the search intention information to obtain a search keyword; Words are searched to get search results.
  • the search module 42 may be configured to: obtain an inverted index corresponding to the feature information and an inverted index corresponding to the search intention information, where the feature information and the search intention information are combined and combined; The inverted index and the inverted index corresponding to the search intention information are weighted to obtain a search keyword.
  • the search module 42 may retrieve the inverted index library by using the feature information and the search intention information, respectively, to obtain an inverted index corresponding to the feature information and an inverted index corresponding to the search intention information.
  • the feature information and the search intent information may share an inverted index library, or a separate inverted index library may be used.
  • the search module 42 may specifically acquire the feature information and the top N inverted indexes corresponding to the search intention information, for the 2N
  • the inverted index is subjected to weighting processing; then the weighting processing results are sorted, and the top M weighting processing results are selected as search keywords.
  • N and M are both natural numbers and M is less than or equal to N.
  • the image search device provided in this embodiment acquires a search image to be searched and search intention information for the image to be searched, and obtains a search result according to the image to be searched and the search intention information. Since the picture search device of the present embodiment can simultaneously search according to the picture to be searched and the search intention information, the search result obtained is the search result closest to the user's search intention, which greatly reduces the search compared with the prior art. The number of results, thus improving the efficiency of users in obtaining useful information in search results.
  • the disclosed system is The method of setting can be implemented in other ways.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in various embodiments of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or in the form of hardware plus software functional units.
  • the above-described integrated unit implemented in the form of a software functional unit can be stored in a computer readable storage medium.
  • the above software functional unit is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor to perform the methods of the various embodiments of the present disclosure. Part of the steps.
  • the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like, which can store program codes. .

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Multimedia (AREA)
  • Mathematical Physics (AREA)
  • Library & Information Science (AREA)
  • Software Systems (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

本公开提供一种图片搜索方法、装置、设备及非易失性计算机存储介质。方法包括:获取待搜索图片及针对所述待搜索图片的搜索意图信息;根据所述待搜索图片和所述搜索意图信息,获得搜索结果。本公开可以提高从搜索结果中获取有用信息的效率。

Description

图片搜索方法、装置、设备及非易失性计算机存储介质
本申请要求了申请日为2015年05月18日,申请号为201510253333.X发明名称为“图片搜索方法及装置”的中国专利申请的优先权。
技术领域
本公开涉及信息搜索技术领域,特别涉及一种图片搜索方法、装置、设备及非易失性计算机存储介质。
背景技术
随着互联网技术的发展,目前用户已经不再满足于只是对文本的搜索,很多用户还希望可以通过搜索引擎对网络图片进行搜索,于是出现了图片搜索技术。
目前的图片搜索技术,大多都只针对图片进行比对处理,搜索得到相似的图片结果和文字结果等。目前这种图片搜索技术并不能准确地理解用户需求,返回的搜索结果通常较多,用户需要在搜索结果中不断搜寻,往往不能快速地从中获取想要的信息,获取有用信息的效率低。
发明内容
本公开的多个方面提供一种图片搜索方法、装置、设备及非易失性计算机存储介质,用以提高从搜索结果中获取有用信息的效率。
本公开的一方面,提供一种图片搜索方法,包括:
获取待搜索图片及针对所述待搜索图片的搜索意图信息;
根据所述待搜索图片和所述搜索意图信息,获得搜索结果。
本公开的另一方面,提供一种图片搜索装置,包括:
获取模块,用于获取待搜索图片及针对所述待搜索图片的搜索意图信息;
搜索模块,用于根据所述待搜索图片和所述搜索意图信息,获得搜索结果。
本公开的另一方面,提供一种设备,包括:
一个或者多个处理器;
存储器;
一个或者多个程序,所述一个或者多个程序存储在所述存储器中,当被所述一个或者多个处理器执行时:
获取待搜索图片及针对所述待搜索图片的搜索意图信息;
根据所述待搜索图片和所述搜索意图信息,获得搜索结果。
本公开的另一方面,提供一种非易失性计算机存储介质,所述非易失性计算机存储介质存储有一个或者多个程序,当所述一个或者多个程序被一个设备执行时,使得所述设备:
获取待搜索图片及针对所述待搜索图片的搜索意图信息;
根据所述待搜索图片和所述搜索意图信息,获得搜索结果。
在本申请中,获取待搜索图片及针对待搜索图片的搜索意图信息,同时根据待搜索图片和搜索意图信息获得搜索结果。由于同时根据待搜索图片和搜索意图信息进行搜索,所获得的搜索结果是最接近用户搜索意图的搜索结果,与现有技术相比,极大的降低了搜索结果的数量,因 此可以提高用户在搜索结果中获取有用信息的效率。
附图说明
为了更清楚地说明本公开实施例中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作一简单地介绍,显而易见地,下面描述中的附图是本公开的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。
图1为本公开一实施例提供的图片搜索方法的流程示意图;
图2为本公开一实施例提供的图片搜索装置的结构示意图。
具体实施方式
为使本公开实施例的目的、技术方案和优点更加清楚,下面将结合本公开实施例中的附图,对本公开实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例是本公开一部分实施例,而不是全部的实施例。基于本公开中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本公开保护的范围。
图1为本公开一实施例提供的图片搜索方法的流程示意图。如图1所示,该方法包括:
101、获取待搜索图片及针对待搜索图片的搜索意图信息。
102、根据上述待搜索图片和上述搜索意图信息,获得搜索结果。
针对现有图片搜索存在的无法准确理解用户需求,搜索结果数量较多,导致用户从搜索结果中获取有用信息的效率较低的问题,本实施例提供一种图片搜索方法,具体的:
当需要进行图片搜索时,图片搜索装置获取待搜索图片,并获取针对该待搜索图片的搜索意图信息,然后同时根据该待搜索图片和搜索意图信息进行搜索,从而获得与待搜索图片相关又满足搜索意图的搜索结果。
上述针对待搜索图片的搜索意图信息主要表示用户对待搜索图片的搜索意图或搜索需求。一般来说,与待搜索图片相关的信息量会很大,而通过搜索意图信息可以从所有与待搜索图片相关的信息中过滤掉不符合搜索意图的信息,减少信息量。以待搜索图片为一种蔬菜的图片为例,则与该图片相关的信息包括:该蔬菜的热量信息、菜谱信息、购买信息、相关新闻、综合百科信息,等等。假设对于该蔬菜,用户的搜索意图是搜索原产地信息,则与该蔬菜有关的信息量相比,同时与该蔬菜相关且又涉及该蔬菜原产地信息的信息量会小很多,例如,同时与该蔬菜相关且又涉及该蔬菜原产地信息的信息可能包括该蔬菜的相关新闻、综合百科信息等,而不会包括该蔬菜的热量信息、菜谱信息等。
由上述可见,在本实施例中,除了将待搜索图片作为搜索条件之外,同时将针对待搜索图片的搜索意图信息作为搜索条件进行搜索,可以直接获取既与待搜索图片相关又满足搜索意图信息的搜索结果,与现有技术中仅基于图片进行搜索的方案相比,本实施例可以极大的降低搜索结果的数量,用户可以快速地从搜索结果中获取想要的信息,提高了用户从中获取有用信息的效率。
在实际应用中,可以通过对待识别对象进行拍照,将待识别对象的照片作为待搜索图片,所述待识别对象可以是植物(如一棵树、一朵花等)、动物、衣物(如背包、上衣、鞋子)、书籍、食品、智能终端(如 手机、平板电脑、打印机等)、服务器、网络设备等。或者,也可以从本地图库中选取包括待识别对象或内容的图片作为待搜索图片。或者,也可以从云端获取包括待识别对象或内容的图片作为待搜索图片。
值得说明的是,上述待搜索图片可以是一张或多张。
在实际应用中,针对上述待搜索图片的搜索意图信息可以包括文本信息、语音信息和视频信息中的至少一种。简单来说,用户可以通过文本、语音和视频中的至少一种方式表达对待搜索图片的搜索意图。举例说明,以内容为儿童读物的待搜索图片为例,对该搜索图片的搜索意图信息包括:以文本形式描述的价格信息,如具体文本内容可以是:该儿童读物的价格,以语音形式描述的出版社信息,如具体语音内容可以是:该儿童读物的出版社是哪家,以及包括以视频形式描述的内容信息,如具体视频内容可以是:连续播放的该儿童读物中的插图。
根据搜索意图信息在实现形式上的不同,获取搜索意图信息的方式也会有所不同。对于搜索意图信息中的文本信息,从用户角度来说,可以通过各种文本输入工具,如键盘、鼠标、输入笔、触摸屏等,输入文本信息;从图片搜索装置角度来说,可以接收用户通过各种文本输入工具输入的文本信息。对于搜索意图信息中的语音信息,从用户角度来说,可以通过语音录制模块,例如麦克风输入语音信息;从图片搜索装置角度来说,可以获取语音录制模块所录制的语音信息。对于搜索意图信息中的视频信息,从用户角度来说,可以通过视频拍摄模块,如摄像头,进行拍摄;从图片搜索装置角度来说,可以获取视频拍摄模块所拍摄的视频信息。
在步骤101,即获取待搜索图片及针对待搜索图片的搜索意图信息 的一种可选实施方式中,可以先启动图片搜索装置,进入搜索页面;然后,图片搜索装置获得待搜索图片;之后图片搜索装置接收用户输入的搜索意图信息。
具体的,可以在搜索页面设置拍摄按钮(该按钮可以是一照相机图标),用于供用户发出拍照指令。当需要进行搜索时,用户可以启动搜索功能(即启动图片搜索装置),启动搜索功能的表现方式为进入图片搜索装置提供的搜索页面。然后,用户通过搜索页面上的拍照按钮发出拍照指令;图片搜索装置接收用户发出的拍照指令,根据该指令启动拍照模块,对待识别对象进行拍照,以获得待搜索图片。具体的,用户可以用拍照模块对准待识别对象,以对待识别对象进行拍照。接下来,图片搜索装置接收用户输入的搜索意图信息。
在图片搜索装置接收用户输入的搜索意图信息的一具体实施方式中,图片搜索装置在获得待搜索图片后,自动启动录音模块对用户的语音进行录制,以获得语音形式的搜索意图信息。
在图片搜索装置接收用户输入的搜索意图信息的另一具体实施方式中,图片搜索装置可以通过同时通过具有摄像功能的拍照模块录制一段视频流,获取该视频流中的音频信息作为搜索意图信息。其中,该视频流中的图片为待搜索图片,该视频流中的音频信息为语音形式的搜索意图信息。
在图片搜索装置接收用户输入的搜索意图信息的又一具体实施方式中,可以在搜索页面设置文本输入框、语音录制按钮和视频拍摄按钮中的至少一个,用于供用户输入搜索意图信息。基于此,用户在拍照结束后,通过搜索页面上的文本输出框输入搜索意图信息;图片搜索装置在 获得待搜索图片后,接收用户输入的搜索意图信息。或者,用户在拍照结束后,通过搜索页面上的录音按钮发出录音指令,图片搜索装置在获得待搜索图片后,接收录音指令,启动语音录音模块,对表达用户搜索意图的语音进行录制,以获得语音形式的搜索意图信息。或者,用户在拍照结束后,通过搜索页面上的视频按钮发出视频拍摄指令,图片搜索装置在获得待搜索图片后,接收视频拍摄指令,启动视频拍摄模块,对表达用户搜索意图的视频进行拍摄,以获得视频形式的搜索意图信息。
在步骤101,即获取待搜索图片及针对待搜索图片的搜索意图信息的另一种可选实施方式中,实时监测用户输入的原始信息;根据该原始信息判断用户是否有搜索需求;当确定用户有搜索需求时,进入搜索页面,获得待搜索图片,并将原始信息作为针对待搜索图片的搜索意图信息。
具体的,图片搜索装置实时监测用户输入的原始信息,根据监测到的原始信息判断用户是否具有搜索需求。其中,用户输入的原始信息可以包括:文本信息、语音信息和视频信息中的至少一种。
在一种实施方式中,图片搜索装置可以预先设定搜索需求词,搜索需求词是反应用户具有搜索需求的词语或句子,例如搜索需求词可以是类似“去哪”“多少钱”、“这种蔬菜的原产地是哪”、“这部电影哪天上映”、“几号了”等词语或句子。基于此,图片搜索装置具体可以判断监测到的原始信息是否属于预设的搜索需求词;当判断结果为是时,确定用户有搜索需求;当判断结果为否时,确定用户没有搜索需求。
当判断出用户具有搜索需求时,图片搜索装置启动搜索功能,启动 搜索功能的表现方式为进入搜索页面。之后,图片搜索装置对待识别对象进行拍照,以获取待搜索图片,并将监测到的原始信息作为待搜索图片的搜索意图信息。
在一可选实施方式中,图片搜索装置在判断出用户具有搜索需求后,自动启动拍照模块,对待识别对象进行拍照,以获得待搜索图片。在该实施方式中,并不限定图片搜索装置启动拍照模块和进入搜索页面的先后顺序。
在另一可选实施方式中,可以在搜索页面设置拍摄按钮(该按钮可以是一照相机图标),用于供用户发出拍照指令。图片搜索装置在判断出用户具有搜索需求后,先进入搜索页面。用户通过搜索页面上的拍照按钮发出拍照指令;图片搜索装置接收用户发出的拍照指令,根据该指令启动拍照模块,对待识别对象进行拍照,以获得待搜索图片。
在步骤102,即根据上述待搜索图片和上述搜索意图信息,获得搜索结果的一种可选实施方式中,图片搜索装置可以先根据上述待搜索图片进行搜索,获得初始搜索结果,根据上述搜索意图信息,在初始搜索结果中进行二次搜索,获得最终的搜索结果。
在步骤102,即根据上述待搜索图片和上述搜索意图信息,获得搜索结果的另一种可选实施方式中,图片搜索装置提取待搜索图片的特征信息;对特征信息和搜索意图信息进行合并处理,以获得搜索关键词;直接根据该搜索关键词进行搜索,以获得搜索结果。
值得说明的是,在上述合并处理过程中,若搜索意图信息包括语音信息,则可以将语音信息转换成文本信息,再将特征信息与文本信息进 行合并处理。
在本公开的实施例中,可以采用很多通用的提取算法提取待搜索图片的特征,例如可以采用词袋(bag of word,BOW)算法,提取待搜索图片中的不定个数的特征,每个特征对应一个特征向量,这样可以从一张图片中提取多个特征。
进一步可选的,图片搜索装置可以获取特征信息对应的倒排索引和搜索意图信息对应的倒排索引;之后,对特征信息对应的倒排索引和搜索意图信息对应的倒排索引进行加权处理,获得上述搜索关键词。
例如,图片搜索装置可以分别采用特征信息和搜索意图信息对倒排索引库进行检索,从而获得特征信息对应的倒排索引和搜索意图信息对应的倒排索引。其中,特征信息和搜索意图信息可以共用一个倒排索引库,也可以使用独立的倒排索引库。
可选的,上述特征信息和上述搜索意图信息一般会对应多个倒排索引,则可以分别获取特征信息和上述搜索意图信息对应的最前面的N个倒排索引,对这2N个倒排索引进行加权处理;然后对加权处理结果进行排序,选择排在最前面的M个加权处理结果作为搜索关键词。其中,N和M均为自然数,且M小于或等于N。
采用本公开实施例提供的方法,当用户在乘坐地铁时看到某个电影的海报后,若想了解该电影的放映及票价等信息,用户可以拿起手机进入搜索页面,拍下这个海报并录制询问放映影院及票价的语音信息,之后采用BOW或其他提取算法提取出该海报的特征信息,并将语音信息转换为文本信息,将特征信息与文本信息进行合并,生成搜索关键词,利用搜索关键词进行搜索,获得与该电影的放映及票价有关的搜索结果, 将所获得的搜索结果展示给用户。或者
采用本公开实施例提供的方法,当用户在乘坐地铁时看到某个电影的海报后,若想了解该电影的放映及票价等信息,用户可以开启手机上的录音功能,录制询问放映影院及票价的语音信息,当根据该语音信息判断出用户有搜索需求时,进入搜索页面并启动拍照模块(如摄像头),用户通过拍照模块拍下这个海报,之后采用BOW或其他提取算法提取出该海报的特征信息,并将语音信息转换为文本信息,将特征信息与文本信息进行合并,生成搜索关键词,利用搜索关键词进行搜索,获得与该电影的放映及票价有关的搜索结果,将所获得的搜索结果展示给用户。
由上述可见,本申请同时根据待搜索图片和搜索意图信息进行搜索,所获得的搜索结果是最接近用户搜索意图的搜索结果,与现有技术相比,极大的降低了搜索结果的数量,因此可以提高用户在搜索结果中获取有用信息的效率。
需要说明的是,对于前述的各方法实施例,为了简单描述,故将其都表述为一系列的动作合并,但是本领域技术人员应该知悉,本公开并不受所描述的动作顺序的限制,因为依据本公开,某些步骤可以采用其他顺序或者同时进行。其次,本领域技术人员也应该知悉,说明书中所描述的实施例均属于优选实施例,所涉及的动作和模块并不一定是本公开所必须的。
在上述实施例中,对各个实施例的描述都各有侧重,某个实施例中没有详述的部分,可以参见其他实施例的相关描述。
图2为本公开一实施例提供的图片搜索装置的结构示意图。如图2所示,该方法包括:获取模块21和搜索模块22。
获取模块21,用于获取待搜索图片及针对所述待搜索图片的搜索意图信息。
搜索模块22,用于根据获取模块21获取的待搜索图片和搜索意图信息,获得搜索结果。
在一可选实施方式中,获取模块21具体用于:
进入搜索页面;
获得所述待搜索图片;
接收用户输入的所述搜索意图信息。
在一可选实施方式中,获取模块21具体用于:
实时监测用户输入的原始信息;
根据所述原始信息判断所述用户是否有搜索需求;
当确定所述用户有搜索需求时,进入搜索页面,获得所述待搜索图片,并将所述原始信息作为针对所述待搜索图片的搜索意图信息。
进一步,获取模块21在根据所述原始信息判断所述用户是否有搜索需求时,具体用于:
判断所述原始信息是否属于预设的搜索需求词;
当判断结果为是时,确定所述用户有搜索需求;
当判断结果为否时,确定所述用户没有搜索需求。
可选的,上述搜索意图信息可以包括:文本信息、语音信息和视频信息中的至少一种。
在一可选实施方式中,搜索模块42具体用于:提取所述搜索图片的特征信息;对所述特征信息和所述搜索意图信息进行合并处理,以获得搜索关键词;根据所述搜索关键词进行搜索,以获得搜索结果。
进一步可选的,搜索模块42在对上述特征信息和搜索意图信息进行组合合并处理时,具体可用于:获取特征信息对应的倒排索引和搜索意图信息对应的倒排索引;对特征信息对应的倒排索引和搜索意图信息对应的倒排索引进行加权处理,获得搜索关键词。
例如,搜索模块42可以分别采用特征信息和搜索意图信息对倒排索引库进行检索,从而获得特征信息对应的倒排索引和搜索意图信息对应的倒排索引。其中,特征信息和搜索意图信息可以共用一个倒排索引库,也可以使用独立的倒排索引库。
例如,在特征信息和上述搜索意图信息分别对应多个倒排索引的情况下,搜索模块42具体可以分别获取特征信息和上述搜索意图信息对应的最前面的N个倒排索引,对这2N个倒排索引进行加权处理;然后对加权处理结果进行排序,选择排在最前面的M个加权处理结果作为搜索关键词。其中,N和M均为自然数,且M小于或等于N。
本实施例提供的图片搜索装置,获取待搜索图片及针对待搜索图片的搜索意图信息,同时根据待搜索图片和搜索意图信息获得搜索结果。由于本实施例的图片搜索装置可以同时根据待搜索图片和搜索意图信息进行搜索,其所获得的搜索结果是最接近用户搜索意图的搜索结果,与现有技术相比,极大的降低了搜索结果的数量,因此可以提高用户在搜索结果中获取有用信息的效率。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的系统,装置和单元的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
在本公开所提供的几个实施例中,应该理解到,所揭露的系统,装 置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本公开各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用硬件加软件功能单元的形式实现。
上述以软件功能单元的形式实现的集成的单元,可以存储在一个计算机可读取存储介质中。上述软件功能单元存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)或处理器(processor)执行本公开各个实施例所述方法的部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
最后应说明的是:以上实施例仅用以说明本公开的技术方案,而非 对其限制;尽管参照前述实施例对本公开进行了详细的说明,本领域的普通技术人员应当理解:其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分技术特征进行等同替换;而这些修改或者替换,并不使相应技术方案的本质脱离本公开各实施例技术方案的精神和范围。

Claims (16)

  1. 一种图像搜索方法,其特征在于,包括:
    获取待搜索图片及针对所述待搜索图片的搜索意图信息;
    根据所述待搜索图片和所述搜索意图信息,获得搜索结果。
  2. 根据权利要求1所述的方法,其特征在于,所述获取待搜索图片及针对所述待搜索图片的搜索意图信息,包括:
    进入搜索页面;
    获得所述待搜索图片;
    接收用户输入的所述搜索意图信息。
  3. 根据权利要求1或2所述的方法,其特征在于,所述获取待搜索图片及针对所述待搜索图片的搜索意图信息,包括:
    实时监测用户输入的原始信息;
    根据所述原始信息判断所述用户是否有搜索需求;
    当确定所述用户有搜索需求时,进入搜索页面,获得所述待搜索图片,并将所述原始信息作为针对所述待搜索图片的搜索意图信息。
  4. 根据权利要求3所述的方法,其特征在于,所述根据所述原始信息判断所述用户是否有搜索需求,包括:
    判断所述原始信息是否属于预设的搜索需求词;
    当判断结果为是时,确定所述用户有搜索需求;
    当判断结果为否时,确定所述用户没有搜索需求。
  5. 根据权利要求1-4任一项所述的方法,其特征在于,所述搜索意图信息包括:文本信息、语音信息和视频信息中的至少一种。
  6. 根据权利要求1-5任一项所述的方法,其特征在于,所述根据所 述待搜索图片和所述搜索意图信息,获得搜索结果,包括:
    提取所述待搜索图片的特征信息;
    对所述特征信息和所述搜索意图信息进行合并处理,以获得搜索关键词;
    根据所述搜索关键词进行搜索,以获得搜索结果。
  7. 根据权利要求6所述的方法,其特征在于,所述对所述特征信息和所述搜索意图信息进行合并处理,以获得搜索关键词,包括:
    获取所述特征信息对应的倒排索引和所述搜索意图信息对应的倒排索引;
    对所述特征信息对应的倒排索引和所述搜索意图信息对应的倒排索引进行加权处理,获得所述搜索关键词。
  8. 一种图像搜索装置,其特征在于,包括:
    获取模块,用于获取待搜索图片及针对所述待搜索图片的搜索意图信息;
    搜索模块,用于根据所述待搜索图片和所述搜索意图信息,获得搜索结果。
  9. 根据权利要求8所述的装置,其特征在于,所述获取模块具体用于:
    进入搜索页面;
    获得所述待搜索图片;
    接收用户输入的所述搜索意图信息。
  10. 根据权利要求8或9所述的装置,其特征在于,所述获取模块具体用于:
    实时监测用户输入的原始信息;
    根据所述原始信息判断所述用户是否有搜索需求;
    当确定所述用户有搜索需求时,进入搜索页面,获得所述待搜索图片,并将所述原始信息作为针对所述待搜索图片的搜索意图信息。
  11. 根据权利要求10所述的装置,其特征在于,所述获取模块进一步具体用于:
    判断所述原始信息是否属于预设的搜索需求词;
    当判断结果为是时,确定所述用户有搜索需求;
    当判断结果为否时,确定所述用户没有搜索需求。
  12. 根据权利要求8-11任一项所述的装置,其特征在于,所述搜索意图信息包括:文本信息、语音信息和视频信息中的至少一种。
  13. 根据权利要求8-12任一项所述的装置,其特征在于,所述搜索模块具体用于:
    提取所述待搜索图片的特征信息;
    对所述特征信息和所述搜索意图信息进行合并处理,以获得搜索关键词;
    根据所述搜索关键词进行搜索,以获得搜索结果。
  14. 根据权利要求13所述的装置,其特征在于,所述搜索模块进一步具体用于:
    获取所述特征信息对应的倒排索引和所述搜索意图信息对应的倒排索引;
    对所述特征信息对应的倒排索引和所述搜索意图信息对应的倒排索引进行加权处理,获得所述搜索关键词。
  15. 一种设备,包括:
    一个或者多个处理器;
    存储器;
    一个或者多个程序,所述一个或者多个程序存储在所述存储器中,当被所述一个或者多个处理器执行时:
    获取待搜索图片及针对所述待搜索图片的搜索意图信息;
    根据所述待搜索图片和所述搜索意图信息,获得搜索结果。
  16. 一种非易失性计算机存储介质,所述非易失性计算机存储介质存储有一个或者多个程序,当所述一个或者多个程序被一个设备执行时,使得所述设备:
    获取待搜索图片及针对所述待搜索图片的搜索意图信息;
    根据所述待搜索图片和所述搜索意图信息,获得搜索结果。
PCT/CN2015/094339 2015-05-18 2015-11-11 图片搜索方法、装置、设备及非易失性计算机存储介质 WO2016184051A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US15/524,544 US20170337222A1 (en) 2015-05-18 2015-11-11 Image searching method and apparatus, an apparatus and non-volatile computer storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510253333.X 2015-05-18
CN201510253333.XA CN104881451A (zh) 2015-05-18 2015-05-18 图片搜索方法及装置

Publications (1)

Publication Number Publication Date
WO2016184051A1 true WO2016184051A1 (zh) 2016-11-24

Family

ID=53948944

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/094339 WO2016184051A1 (zh) 2015-05-18 2015-11-11 图片搜索方法、装置、设备及非易失性计算机存储介质

Country Status (3)

Country Link
US (1) US20170337222A1 (zh)
CN (1) CN104881451A (zh)
WO (1) WO2016184051A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109783679A (zh) * 2019-01-14 2019-05-21 广东小天才科技有限公司 一种学习辅助方法及学习设备

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104881451A (zh) * 2015-05-18 2015-09-02 百度在线网络技术(北京)有限公司 图片搜索方法及装置
CN107305566B (zh) * 2016-04-21 2019-10-18 北京搜狗科技发展有限公司 一种为搜索信息匹配图片的方法及装置
CN106980640B (zh) * 2017-02-08 2020-04-24 网易(杭州)网络有限公司 针对照片的交互方法、设备和计算机可读存储介质
TWI647580B (zh) * 2017-06-01 2019-01-11 正修學校財團法人正修科技大學 提升文圖搜尋結果相契合最佳化的搜尋過濾方法
CN107944954A (zh) * 2017-11-15 2018-04-20 联想(北京)有限公司 信息处理方法及其装置
CN107895050A (zh) * 2017-12-07 2018-04-10 联想(北京)有限公司 图片搜索方法及系统
CN110119461B (zh) * 2018-01-25 2022-01-14 阿里巴巴(中国)有限公司 一种查询信息的处理方法及装置
CN109348275B (zh) * 2018-10-30 2021-07-30 百度在线网络技术(北京)有限公司 视频处理方法和装置
CN109756676B (zh) * 2019-01-16 2021-06-25 广东小天才科技有限公司 一种图像处理方法及电子设备
CN109800301B (zh) * 2019-01-23 2020-12-01 广东小天才科技有限公司 一种薄弱知识点的挖掘方法及学习设备
CN112541091A (zh) * 2019-09-23 2021-03-23 杭州海康威视数字技术股份有限公司 图像搜索方法、装置、服务器和存储介质
CN111638846A (zh) * 2020-05-26 2020-09-08 维沃移动通信有限公司 图像识别方法、装置及电子设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103064936A (zh) * 2012-12-24 2013-04-24 北京百度网讯科技有限公司 一种基于语音输入的图像信息提取分析方法及装置
CN103778227A (zh) * 2014-01-23 2014-05-07 西安电子科技大学 从检索图像中筛选有用图像的方法
CN104462325A (zh) * 2014-12-02 2015-03-25 百度在线网络技术(北京)有限公司 搜索推荐方法及装置
CN104881451A (zh) * 2015-05-18 2015-09-02 百度在线网络技术(北京)有限公司 图片搜索方法及装置

Family Cites Families (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7555705B2 (en) * 2003-09-10 2009-06-30 Microsoft Corporation Annotation management in a pen-based computing system
US7346613B2 (en) * 2004-01-26 2008-03-18 Microsoft Corporation System and method for a unified and blended search
CN101329677A (zh) * 2008-05-07 2008-12-24 裴亚军 基于图像内容的图像搜索引擎
WO2011091402A1 (en) * 2010-01-25 2011-07-28 Justin Mason Voice electronic listening assistant
US8447752B2 (en) * 2010-09-16 2013-05-21 Microsoft Corporation Image search by interactive sketching and tagging
CN102467541B (zh) * 2010-11-11 2016-06-15 深圳市世纪光速信息技术有限公司 一种情境搜索方法及系统
CN102135985B (zh) * 2011-01-28 2013-03-06 百度在线网络技术(北京)有限公司 调用第三方搜索引擎搜索结果的搜索方法和搜索系统
CN102207966B (zh) * 2011-06-01 2013-07-10 华南理工大学 基于对象标签的视频内容快速检索方法
CN102521258A (zh) * 2011-11-18 2012-06-27 百度在线网络技术(北京)有限公司 一种用于提供壁纸图片的方法与设备
CN102609458B (zh) * 2012-01-12 2015-08-05 北京搜狗信息服务有限公司 一种图片推荐方法和装置
CN102708185A (zh) * 2012-05-11 2012-10-03 广东欧珀移动通信有限公司 一种图片语音搜索方法
CN103793434A (zh) * 2012-11-02 2014-05-14 北京百度网讯科技有限公司 一种基于内容的图片搜索方法和装置
CN103995848B (zh) * 2014-05-06 2017-04-05 百度在线网络技术(北京)有限公司 图片搜索方法及装置
CN104462510B (zh) * 2014-12-22 2018-09-11 北京奇虎科技有限公司 基于用户搜索意图的搜索方法及装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103064936A (zh) * 2012-12-24 2013-04-24 北京百度网讯科技有限公司 一种基于语音输入的图像信息提取分析方法及装置
CN103778227A (zh) * 2014-01-23 2014-05-07 西安电子科技大学 从检索图像中筛选有用图像的方法
CN104462325A (zh) * 2014-12-02 2015-03-25 百度在线网络技术(北京)有限公司 搜索推荐方法及装置
CN104881451A (zh) * 2015-05-18 2015-09-02 百度在线网络技术(北京)有限公司 图片搜索方法及装置

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109783679A (zh) * 2019-01-14 2019-05-21 广东小天才科技有限公司 一种学习辅助方法及学习设备
CN109783679B (zh) * 2019-01-14 2021-03-12 广东小天才科技有限公司 一种学习辅助方法及学习设备

Also Published As

Publication number Publication date
US20170337222A1 (en) 2017-11-23
CN104881451A (zh) 2015-09-02

Similar Documents

Publication Publication Date Title
WO2016184051A1 (zh) 图片搜索方法、装置、设备及非易失性计算机存储介质
US10789525B2 (en) Modifying at least one attribute of an image with at least one attribute extracted from another image
US10810252B2 (en) Searching using specific attributes found in images
EP2402867B1 (en) A computer-implemented method, a computer program product and a computer system for image processing
US9898847B2 (en) Multimedia picture generating method, device and electronic device
US9317531B2 (en) Autocaptioning of images
CN102609458B (zh) 一种图片推荐方法和装置
JP2008257460A (ja) 情報処理装置、情報処理方法、およびプログラム
Le et al. NII-HITACHI-UIT at TRECVID 2016.
JP2019185738A (ja) コンテンツメディアにテキスト要約を関連付けるためのシステム及び方法、プログラム、及びコンピュータ装置
WO2015188719A1 (zh) 结构化数据与图片的关联方法与关联装置
Wang et al. JIGSAW: interactive mobile visual search with multimodal queries
CN104239568A (zh) 一种基于图片说明的快速照片搜索系统和方法
GB2542891A (en) Modifying at least one attribute of an image with at least one attribute extracted from another image
US10321167B1 (en) Method and system for determining media file identifiers and likelihood of media file relationships
US20160012078A1 (en) Intelligent media management system
Trang-Trung et al. Flexible interactive retrieval system 2.0 for visual lifelog exploration at LSC 2021
JP2017021672A (ja) 検索装置
JP5794001B2 (ja) 情報検索方法、情報検索装置および情報検索プログラム
CN113672768A (zh) 用于内容发现的人工智能
El-Saban et al. Seamless annotation and enrichment of mobile captured video streams in real-time
Rabbath et al. Multimedia retrieval in social networks for photo book creation
WO2015185479A1 (en) Method of and system for determining and selecting media representing event diversity
Lin et al. Smartphone landmark image retrieval based on Lucene and GPS
JP7287459B2 (ja) 発話画像化装置、発話画像化方法及びプログラム

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15892427

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15892427

Country of ref document: EP

Kind code of ref document: A1