CN102955848A - Semantic-based three-dimensional model retrieval system and method - Google Patents
Semantic-based three-dimensional model retrieval system and method Download PDFInfo
- Publication number
- CN102955848A CN102955848A CN2012104213402A CN201210421340A CN102955848A CN 102955848 A CN102955848 A CN 102955848A CN 2012104213402 A CN2012104213402 A CN 2012104213402A CN 201210421340 A CN201210421340 A CN 201210421340A CN 102955848 A CN102955848 A CN 102955848A
- Authority
- CN
- China
- Prior art keywords
- semantic
- concept
- semantic concept
- dimensional model
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a semantic-based three-dimensional model retrieval system and method. The system comprises a semantic retrieval module and a semantic database. The semantic retrieval module is used to segment and filter semanteme of a retrieved sentence so as to obtain a semantic concept. The method comprises the steps of inquiring the synonymous semantic concept in the WordNet according to the semantic concept, calculating the similarity between the semantic concept and the inquired semantic concept and screening the semantic concept according to the similarity to obtain other semantic concepts associated with the semantic concept; and combining the semantic concept with other associated semantic concepts to retrieve corresponding three-dimensional information in the semantic database to obtain the retrieval result. According to the three-dimensional model retrieval system and method provided by the invention, the semantic gap of high-level semantic information and low-level semantic information is complemented, the retrieval range of the three-dimensional model is expanded, and the retrieval precision is improved.
Description
Technical field
The present invention relates generally to technical field of information retrieval, is specifically related to a kind of three-dimensional model searching system and method for semantic-based.
Background technology
Along with the rapid growth of internet information and multimedia technology popularizing in daily life, information retrieval not only is confined to text retrieval, but can carry out multimedia retrieval according to multimedia file, namely progressively to Directional Extensions such as image retrieval, audio retrieval, video frequency searchings.Simultaneously, for the ease of improving people to the use of searching system, text retrieval also should progressively be merged semantic meaning, thereby dwindles the semantic gap of high-level semantic and low layer data.
Three-dimensional model progressively is applied to a plurality of fields such as industry, design industry, microorganism industry.Three-dimensional model can be vivid structure and the composition of expression object, but because data volume is large, so it can not be retrieved as text message fast.A large amount of three-dimensional models repeats to generate, storage with regard to having caused for this, is not easy to reusing of three-dimensional model.Present stage, the content-based retrieval method is mainly adopted in the retrieval of three-dimensional model.Its essence is according to the visual signature of three-dimensional model and set up aspect indexing, the aspect indexing of setting up meets the human vision requirement, and namely result for retrieval visually is similar.Content-based three-dimensional model search mainly adopts following steps: at first, carry out pre-service by the visual signature to three-dimensional model to be retrieved, extract the data of vacuate three-dimensional model; Subsequently the data of vacuate are carried out the three-dimensional model feature extraction, extract the proper vector of this three-dimensional model; Calculate at last the similarity degree of the proper vector in three-dimensional model to be retrieved and the characteristic vector data storehouse, in 3 d model library, obtain result for retrieval according to similarity degree.
Current, three-dimensional model searching system mainly comprises text retrieval, the retrieval of two-dimentional sketch and content-based three kinds of modes of three-dimensional model search.Wherein, the retrieval of two-dimentional sketch and the content-based three-dimensional model search essential characteristic that all is based on image graphics is retrieved.And text retrieval still rests on the semantic concept mark that three-dimensional model is had and carries out keyword retrieval, is not semantic retrieval truly.Wherein, semantic concept mark is a kind of of text key word mark, be user or expert to a kind of textual description with semantic concept of three-dimensional model, often formed by brief word, can be used for the three-dimensional model semantic retrieval.Its objective is by three-dimensional model being carried out the mark on the semantic concept, namely can be three-dimensional model and increase semantic concept, the method for searching three-dimension model of semantic-based finally can be provided for the user.For text retrieval, if two complete dissimilar three-dimensional models, but but having identical semantic concept mark, when retrieving so this semantic concept mark, these two three-dimensional models will be exported as result for retrieval jointly.On the contrary, if the semantic concept of three-dimensional model mark is not identical, but relevant at semantic level, for example, retrieval " furniture " word should have cupboard, chair, desk etc., and is not a class three-dimensional model that " furniture " semantic concept mark only occurs having.This situation can be described to " semantic gap " problem, i.e. the generation gap of high-level semantic and low layer three-dimensional model.This situation often comes across in the middle of the text retrieval, thereby causes the as a result precision of text retrieval not high.
The works and expressions for everyday use that the three-dimensional model search of semantic-based can be avoided " semantic gap " problem, can be close to the users, for example: the user inputs " quadrupeds ", system will retrieve " dog ", " cat " and three-dimensional models such as " horses ", and the content that comprises on the semantic concept of these three-dimensional models mark may be " dog ", " cat " and item names such as " horses ", and mark " quadrupeds ", but belong to " quadrupeds " this semantic concept in these contents semantically.At present, existing 3D modelling system is single for the mark that the semantic concept of three-dimensional model carries out, the mark that perhaps just based on expert's angle three-dimensional model is carried out, because when the user really uses, because everyone is to the difference of the view of things, angle, stock of knowledge, and the mode that the user is convenient to take to press close to own works and expressions for everyday use provides Search Requirement, thereby causes can producing larger deviation when carrying out the retrieval of semantic concept mark.
The three-dimensional model search of existing use semantic-based has Chinese patent application number 200810115698.6 disclosed three-dimensional model searching system and methods, and Fig. 1 shows the structural drawing of this searching system.This patent has proposed a kind of multi-level relevance feedback algorithm MRF, MRF mainly utilizes feedback mechanism to obtain the high-layer semantic information of user feedback, by constantly decomposing, high-layer semantic information is converted into weight relationship between the different characteristic vector and the correlation information of proper vector inside, namely upgrades according to user's feedback information weighted value and the correlation information to each proper vector of retrieval model.Utilize weighted value and the correlation information of each proper vector after upgrading, calculate the similarity distance between retrieval model and the Matching Model, the three-dimensional model similar to retrieval model apart from acquisition according to similarity.Thereby this method still is based on the retrieval on the three-dimensional model content, judges to assist a ruler in governing a country by user's correlativity and revises the three-dimensional model proper vector, can not finish semantic-based to the retrieval of three-dimensional model.
Summary of the invention
For the problems referred to above, the present invention proposes a kind of three-dimensional model searching system and method for semantic-based, solve the semantic gap problem of high-layer semantic information and low layer three-dimensional model characteristic information in the mode that adopts semantic retrieval truly.
According to one embodiment of the invention, a kind of three-dimensional model searching system is provided, comprise semantic retrieval module and semantic database, wherein:
Described semantic retrieval module also comprises: participle submodule, semantic association calculating sub module and information are transmitted submodule, wherein
The retrieve statement that described participle submodule is used for the user is inputted carries out semantic participle and filtration, and obtains semantic concept;
Described semantic association calculating sub module is used for according to described semantic concept at a plurality of semantic concepts of semantic dictionary WordNet inquiry with its synonym, and described semantic concept and the semantic concept that inquires are carried out similarity calculate, carry out the semantic concept screening according to this similarity, obtain the semantic concept relevant with described semantic concept;
Described information is transmitted submodule and is used for retrieving corresponding three-dimensional model information in conjunction with described semantic concept and relative semantic concept at semantic database, and obtains result for retrieval;
Described semantic database is used for depositing the corresponding relation of semantic concept information and semantic concept and three-dimensional model.
In one embodiment, described system also comprises three-dimensional modeling data storehouse and display module, wherein
Described three-dimensional modeling data storehouse is used for storage about the data of three-dimensional model;
Described display module is used for supporting that the user submits retrieve statement to, and this retrieve statement is passed to described semantic retrieval module; And be used for transmitting the result for retrieval that submodule obtains according to described information, read the three-dimensional modeling data in the described three-dimensional modeling data storehouse, and show three-dimensional modeling data;
Described information is transmitted submodule result for retrieval is passed to display module.
In one embodiment, described semantic association calculating sub module is carried out the following formula of similarity calculating employing with described semantic concept and the semantic concept that inquires:
Wherein, two semantic concepts that A, B indicate to calculate, most recent co mmon ancestor node in the WordNet semantic dictionary of lcs (A, B) expression A, B, IC (lcs (A, B)) represents the information content value that this common ancestor's node has, structure factor
Wherein, depth (A) and depth (B) represent respectively semantic concept A, the B residing degree of depth in the WordNet semantic dictionary.
In one embodiment, described semantic association calculating sub module is carried out the semantic concept screening according to similarity, obtains the semantic concept relevant with described semantic concept and comprises:
According to predefined similarity threshold, to being screened by the resulting similarity result of semantic association computational algorithm, select the semantic concept with the similarity that is higher than described predefined similarity threshold by the semantic association calculating sub module,
In one embodiment, described information is transmitted submodule and is retrieved corresponding three-dimensional model information in conjunction with described semantic concept and relative semantic concept in semantic database, and the acquisition result for retrieval comprises:
Be present in the described semantic database if transmit the semantic concept that comes, then retrieved and returned the corresponding three-dimensional model information of this semantic concept;
If transmit next semantic concept not in semantic database, then do not return three-dimensional model information.
In one embodiment, described system also comprises the semantic tagger module, and this semantic tagger module comprises user interface;
The three-dimensional model semantic concept mark that the user submits to is examined by the data base administrator by user interface, check whether mark meets standard and do not repeat with existing mark, if by audit, then this semantic concept mark is added in the semantic concept mark of corresponding three-dimensional model in the described semantic database; Otherwise, then will return miscue information to display module.
In a further embodiment, described display module is used for also supporting that the user submits the semantic concept mark to, and the semantic concept mark that the user submits to is passed to the semantic tagger module.
According to one embodiment of the invention, a kind of method for searching three-dimension model based on described three-dimensional model searching system is proposed, comprising:
Step 1), retrieve statement is carried out semantic participle and filtration, obtain semantic concept;
Step 2), obtain relative semantic concept according to described semantic concept, in semantic database, retrieve corresponding three-dimensional model information according to described semantic concept and relative semantic concept, obtain result for retrieval.
In one embodiment, also comprise before the step 1):
Step 0), submit retrieve statement at display module.
According to one embodiment of the invention, step 2) after also comprise:
Step 3), result for retrieval is inquired about in the three-dimensional modeling data storehouse, obtained the data of three-dimensional model, and show this three-dimensional modeling data.
In one embodiment, obtaining relative semantic concept according to described semantic concept step 2) comprises:
Step 21), in the WordNet semantic dictionary, inquire a plurality of semantic concept vocabulary with its synonym according to described semantic concept;
Step 22), adopting following formula to carry out similarity described semantic concept and the semantic concept that inquires calculates:
Wherein, two semantic concepts that A, B indicate to calculate, most recent co mmon ancestor node in the WordNet semantic dictionary of lcs (A, B) expression A, B, IC (lcs (A, B)) represents the information content value that this common ancestor's node has, structure factor
Wherein, depth (A) and depth (B) represent respectively semantic concept A, the B residing degree of depth in the WordNet semantic dictionary.
Step 23) carries out the semantic concept screening according to the similarity of calculating gained, obtain the semantic concept relevant with described semantic concept.
In a further embodiment, carrying out the semantic concept screening according to the similarity of calculating gained step 23) comprises:
According to predefined similarity threshold, the similarity of gained is screened, select to have the semantic concept of the similarity that is higher than this similarity threshold as relevant semantic concept.
In one embodiment, step 2) in semantic database, retrieve corresponding three-dimensional model information according to described semantic concept and relative semantic concept in, obtain result for retrieval and comprise:
If these semantic concepts have been present in the semantic database, then from the semantic data library searching and return the corresponding three-dimensional model information of this semantic concept; As not existing, then do not return three-dimensional model information
In sum, the three-dimensional model searching system of semantic-based provided by the invention and method have following beneficial effect:
(1) only avoided limitation based on the retrieval mode of the content characteristic of three-dimensional model and text marking, the Global Information from different level Consideration of Three-dimensional models enlarged range of search
(2)。Remedy the semantic gap of high-layer semantic information and low layer three-dimensional model characteristic information, improved retrieval precision.
Description of drawings
Fig. 1 is the three-dimensional model searching system structural drawing of existing semantic-based;
Fig. 2 is the three-dimensional model searching system structural drawing of according to an embodiment of the invention semantic-based;
Fig. 3 is the semantic concept mark schematic diagram that adopts the RDF markup language;
Fig. 4 is the method for searching three-dimension model process flow diagram of according to an embodiment of the invention semantic-based.
Embodiment
Present invention is described below in conjunction with the drawings and specific embodiments.
According to one embodiment of the invention, provide a kind of three-dimensional model searching system of semantic-based.Fig. 2 shows the general structure of this system, comprises server end and client.Described server end comprises semantic retrieval module, semantic database, three-dimensional modeling data storehouse and semantic tagger module; Described client comprises display module.Wherein
Described semantic retrieval module also comprises: the participle submodule, and the retrieve statement that is used for the user is inputted carries out semantic participle and filtration, and obtains semantic concept; The semantic association calculating sub module, be used for inquiring with it a plurality of semantic concepts of synonym (being semantic concept vocabulary) according to described semantic concept at semantic dictionary WordNet, and described semantic concept and the semantic concept that inquires are carried out similarity calculate, carry out the semantic concept screening according to the similarity of calculating, obtain other semantic concepts relevant with described semantic concept; Information is transmitted submodule, is used for retrieving corresponding three-dimensional model information in conjunction with described semantic concept and the semantic concept relevant with described semantic concept at semantic database, obtains result for retrieval, and the result is passed to display module.
Described semantic database is used for the information (being the semantic concept mark) of storage semantic concept and the corresponding relation of semantic concept and three-dimensional model, i.e. semantic concept, corresponding a plurality of three-dimensional models.
Described three-dimensional modeling data storehouse is used for storage about data, three-dimensional model classified information, the two dimensional image of three-dimensional model and the essential information of three-dimensional model etc. of three-dimensional model.
Described semantic tagger module, the three-dimensional model semantic concept mark that is used for the user is submitted to is examined, check whether this mark (for example meets format standard, mark can not comprise numbers and symbols etc.) and with existing mark (namely, the semantic concept mark that this three-dimensional model has in the semantic database) do not repeat, if by audit, then this semantic concept mark is added in the described semantic database in should the semantic concept mark of three-dimensional model; Otherwise, then will return miscue information.
Described display module is used for every interactive operation of completing user and system, and it comprises: system's page is provided, support that the user submits retrieve statement to, and the retrieve statement that the user is submitted to is passed to the semantic retrieval module; Display page (can comprise the three-dimensional model information page) is provided, transmit result's (three-dimensional model information of retrieval gained that the submodule transmission comes according to the information in the semantic retrieval module, the results list that namely is made of the three-dimensional model title) data read the three-dimensional modeling data in the three-dimensional modeling data storehouse, and show this three-dimensional modeling data; Described display module also provides and adds the semantic concept mark page, supports the user to submit the semantic concept mark to, and the semantic concept mark that the user submits to is passed to the semantic tagger module.Described display module arranges prompt facility, can guides user operate.
Below the modules of the three-dimensional model searching system of semantic-based is described respectively.
One. the semantic retrieval module
The semantic retrieval module comprises participle submodule, semantic association calculating sub module and information transmission submodule.
1. described participle submodule comprises participle instrument and filter word tabulation.Described participle instrument is supported Chinese word segmentation and English string segmentation, and existing participle instrument comprises: ictclas4j, IKAnalyzer, dismember an ox as skillfully as a butcher etc.The groundwork of participle instrument be will input literal split according to part of speech and the meaning of a word, the result after the fractionation is a series of words, and automatically omits the word that ",, " etc. do not have actual semantic meaning.For example, the participle submodule reads the retrieve statement that the user submits at system's page: " three-dimensional model of quadrupeds ", and utilize the participle instrument that this statement is split into a plurality of words, such as " four limbs ", " animal ", " three-dimensional ", " model ".Then, the stop word in the word after the participle submodule will split according to filter word tabulation filters, and filters afterwards that the word (such as " four limbs ", " animal ") of gained can be used as semantic concept, and this semantic concept is passed to the semantic association calculating sub module.
2. the semantic association calculating sub module is at first inquiring a plurality of semantic concept vocabulary (for example 2-5) according to the semantic concept that obtains from word-dividing mode among semantic dictionary WordNet.WordNet is the English semantic dictionary by a kind of hierarchical structure of Princeton university exploitation.In WordNet, between semantic concept be by synonym (synonym), antisense (antonym), whole (holonym), partly (meronym), upper (hypernym), the next (hypony), contain multiple semantic relations such as (entailment) and make up relational network, whole WordNet is exactly the semantic network that is made into by numerous net.By interpretative tool (such as from the internet obtainable interpretative tool JAWL etc.) among WordNet, inquire the synonym semantic concept about described semantic concept (semantic concept that obtains from word-dividing mode).Then, will carry out similarity from participle the submodule semantic concept that obtains and the semantic concept that from WordNet, inquires and calculate, carry out the semantic concept screening according to the similarity of calculating gained, obtain other semantic concepts relevant with described semantic concept.
And then, will all be passed to information transmission submodule from semantic concept and the relative semantic concept that the participle submodule obtains.Wherein
The semantic association computational algorithm can adopt algorithm based on the information content, such as characteristic model that Lin algorithm and Rodriguez and Egenhofer propose etc.
According to one embodiment of present invention, provide W﹠amp; The IC algorithm is as the semantic association computational algorithm of semantic association calculating sub module, this W﹠amp; The IC algorithm is structured on the characteristic model of Rodriguez and Egenhofer (characteristic model described here is judged similarity degree between the semantic concept according to the quantity of the common feature set of semantic concept) based on the WordNet semantic dictionary.The information content (IC) refers to synonym set (synset, minimum unit among the WordNet, it is the integrated definition for a vocabulary, concept, synonym, directional information and the side-play amount etc. that comprise this word itself) frequency that the information that comprises occurs in specific corpus, can utilize the shared rate of the information content between two synset to calculate similarity.The method of present existing computing semantic similarity has the semantic-based dictionary and based on the semantic similarity computing method of the information content.Wherein, the semantic computation method of semantic-based dictionary utilizes two physical path lengths between the concept to judge between the two similarity value, the nearlyer similarity between the two of semantic distance is larger, yet but there is certain restrictive condition in this theory, for example for the concept of same path length between, abstract concept between semantic distance should greater than concrete concept between semantic distance.Thereby and based on the method for the information content relatively the information content between two concepts calculate between the two semantic similarity, yet the value of traditional information content all is a certain concept of calculating appears at the frequency in the large corpora (such as Blang's corpus, LOB, Kolhapur Corpus etc.), contain much information in the large corpora, frequency values can not accurately reflect semantic information, and depending on corpus unduly is a defective of the method.For the problem of prior art, W﹠amp provided by the invention; The IC algorithm utilizes characteristic model when the value of computing information content, only carry out the calculating of semantic distance for the hierarchical structure of WordNet.The WordNet data volume is less, and semantic structure is clear and semantic concept is accurate, can reduce the accuracy of computation complexity, quickening computing velocity, lifting result of calculation.
W﹠amp provided by the invention; The IC algorithm is based on traditional characteristic model by Rodriguez and Egenhofer proposition, and this model utilizes feature quantity to calculate similarity between two concepts, and computing formula is as follows:
Wherein A, B are two semantic concepts that will calculate similarity, and the word that namely comprises in two semantic concepts is as characteristic item, the characteristic set of composition; A ∩ B represents not only to appear among the A but also appears at feature set among the B, and A-B expresses among the present A, but does not appear at the characteristic set among the B, and in like manner B-A expresses among the present B but do not appear at characteristic set among the A;
In this scope is selected
Value be for make public characteristic to the importance of similarity greater than non-public characteristic, namely weakened non-public characteristic to the impact of similarity, promoted the accuracy of similarity.
Above-mentioned formula is the characteristic model that a concept similarity calculates, wherein, feature quantity and IC value have common part, all represent the quantity of information that a concept contains, the feature quantity of imagination semantic concept A is its IC value, concept A, the common trait quantity of B is the IC value of both most recent co mmon ancestor nodes (Least Common Subsumer is called for short LCS), then A ∩ B, | A-B|, | the computing formula of B-A| is as follows:
A∩B≈sim
res(A,B)=IC(lcs(A,B))
|A-B|=Feature((A)-(A∩B))≈IC(A)-IC(lcs(A,B))
|B-A|=Feature((B)-(A∩B))≈IC(B)-IC(lcs(A,B))
Bring above three formula into sim (A, B) and can obtain W﹠amp; The IC algorithmic formula:
Wherein lcs (A, B) represents semantic concept A, B most recent co mmon ancestor node in the WordNet semantic dictionary, and IC (lcs (A, B)) represents the information content (IC) value that this common ancestor's node has.Because it is not use corpus that IC value described herein is calculated prerequisite, wherein, the computing formula of structure factor a is as follows:
Wherein, the span of a still is 0 to 0.5, namely hypothesis still is public characteristic to the importance of similarity greater than non-public characteristic, wherein A, B are the characteristic sets of two semantic concepts, and depth (A) and depth (B) represent respectively A, the B residing degree of depth in the WordNet semantic dictionary.
W﹠amp; The IC algorithm is compared with traditional algorithm based on the information content, only rely on the hierarchical structure among the WordNet to come computing information content (IC) value, cost and the complexity calculated have been reduced, considering the impact on the similarity value of public characteristic factor and non-public characteristic factor simultaneously, the subjective judgement of finally calculating income value and people is also more approaching.
According to one embodiment of present invention, according to predefined similarity threshold, by the semantic association calculating sub module to being screened by the resulting similarity result of semantic association computational algorithm, select to have and be higher than the semantic concept that this presets the similarity of similarity threshold, selected semantic concept is relevant with splitting the semantic concept that obtains according to retrieve statement.The semantic concept of these relevant semantic concepts and retrieve statement fractionation all is passed to information transmission submodule.For example, this similarity threshold can be redefined for 85%, and split the semantic concept obtain according to retrieve statement and carry out after similarity calculates, select to be higher than 85% the corresponding semantic concept of similarity as relevant semantic concept.
3. information is transmitted submodule and is received the content (semantic concept and relative semantic concept that retrieve statement splits) that the transmission of semantic association calculating sub module comes, the semantic content that in semantic database three-dimensional model is had is retrieved, search semantic concept and three-dimensional model (namely according to existing semantic concept, finding the model name of corresponding three-dimensional model) corresponding to relative semantic concept that retrieve statement splits.
Because the semantic concept in the semantic database is limited, be present in the semantic database if transmit the semantic concept that comes, then retrieve and return the corresponding three-dimensional model information of this semantic concept model name of the three-dimensional model of result for retrieval (namely as); If transmit next semantic concept not in semantic database, then do not return three-dimensional model information.If need to show the result at display page, then result's (model name of three-dimensional model) is passed to display module, carried out again three-dimensional model semantic retrieval result's demonstration by display module.
Two. semantic database
Semantic database is used for depositing the corresponding relation of semantic concept information and semantic concept and three-dimensional model, i.e. semantic concept, corresponding a plurality of three-dimensional models, its storage mode be<semantic concept, model name 1 ..., model name n 〉.A semantic concept can be had by a plurality of models, and a plurality of model also can corresponding a plurality of semantic concepts, but semantic concept unique existence in database, can not repeat.According to semantic concept, in semantic database, can retrieve three-dimensional model corresponding to this semantic concept.
Three. the three-dimensional modeling data storehouse
The three-dimensional modeling data storehouse is used for storage three-dimensional modeling data, three-dimensional model classified information, the two dimensional image of three-dimensional model and the essential information of three-dimensional model.Wherein said essential information comprises: the boundary value of three-dimensional model name, three-dimensional model (x axle maximin, y axle maximin, z axle maximin), three-dimensional model barycentric coordinates value, three-dimensional model main shaft etc.In this database, model name is as major key, unique existence, not reproducible.According to unique model name, can retrieve the information of three-dimensional model, obtain three-dimensional model file, two dimensional image in Url and the essential information of server end, utilize these information, can be used as result for retrieval and be passed to display module and show.
Four. the semantic tagger module
The semantic tagger module supports the user to add the semantic concept mark for three-dimensional model, at first examine for the semantic concept mark that comes from the display module transmission, check and examine this mark whether meet standard and with semantic database in already present mark about this model whether repeat.If comprise numeral or symbol in this mark, perhaps comprise bad, invalid information in this mark, think that then this mark does not meet standard, do not pass through audit; In addition, if this model has had this mark in semantic database, think that then this mark is not by audit; Otherwise, then by audit.If by audit, then this semantic concept mark is added in the semantic concept mark of corresponding three-dimensional model in the semantic database: at first in semantic database, add this semantic concept, then according to<semantic concept, model name 1, ..., model name n〉storage mode, the model name of model is added on this semantic concept after, be back to display module with adding successful information at last, add semantic concept by the display module prompting user and mark successfully; If by audit, then do not do and add operation, interpolations failure information is back to display module, add this semantic concept by the display module prompting user and mark unsuccessfully, and require the user again to add or abandon interpolation.
According to one embodiment of present invention, the semantic tagger module also comprises user interface, is used for the semantic concept mark is offered semantic database managerial personnel audit, and examination result is returned the semantic tagger module.
Five. display module
Display module is used for every interactive operation of completing user and system, and display module arranges prompt facility, can guides user operates to comprise:
1. system's page is provided, support that the user submits retrieve statement to, and the retrieve statement that the user is submitted to is passed to the semantic retrieval module.System's page provides input retrieve statement square frame, submits the retrieve statement button in the square frame placed around.After the user inputted retrieve statement, click on submission button was committed to statement the semantic retrieval module of server end.
2. provide display page and three-dimensional model information page to show three-dimensional model.Display module can be retrieved according to result's (model name of three-dimensional model) that the transmission of semantic retrieval module comes the three-dimensional modeling data in the three-dimensional modeling data storehouse: utilize the model name of three-dimensional model, search three-dimensional model file Url corresponding to this model name, the information such as two dimensional image Url of three-dimensional model in database; According to Url, when showing this three-dimensional model, can demonstrate the three-dimensional picture of three-dimensional model two dimensional image and this model at the page; In addition, in the process that the user marks the semantic concept of three-dimensional model, also provide Presentation Function.
According to one embodiment of the invention, the display page of described display module shows the demonstration that can be divided into that three-dimensional model independently shows and three-dimensional model is tabulated of the mode of three-dimensional model.Wherein:
Three-dimensional model tabulation demonstration is mainly used in result for retrieval output and category is checked three-dimensional model.When showing result for retrieval, three-dimensional model can sort according to the large young pathbreaker's result for retrieval of the similarity of calculating (for example from high to low ordering), and the result for retrieval of highlighted demonstration the most similar (for example front 1-5).Original list shows the corresponding two dimensional image of three-dimensional model, and shows that three-dimensional model relevant information (such as the three-dimensional model name etc.) and part of semantic concept mark.The two dimensional image that the user clicks in the original list can enter the three-dimensional model information page, details (comprising three-dimensional model spatial information, coordinate axis value, center of gravity etc.) and two dimensional image that three-dimensional model is arranged in this page, button have the semantic concept of interpolation mark, 3-D display and return.
Check three-dimensional model for the ease of user's category, in one embodiment, display module can show, supports that with the tree structure classification user reads three-dimensional modeling data, utilizes WebGL to browse three-dimensional model with three-dimensional model, and supports the mouse rotary manipulation.
Three-dimensional model independently shows and is mainly used in checking independent three-dimensional model, calls the auxiliary three-dimensional model that shows of WebGL control.Mainly can trigger by the 3-D display button.In three-dimensional model stereo display frame, three-dimensional model has been removed the much informations such as color, texture, pattern, reduces the data volume of server process, is convenient to the user and checks fast model structure.Simultaneously, display mode is supported the user checks three-dimensional model by the mouse action adjustment angle, namely supports rotary manipulation, translation, zoom operations etc.
3. provide and add the semantic concept mark page, support the user to submit the semantic concept mark to, and the semantic concept mark that the user submits to is passed to the semantic tagger module.
Wherein, system's page provides the button of selecting the semantic concept mark, when the user enters system's page and selects the semantic concept mark.System enters the display page of three-dimensional model automatically, and for example the three-dimensional model original list is the tree-shaped classification chart of three-dimensional model on the left of the page, and the right side is used for showing three-dimensional model.The tree-shaped classification chart of three-dimensional model is according to the demonstration of the sort file in the three-dimensional modeling data storehouse, and each classification is as a file, and all categories is under the root root directory.Click arbitrary classification, the right side page shows two dimensional image and the essential information of all three-dimensional models under this classification, comprises three-dimensional model name, three-dimensional model space characteristics etc.The user can click arbitrary two dimensional image this moment, enter the corresponding three-dimensional model information page of this two dimensional image, the details of three-dimensional model are arranged in this page, and the semantic concept of interpolation mark and two buttons of 3-D display are arranged, can check three-dimensional model by three-dimensional display key.
In addition, the user can enter the interpolation semantic concept mark page by clicking interpolation semantic concept mark button.In one embodiment of the invention, the semantic concept mark page comprises three-dimensional model stereo display mode, supports mouse action and unspecified angle to check model; Also comprise semantic concept mark and add frame, need input text in this frame, text should not comprise symbol, but can comprise Chinese and English and numeral, and wherein a plurality of semantic concepts marks are used the space and separated; The semantic concept mark adds the frame top the Chinese and English information, helps the user correctly to add the semantic concept mark.The user is after clicking affirmation interpolation button, and this semantic concept mark is passed to the semantic tagger module of server end by the semantic concept mark page,
Semantic concept mark is the means by a kind of mark (Tag), in HTML or XML the metadata of resource, i.e. the semantic information process of getting up with corresponding resource relationship.Employed markup language can be the markup languages that is applicable to this well known in the art such as XML, RDF.
The example that adopts the RDF technology three-dimensional model to be added the semantic concept mark has been shown among Fig. 3.In three-dimensional model, can pass through RDF tlv triple<resource, attribute, value〉three-dimensional model is described, take Fig. 3 as example, following form provides the example that adopts RDF mark three-dimensional model semantic concept:
According to one embodiment of present invention, also provide a kind of method for searching three-dimension model based on above-mentioned three-dimensional model searching system, Fig. 4 shows the process flow diagram of the method, comprising:
The first step, at system's page input retrieve statement, click on submission button is submitted retrieve statement to or is selected click semantic concept mark button to carry out the semantic concept mark;
If in first step input retrieve statement and submission, then carry out following steps:
Second step carries out semantic participle to described retrieve statement, namely utilizes the participle instrument that retrieve statement is carried out participle, filtration, thereby obtains semantic concept.
The 3rd step obtained relative semantic concept according to described semantic concept, retrieved corresponding three-dimensional model information in conjunction with described semantic concept and relative semantic concept in semantic database.
If these semantic concepts have been present in the semantic database, then from the semantic data library searching and return the corresponding three-dimensional model information of this semantic concept (this result for retrieval is the model name of three-dimensional model); As not existing, then do not return three-dimensional model information.
Wherein, obtaining relative semantic concept according to described semantic concept comprises:
1. in the WordNet semantic dictionary, inquire and a plurality of semantic concept vocabulary of its synonym (being semantic concept, for example 2-5) according to described semantic concept, can for example inquire about by the available interpretative tool of network.
2. described semantic concept and the semantic concept that inquires being carried out similarity calculates.According to one embodiment of present invention, computing semantic similarity can adopt the method based on the information content, for example W﹠amp; The IC algorithm.
3. carry out the semantic concept screening according to the similarity of calculating gained, obtain other semantic concepts relevant with described semantic concept.According to one embodiment of present invention, according to predefined similarity threshold, similarity result is screened, select to have the semantic concept of the similarity that is higher than this similarity threshold as relevant semantic concept.For example, set 85% as similarity threshold.
The 4th step, according to the model name of the 3rd three-dimensional model that obtain of step, in the three-dimensional modeling data storehouse, inquire about, obtain the data of this three-dimensional model, again it is shown as the display module of result for retrieval in client.
Wherein, can adopt the three-dimensional model tabulation to show.Can sort according to the large young pathbreaker's result for retrieval of the similarity of calculating, and the most similar result for retrieval of highlighted demonstration.Display page (the one-level page) can be comprised of the form of 5 row, 4 row, the model name that comprises two dimensional image and the three-dimensional model of three-dimensional model in the form, click this image and can enter this three-dimensional model information page (the secondary page), the 3-D display that comprises this three-dimensional model in the secondary page, the operations such as convergent-divergent are carried out in support to three-dimensional model, be included in simultaneously the essential information of storing in the three-dimensional modeling data storehouse, such as boundary value, main shaft, center of gravity etc.
If select the semantic concept mark in the first step, then carry out following steps:
Second step, the three-dimensional model that user selection will mark, and enter the interpolation semantic concept mark page or leaf of display module.
In the 3rd step, the user submits the semantic concept mark at the semantic concept mark page, by the semantic concept mark page semantic concept mark of submitting to is passed to the semantic tagger module.
Wherein can adopt the markup language of XML, RDF for example to carry out the semantic tagger of three-dimensional model.
The 4th step, in the semantic tagger module, and the semantic concept that judge to obtain mark whether meet standard and with semantic database in already present mark about this model (for example whether repeat, can this semantic concept mark be offered the semantic data library manager by user interface examines), if comprise numeral or symbol in this mark, perhaps change mark and comprise bad, invalid information, think that then this mark does not meet standard, do not pass through audit; Perhaps this model has had this mark in semantic database, thinks that then this mark is not by audit; Otherwise, then by audit.
If by audit, by the semantic tagger module this semantic concept mark is added in the semantic database in the corresponding three-dimensional model semantic concept mark, comprising:
At first in semantic database, add this semantic concept, then according to<semantic concept, model name 1, ..., model name n〉storage mode, after the model name of model is added on this semantic concept, be back to display module with adding successful information at last, marked successfully by display module prompting user interpolation semantic concept; If by audit, then do not do and add operation, interpolations failure information is back to display module, add semantic concept by the display module prompting user and mark unsuccessfully, and can also require the user again to add or abandon interpolation.
In above-mentioned semantic concept mark process, the ability to express that the user add semantic concept of second step mark process need is described three-dimensional model by the people is analyzed and is processed for corresponding three-dimensional model mark object, usually needs through following process:
1. user profile, mark three-dimensional model to be marked.The user can provide the attribute that paragraph description and mark three-dimensional model should possess according to the understanding of the information such as the vision that three-dimensional model is had, content, structure, provides property value.By analysis and the description to the mark object, can obtain marking the key words content of object.
2. extract the concept of describing in the content.For described content, by with Wordnet in semantic concept compare one by one after, it is extracted in order to further process.For example in the three-dimensional model semantic tagger, " quadrupeds ", " ox ", " furniture ", etc. noun all be concept among the Wordnet, when analyzing, they are extracted.The series of concepts that obtain this moment can not really embody the semantic description content of three-dimensional model, but simply concept is enumerated and piled up, and therefore also needs to further process.
3. the extraction of property value in the mark object.Content for described in the three-dimensional model by a series of processing, obtains their described property values, thereby obtains the semantic object of three-dimensional model.In the extracting method of semantic example, the semantic concept marked content of three-dimensional model is exactly the property value composition of a plurality of attributes, for example uses the RDF language to mark, concrete property value in the semantic concept corresponding three-dimensional model of each mark.
Should be noted that and understand, in the situation that do not break away from the desired the spirit and scope of the present invention of accompanying claim, can make to the present invention of foregoing detailed description various modifications and improvement.Therefore, the scope of claimed technical scheme is not subjected to the restriction of given any specific exemplary teachings.
Claims (13)
1. a three-dimensional model searching system comprises semantic retrieval module and semantic database, wherein
Described semantic retrieval module also comprises: participle submodule, semantic association calculating sub module and information are transmitted submodule, wherein
The retrieve statement that described participle submodule is used for the user is inputted carries out semantic participle and filtration, and obtains semantic concept;
Described semantic association calculating sub module is used for according to described semantic concept at a plurality of semantic concepts of semantic dictionary WordNet inquiry with its synonym, and described semantic concept and the semantic concept that inquires are carried out similarity calculate, carry out the semantic concept screening according to this similarity, obtain the semantic concept relevant with described semantic concept;
Described information is transmitted submodule and is used for retrieving corresponding three-dimensional model information in conjunction with described semantic concept and relative semantic concept at semantic database, and obtains result for retrieval;
Described semantic database is used for depositing the corresponding relation of semantic concept information and semantic concept and three-dimensional model.
2. system according to claim 1 also comprises three-dimensional modeling data storehouse and display module, wherein
Described three-dimensional modeling data storehouse is used for storage about the data of three-dimensional model;
Described display module is used for supporting that the user submits retrieve statement to, and this retrieve statement is passed to described semantic retrieval module; And be used for transmitting the result for retrieval that submodule obtains according to described information, read the three-dimensional modeling data in the described three-dimensional modeling data storehouse, and show three-dimensional modeling data;
Described information is transmitted submodule result for retrieval is passed to display module.
3. system according to claim 1 and 2, wherein, described semantic association calculating sub module is carried out similarity with described semantic concept and the semantic concept that inquires and is calculated and adopt following formula:
Wherein, two semantic concepts that A, B indicate to calculate, most recent co mmon ancestor node in the WordNet semantic dictionary of lcs (A, B) expression A, B, IC (lcs (A, B)) represents the information content value that this common ancestor's node has, structure factor
Wherein, depth (A) and depth (B) represent respectively semantic concept A, the B residing degree of depth in the WordNet semantic dictionary.
4. system according to claim 1 and 2, wherein, described semantic association calculating sub module is carried out the semantic concept screening according to similarity, obtains the semantic concept relevant with described semantic concept and comprises:
According to predefined similarity threshold, to being screened by the resulting similarity result of semantic association computational algorithm, select the semantic concept with the similarity that is higher than described predefined similarity threshold by the semantic association calculating sub module.
5. system according to claim 1 and 2, wherein, described information is transmitted submodule and retrieve corresponding three-dimensional model information in conjunction with described semantic concept and relative semantic concept in semantic database, and the acquisition result for retrieval comprises:
Be present in the described semantic database if transmit the semantic concept that comes, then retrieved and returned the corresponding three-dimensional model information of this semantic concept;
If transmit next semantic concept not in semantic database, then do not return three-dimensional model information.
6. system according to claim 2 also comprises the semantic tagger module, and this semantic tagger module comprises user interface;
The three-dimensional model semantic concept mark that the user submits to is examined by the data base administrator by user interface, check whether mark meets standard and do not repeat with existing mark, if by audit, then this semantic concept mark is added in the semantic concept mark of corresponding three-dimensional model in the described semantic database; Otherwise, then will return miscue information to display module.
7. system according to claim 6, wherein, described display module is used for also supporting that the user submits the semantic concept mark to, and the semantic concept mark that the user submits to is passed to the semantic tagger module.
8. method for searching three-dimension model based on the three-dimensional model searching system of claim 1 comprises:
Step 1), retrieve statement is carried out semantic participle and filtration, obtain semantic concept;
Step 2), obtain relative semantic concept according to described semantic concept, in semantic database, retrieve corresponding three-dimensional model information according to described semantic concept and relative semantic concept, obtain result for retrieval.
9. method according to claim 8 wherein also comprises before the step 1):
Step 0), submit retrieve statement at display module.
According to claim 8 or 9 described method, wherein steps 2) after also comprise:
Step 3), result for retrieval is inquired about in the three-dimensional modeling data storehouse, obtained the data of three-dimensional model, and show this three-dimensional modeling data.
11. according to claim 8 or 9 described methods, obtaining relative semantic concept according to described semantic concept step 2) comprises:
Step 21), in the WordNet semantic dictionary, inquire a plurality of semantic concept vocabulary with its synonym according to described semantic concept;
Step 22), adopting following formula to carry out similarity described semantic concept and the semantic concept that inquires calculates:
Wherein, two semantic concepts that A, B indicate to calculate, most recent co mmon ancestor node in the WordNet semantic dictionary of lcs (A, B) expression A, B, IC (lcs (A, B)) represents the information content value that this common ancestor's node has, structure factor
Wherein, depth (A) and depth (B) represent respectively semantic concept A, the B residing degree of depth in the WordNet semantic dictionary;
Step 23), carry out the semantic concept screening according to the similarity of calculating gained, obtain the semantic concept relevant with described semantic concept.
12. method according to claim 11, step 23) carrying out the semantic concept screening according to the similarity of calculating gained in comprises:
According to predefined similarity threshold, the similarity of gained is screened, select to have the semantic concept of the similarity that is higher than this similarity threshold as relevant semantic concept.
13. according to claim 8 or 9 described methods, step 2) in semantic database, retrieve corresponding three-dimensional model information according to described semantic concept and relative semantic concept in, obtain result for retrieval and comprise:
If these semantic concepts have been present in the semantic database, then from the semantic data library searching and return the corresponding three-dimensional model information of this semantic concept; As not existing, then do not return three-dimensional model information.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210421340.2A CN102955848B (en) | 2012-10-29 | 2012-10-29 | A kind of three-dimensional model searching system based on semanteme and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210421340.2A CN102955848B (en) | 2012-10-29 | 2012-10-29 | A kind of three-dimensional model searching system based on semanteme and method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102955848A true CN102955848A (en) | 2013-03-06 |
CN102955848B CN102955848B (en) | 2015-11-18 |
Family
ID=47764654
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210421340.2A Active CN102955848B (en) | 2012-10-29 | 2012-10-29 | A kind of three-dimensional model searching system based on semanteme and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102955848B (en) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015021907A1 (en) * | 2013-08-15 | 2015-02-19 | International Business Machines Corporation | Utilization of a concept to obtain data of specific interest to a user from one or more data storage locations |
CN104461525A (en) * | 2014-11-27 | 2015-03-25 | 韩慧健 | Intelligent user-defined consulting platform generating system |
CN105335487A (en) * | 2015-10-16 | 2016-02-17 | 北京农业信息技术研究中心 | Agricultural specialist information retrieval system and method on basis of agricultural technology information ontology library |
CN107169051A (en) * | 2017-04-26 | 2017-09-15 | 山东师范大学 | Based on semantic related method for searching three-dimension model and system between body |
US9767222B2 (en) | 2013-09-27 | 2017-09-19 | International Business Machines Corporation | Information sets for data management |
CN107908760A (en) * | 2017-11-23 | 2018-04-13 | 北京航空航天大学 | A kind of method for searching three-dimension model for considering design knowledge and the aspect of model |
CN108021683A (en) * | 2017-12-11 | 2018-05-11 | 首都航天机械公司 | A kind of scale model retrieval implementation method based on three-dimensional labeling |
CN108170823A (en) * | 2018-01-04 | 2018-06-15 | 江西师范大学 | Hand-drawn interactive three-dimensional model retrieval method based on high-level semantic attribute understanding |
US10223401B2 (en) | 2013-08-15 | 2019-03-05 | International Business Machines Corporation | Incrementally retrieving data for objects to provide a desired level of detail |
CN109753655A (en) * | 2018-12-29 | 2019-05-14 | 上海半坡网络技术有限公司 | Based on the associated document retrieval method of semantic concept, system, storage medium and terminal |
CN110199354A (en) * | 2017-01-09 | 2019-09-03 | 金善中 | Biosystem information retrieval system and method |
CN110348470A (en) * | 2019-05-21 | 2019-10-18 | 湖南大学 | Semantic retrieving method for industrial fault message Rapid matching |
CN110704653A (en) * | 2019-09-09 | 2020-01-17 | 上海慧之建建设顾问有限公司 | Method for searching component by graph in BIM (building information modeling) model and graph-text searching system |
CN112487239A (en) * | 2020-11-27 | 2021-03-12 | 北京百度网讯科技有限公司 | Video retrieval method, model training method, device, equipment and storage medium |
CN112765363A (en) * | 2021-01-19 | 2021-05-07 | 昆明理工大学 | Demand map construction method for scientific and technological service demand |
CN114547357A (en) * | 2022-01-14 | 2022-05-27 | 北京师范大学 | Furniture model retrieval method based on voice and sketch |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090089277A1 (en) * | 2007-10-01 | 2009-04-02 | Cheslow Robert D | System and method for semantic search |
CN102207973A (en) * | 2011-06-22 | 2011-10-05 | 上海互联网软件有限公司 | Fuzzy search system and search method |
CN102722501A (en) * | 2011-03-31 | 2012-10-10 | 北京百度网讯科技有限公司 | Search engine and realization method thereof |
-
2012
- 2012-10-29 CN CN201210421340.2A patent/CN102955848B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090089277A1 (en) * | 2007-10-01 | 2009-04-02 | Cheslow Robert D | System and method for semantic search |
CN102722501A (en) * | 2011-03-31 | 2012-10-10 | 北京百度网讯科技有限公司 | Search engine and realization method thereof |
CN102207973A (en) * | 2011-06-22 | 2011-10-05 | 上海互联网软件有限公司 | Fuzzy search system and search method |
Non-Patent Citations (1)
Title |
---|
赵医娟等: "基于同义词扩展的文本检索改进研究", 《中国教育技术装备》, no. 18, 30 June 2010 (2010-06-30), pages 95 * |
Cited By (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10223401B2 (en) | 2013-08-15 | 2019-03-05 | International Business Machines Corporation | Incrementally retrieving data for objects to provide a desired level of detail |
WO2015021907A1 (en) * | 2013-08-15 | 2015-02-19 | International Business Machines Corporation | Utilization of a concept to obtain data of specific interest to a user from one or more data storage locations |
US10521416B2 (en) | 2013-08-15 | 2019-12-31 | International Business Machines Corporation | Incrementally retrieving data for objects to provide a desired level of detail |
US10445310B2 (en) | 2013-08-15 | 2019-10-15 | International Business Machines Corporation | Utilization of a concept to obtain data of specific interest to a user from one or more data storage locations |
US10515069B2 (en) | 2013-08-15 | 2019-12-24 | International Business Machines Corporation | Utilization of a concept to obtain data of specific interest to a user from one or more data storage locations |
US9767222B2 (en) | 2013-09-27 | 2017-09-19 | International Business Machines Corporation | Information sets for data management |
CN104461525A (en) * | 2014-11-27 | 2015-03-25 | 韩慧健 | Intelligent user-defined consulting platform generating system |
CN104461525B (en) * | 2014-11-27 | 2018-01-23 | 韩慧健 | A kind of intelligent consulting platform generation system that can customize |
CN105335487A (en) * | 2015-10-16 | 2016-02-17 | 北京农业信息技术研究中心 | Agricultural specialist information retrieval system and method on basis of agricultural technology information ontology library |
CN110199354B (en) * | 2017-01-09 | 2023-08-04 | 金善中 | Biological system information retrieval system and method |
CN110199354A (en) * | 2017-01-09 | 2019-09-03 | 金善中 | Biosystem information retrieval system and method |
CN107169051A (en) * | 2017-04-26 | 2017-09-15 | 山东师范大学 | Based on semantic related method for searching three-dimension model and system between body |
CN107169051B (en) * | 2017-04-26 | 2019-09-24 | 山东师范大学 | Based on relevant method for searching three-dimension model semantic between ontology and system |
CN107908760A (en) * | 2017-11-23 | 2018-04-13 | 北京航空航天大学 | A kind of method for searching three-dimension model for considering design knowledge and the aspect of model |
CN108021683B (en) * | 2017-12-11 | 2021-06-22 | 首都航天机械公司 | Three-dimensional labeling-based similar model retrieval implementation method |
CN108021683A (en) * | 2017-12-11 | 2018-05-11 | 首都航天机械公司 | A kind of scale model retrieval implementation method based on three-dimensional labeling |
CN108170823A (en) * | 2018-01-04 | 2018-06-15 | 江西师范大学 | Hand-drawn interactive three-dimensional model retrieval method based on high-level semantic attribute understanding |
CN109753655A (en) * | 2018-12-29 | 2019-05-14 | 上海半坡网络技术有限公司 | Based on the associated document retrieval method of semantic concept, system, storage medium and terminal |
CN110348470A (en) * | 2019-05-21 | 2019-10-18 | 湖南大学 | Semantic retrieving method for industrial fault message Rapid matching |
CN110348470B (en) * | 2019-05-21 | 2022-11-22 | 湖南大学 | Semantic retrieval method for industrial fault information rapid matching |
CN110704653A (en) * | 2019-09-09 | 2020-01-17 | 上海慧之建建设顾问有限公司 | Method for searching component by graph in BIM (building information modeling) model and graph-text searching system |
CN112487239A (en) * | 2020-11-27 | 2021-03-12 | 北京百度网讯科技有限公司 | Video retrieval method, model training method, device, equipment and storage medium |
CN112487239B (en) * | 2020-11-27 | 2024-04-05 | 北京百度网讯科技有限公司 | Video retrieval method, model training method, device, equipment and storage medium |
CN112765363A (en) * | 2021-01-19 | 2021-05-07 | 昆明理工大学 | Demand map construction method for scientific and technological service demand |
CN114547357A (en) * | 2022-01-14 | 2022-05-27 | 北京师范大学 | Furniture model retrieval method based on voice and sketch |
Also Published As
Publication number | Publication date |
---|---|
CN102955848B (en) | 2015-11-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102955848B (en) | A kind of three-dimensional model searching system based on semanteme and method | |
CN111143479B (en) | Knowledge graph relation extraction and REST service visualization fusion method based on DBSCAN clustering algorithm | |
CN110633409B (en) | Automobile news event extraction method integrating rules and deep learning | |
KR102158352B1 (en) | Providing method of key information in policy information document, Providing system of policy information, and computer program therefor | |
US7421418B2 (en) | Method and apparatus for fundamental operations on token sequences: computing similarity, extracting term values, and searching efficiently | |
CN107180045B (en) | Method for extracting geographic entity relation contained in internet text | |
CN109271505A (en) | A kind of question answering system implementation method based on problem answers pair | |
CN111177591B (en) | Knowledge graph-based Web data optimization method for visual requirements | |
CN111190900B (en) | JSON data visualization optimization method in cloud computing mode | |
CN105045875B (en) | Personalized search and device | |
CN102663129A (en) | Medical field deep question and answer method and medical retrieval system | |
US8577887B2 (en) | Content grouping systems and methods | |
CN103064956A (en) | Method, computing system and computer-readable storage media for searching electric contents | |
WO2008008213A2 (en) | Interactively crawling data records on web pages | |
CN102890702A (en) | Internet forum-oriented opinion leader mining method | |
CN112925901B (en) | Evaluation resource recommendation method for assisting online questionnaire evaluation and application thereof | |
CN114090861A (en) | Education field search engine construction method based on knowledge graph | |
CN114997288A (en) | Design resource association method | |
CN112417868A (en) | Block chain news visualization method based on emotion scores and topic models | |
CN112989811B (en) | History book reading auxiliary system based on BiLSTM-CRF and control method thereof | |
CN107908749B (en) | Character retrieval system and method based on search engine | |
JP2008009671A (en) | Data display device, data display method and data display program | |
KR102280028B1 (en) | Method for managing contents based on chatbot using big-data and artificial intelligence and apparatus for the same | |
US9305103B2 (en) | Method or system for semantic categorization | |
CN116257618A (en) | Multi-source intelligent travel recommendation method based on fine granularity emotion analysis |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB03 | Change of inventor or designer information |
Inventor after: Li Haisheng Inventor after: Cao Qian Inventor after: Cao Jian Inventor after: Cai Qiang Inventor before: Li Haisheng Inventor before: Liu Xuan Inventor before: Cao Jian Inventor before: Cai Qiang |
|
COR | Change of bibliographic data | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |