Animation model retrieval method, device, equipment and storage medium

Abstract

The embodiment of the present application provides an animation model retrieval method, device, equipment and storage medium, the method comprising: obtaining a text to be retrieved received through an application interface; performing word segmentation on the text to be retrieved to obtain multiple word units, and performing part-of-speech tagging on each word unit to obtain a first word unit marked as a noun and a second word unit marked as a verb; performing syntactic analysis on the text to be retrieved, determining the character subject and the corresponding character behavior from the first word unit and the second word unit; retrieving animation models based on the character subject and the character behavior in a set database to obtain a first animation model set; loading the first animation model set into the application interface to display the search results corresponding to the text to be retrieved to the user. The method improves the efficiency of model retrieval, accurately screens out an animation model set that meets the user's description, shortens the time for the user to select the target animation model, and improves the fluency of the user's art production.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of computer technology, and in particular to an animation model retrieval method, apparatus, device and storage medium. Background Art

[0002] In game development, animation models, as an important part of the game's visual presentation, can make the game screen more vivid and realistic, enhance the game's immersion and interactivity, and enrich the user's visual experience. In order to facilitate the art production of the game, multiple animation models that can be reused or modified can be pre-set. Developers can use the animation models to carry out secondary development adapted to different game businesses and projects, simplify the development process, and greatly shorten the art production cycle.

[0003] However, the related technology simply classifies and stores the existing animation models, requiring developers to manually search for the required animation models based on the classification labels. Since the number of stored animation models may be large, there may be inefficiencies, affecting the smoothness of art production. Summary of the invention

[0004] The embodiments of the present application provide an animation model retrieval method, apparatus, device and storage medium, which solve the problem that developers need to manually search for required animation models according to classification labels, resulting in low efficiency and affecting the fluency of art production. The method improves the model retrieval efficiency, accurately screens out a set of animation models that meet the user's description, shortens the time for users to select target animation models, and improves the fluency of users' art production.

[0005] In a first aspect, an embodiment of the present application provides an animation model retrieval method, the method comprising:

[0006] Obtaining the text to be retrieved received through the application interface;

[0007] Performing word segmentation on the text to be searched to obtain a plurality of word units, and performing part-of-speech tagging on each of the word units to obtain a first word unit marked as a noun and a second word unit marked as a verb;

[0008] Performing syntactic analysis on the text to be retrieved, and determining a role subject and a corresponding role behavior from the first word unit and the second word unit;

[0009] Retrieving the animation model in a set database based on the character subject and the character behavior to obtain a first animation model set;

[0010] The first animation model set is loaded into the application interface to display the search results corresponding to the text to be searched to the user.

[0011] In a second aspect, the embodiment of the present application further provides an animation model retrieval device, comprising:

[0012] A text acquisition module, configured to acquire the text to be retrieved received through the application interface;

[0013] A text segmentation and tagging module is configured to perform segmentation processing on the text to be searched to obtain a plurality of word units, and to perform part-of-speech tagging on each of the word units to obtain a first word unit marked as a noun and a second word unit marked as a verb;

[0014] A role information determination module, configured to perform syntactic analysis on the text to be retrieved, and determine a role subject and a corresponding role behavior from the first word unit and the second word unit;

[0015] An animation model retrieval module is configured to retrieve animation models based on the character subject and the character behavior in a set database to obtain a first animation model set;

[0016] The model result loading module is configured to load the first animation model set into the application interface to display the search results corresponding to the text to be searched to the user.

[0017] In a third aspect, an embodiment of the present application further provides an animation model retrieval device, the device comprising:

[0018] one or more processors;

[0019] a storage device configured to store one or more programs,

[0020] When the one or more programs are executed by the one or more processors, the one or more processors implement the animation model retrieval method described in the embodiment of the present application.

[0021] In a fourth aspect, an embodiment of the present application further provides a non-volatile storage medium storing computer executable instructions, wherein the computer executable instructions, when executed by a computer processor, are configured to execute the animation model retrieval method described in the embodiment of the present application.

[0022] In the embodiment of the present application, the text to be retrieved received through the application interface is obtained; the text to be retrieved is segmented to obtain multiple word units, and each word unit is tagged with a part of speech to obtain a first word unit marked as a noun and a second word unit marked as a verb; the text to be retrieved is syntactically analyzed, and the character subject and the corresponding character behavior are determined from the first word unit and the second word unit; the animation model is retrieved based on the character subject and the character behavior in the set database to obtain a first animation model set; the first animation model set is loaded into the application interface to display the search results corresponding to the text to be retrieved to the user. In the above scheme, by providing the user with an application interface to receive the text to be retrieved, it is convenient for the user to provide more relevant description information of the target animation model; by segmenting and syntactically analyzing the text to be retrieved to obtain the character subject and the character behavior, the key description information of the text to be retrieved can be accurately captured, which is conducive to accurately retrieving the animation model set that meets the user's description, improving the model retrieval efficiency, shortening the time for the user to select the target animation model, and improving the fluency of the user's art production. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A flowchart of an animation model retrieval method provided in an embodiment of the present application;

[0024] Figure 2 A flowchart of an animation model retrieval method including a process of generating a first animation model set based on multiple character subjects provided in an embodiment of the present application;

[0025] Figure 3 A flowchart of an animation model retrieval method including a process of retrieving and determining a first target animation model provided in an embodiment of the present application;

[0026] Figure 4 A flowchart of an animation model retrieval method including a process of loading a first animation model set in batches provided in an embodiment of the present application;

[0027] Figure 5 A flowchart of another animation model retrieval method including a process of retrieving and determining a first target animation model provided in an embodiment of the present application;

[0028] Figure 6 A flowchart of another animation model retrieval method including a process of loading a first animation model set in batches provided by an embodiment of the present application;

[0029] Figure 7 A flowchart of an animation model retrieval method including a process of retrieving and generating a second animation model set provided in an embodiment of the present application;

[0030] Figure 8A structural block diagram of an animation model retrieval device provided in an embodiment of the present application;

[0031] Fig. 9 A schematic diagram of the structure of an animation model retrieval device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0032] The embodiments of the present application are further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the embodiments of the present application, rather than to limit the embodiments of the present application. It should also be noted that, for ease of description, only parts related to the embodiments of the present application are shown in the accompanying drawings, rather than all structures.

[0033] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0034] The animation model retrieval method provided in the embodiment of the present application is used to provide the user with the retrieval results of the required animation model. It receives the text to be retrieved that describes the target animation model input by the user through the application interface, and parses the text to be retrieved, and then retrieves the animation model set that may meet the user's needs based on the parsing results, and finally loads the animation model set into the application interface. Relevant application scenarios include: game development, web page development, advertising or film and television production, etc. The several application scenarios listed above are only exemplary and explanatory. In actual applications, the animation model retrieval method can also be used in art production in other scenarios, and the embodiment of the present application does not limit this. The present application aims to provide an animation model retrieval method, device, equipment and storage medium to solve the problem that developers need to manually search for the required animation model according to the classification label, resulting in low efficiency and affecting the fluency of art production.

[0035] In the animation model retrieval method provided in the embodiment of the present application, the executor of each step can be a computer device, which refers to any electronic device with data calculation, processing and storage capabilities, such as mobile phones, PCs (Personal Computers), tablet computers and other terminal devices, and can also be servers and other devices, which are not limited in the embodiment of the present application.

[0036] Figure 1 The flowchart of an animation model retrieval method provided in an embodiment of the present application can be implemented with an animation model retrieval device as the execution subject. Figure 1 As shown, the animation model retrieval method specifically includes the following steps:

[0037] Step S101: Acquire the text to be searched received through the application interface.

[0038] The application interface can be a visual operating environment for users to interact and exchange information, and is used to receive the text to be searched input by the user and to feedback the search results of the animation model corresponding to the text to be searched. The text to be searched can be the description information generated by the animation model based on the needs input by the user.

[0039] Step S102: segment the text to be searched to obtain a plurality of word units, and perform part-of-speech tagging on each word unit to obtain a first word unit marked as a noun and a second word unit marked as a verb.

[0040] Wherein, the word segmentation process can be to cut the text to be retrieved into a plurality of meaningful word units. In one embodiment, the word segmentation process of the text to be retrieved can be performed based on the dictionary and word segmentation rules set. Optionally, the forward maximum matching method can be selected to match the text to be retrieved with the words in the dictionary. If the word exists in the dictionary, it is cut out. If the word does not exist in the dictionary, one word is reduced to continue matching until the match is successful or the number of words is reset to zero. For example, "man chopping wood" - "man chopping" - "man", "man chopping wood" can be cut into "man" and "chopping wood". Optionally, it is also possible to choose to construct a directed acyclic graph of the text to be retrieved, wherein each node represents a word and the edge represents a possible word. Then find all paths from the starting node to the end node, and select the shortest path as the word segmentation result, for example, "man chopping wood", possible word segmentation paths are, "man / people / chopping / wood" and "man / chopping wood", and the latter is selected because the path is shorter. In one embodiment, a corpus can also be constructed for statistical learning, and a model can be constructed by statistically analyzing the transition probability between words and the probability of occurrence of words, such as a hidden Markov model, a maximum entropy model, a conditional random field, etc. By using the trained model, the probability of words at each position in the text is calculated, and the word sequence with the largest probability is selected as the word segmentation result. Of course, developers can also train convolutional neural networks, recurrent neural networks, long short-term memory networks, etc. to perform word segmentation processing of the text to be retrieved, which is not limited in this application. The part-of-speech tagging can be used to mark the word unit as a noun, verb, adjective, adverb, preposition, etc. The part-of-speech information. Optionally, the part-of-speech of the word unit can be determined by querying a dictionary. Optionally, a sequence tagging model can be used, such as a recurrent neural network or a Transformer, etc., and a part-of-speech tag sequence can be output by inputting the obtained word unit in a sequence form. Thus, the first word unit marked as a noun and the second word unit marked as a verb can be screened out to achieve the preliminary extraction of key information of the text to be retrieved.

[0041] Step S103: perform syntactic analysis on the text to be searched, and determine the role subject and the corresponding role behavior from the first word unit and the second word unit.

[0042] Among them, the syntactic analysis can be based on the grammatical rules of the language corresponding to the text to be retrieved, and the grammatical relationship and hierarchical structure analysis between the word units of each sentence in the text to be retrieved, such as subject, predicate and object, etc. Optionally, a statistical model can be used for syntactic analysis, such as a random context-free grammar, etc., and a neural network model can be trained to learn the mapping relationship from vocabulary to sentence structure, and syntactic analysis can be performed based on the trained neural network model. A bottom-up grammatical analysis method can also be used, such as a shift-reduce method, which is not limited in this application. For example, take the analysis of "the athlete is running while holding a baton" as an example, the word segmentation result is "athlete / while / running / while / holding / baton", and after the word segmentation processing and part-of-speech tagging in step S102, "athlete (noun) / while (adverb) / running (verb) / while (adverb) / holding (verb) / baton (noun)" can be obtained. The first word unit marked as a noun is "athlete" and "baton", and the second word unit marked as a verb is "running" and "holding". Through syntactic analysis, it can be determined that the character subject is "athlete", and the corresponding character behaviors are "running" and "holding a baton". In this way, the key description information of the animation model required by the user can be determined.

[0043] Step S104: retrieve animation models based on the character subject and the character behavior in the set database to obtain a first animation model set.

[0044] Among them, the database stores a plurality of pre-made animation models that can be used as development materials, and the character subject and the character behavior can be used as key description information to retrieve the animation model that meets the user's needs from the database. In one embodiment, each animation model in the database can be provided with a description text, and the character subject and the character behavior obtained by step S103 can be used as feature description information, and the feature description information and the description text are similarity calculated, and then the animation models whose similarity reaches the expected value are collected to obtain the first animation model set. In one embodiment, each animation model in the database can be provided with a plurality of description codes, and each description code can be used to characterize information of different aspects of the animation model, such as character information, action information, object information, scene information, etc. The description code can be set based on a multi-level classification category, such as: character-athlete-running athlete, action-sports-running, etc. The description code is generated based on the character subject and the character behavior obtained by step S103 based on a preset coding rule, and the animation model that hits the aforementioned description code is screened out from the database to obtain the first animation model set. Of course, developers can also set other forms of identification information for the animation models in the database, such as keywords, index tags, etc., and convert the character subject and character behavior into the same form of information for comparison and retrieval, which is not limited in this application.

[0045] Step S105: Load the first animation model set into the application interface to display the search results corresponding to the text to be searched to the user.

[0046] The first animation model set may include candidate animation models that have a high degree of matching with user requirements. By loading the first animation model set into the application interface, the dynamic effect of each first animation model can be intuitively displayed to the user, and the user can select and download the target animation model in the application interface according to their own needs, so as to facilitate subsequent production and development.

[0047] In the above, the text to be retrieved received through the application interface is obtained; the text to be retrieved is segmented to obtain multiple word units, and each word unit is tagged with a part of speech to obtain a first word unit marked as a noun and a second word unit marked as a verb; the text to be retrieved is syntactically analyzed to determine the character subject and the corresponding character behavior from the first word unit and the second word unit; the animation model is retrieved based on the character subject and the character behavior in the set database to obtain a first animation model set; the first animation model set is loaded into the application interface to display the search results corresponding to the text to be retrieved to the user. In the above scheme, by providing the user with an application interface to receive the text to be retrieved, it is convenient for the user to provide more relevant description information of the target animation model; by segmenting and syntactically analyzing the text to be retrieved to obtain the character subject and the character behavior, the key description information of the text to be retrieved can be accurately captured, which is conducive to accurately retrieving the animation model set that meets the user's description, improving the model retrieval efficiency, shortening the time for the user to select the target animation model, and improving the fluency of the user's art production.

[0048] Figure 2 A flowchart of an animation model retrieval method including a process of generating a first animation model set based on multiple character subjects is provided in an embodiment of the present application. The character subjects include a first character subject and a second character subject, and the character behavior includes a first character behavior corresponding to the first character subject and a second character behavior corresponding to the second character subject, such as Figure 2 As shown, the animation model retrieval method specifically includes the following steps:

[0049] Step S201: Acquire the text to be searched received through the application interface.

[0050] Step S202: segment the text to be searched to obtain a plurality of word units, and perform part-of-speech tagging on each word unit to obtain a first word unit marked as a noun and a second word unit marked as a verb.

[0051] Step S203: perform syntactic analysis on the text to be searched, and determine the role subject and the corresponding role behavior from the first word unit and the second word unit.

[0052] Step S204: retrieve the animation model in the set database based on the first character subject and the first character behavior to obtain a first candidate model set.

[0053] Step S205: retrieve the animation model based on the second character subject and the second character behavior in the set database to obtain a second candidate model set.

[0054] The animation model required by the user may correspond to multiple dynamic character objects, such as different characters, animals, etc. Therefore, multiple character subjects may be parsed from the text to be retrieved, and each character subject may have corresponding character behaviors. In this embodiment, the animation model may be retrieved for each character subject and the corresponding character behavior, thereby obtaining a first candidate model set and a second candidate model set, respectively.

[0055] Step S206: extracting a first animation model set by intersecting the first candidate model set and the second candidate model set.

[0056] By performing intersection extraction on the first candidate model set and the second candidate model set, animation models including a first character subject performing a first character behavior and a second character subject performing a second character behavior can be screened out to obtain a first animation model set.

[0057] Step S207: Load the first animation model set into the application interface to display the search results corresponding to the text to be searched to the user.

[0058] As mentioned above, by performing parallel retrieval of models of different character subjects and corresponding character behaviors respectively, it is possible to reasonably split and retrieve the texts to be retrieved corresponding to multiple character objects, reduce the complexity of a single search, and perform intersection extraction on the obtained candidate model set to accurately obtain the first animation model set that meets the needs.

[0059] Figure 3 A flowchart of an animation model retrieval method including a process of retrieving and determining a first target animation model is provided for an embodiment of the present application. Each animation model stored in the database is provided with a description text, such as Figure 3 As shown, the animation model retrieval method specifically includes the following steps:

[0060] Step S301: Acquire the text to be searched received through the application interface.

[0061] Step S302: segment the text to be searched to obtain a plurality of word units, and perform part-of-speech tagging on each word unit to obtain a first word unit marked as a noun and a second word unit marked as a verb.

[0062] Step S303: perform syntactic analysis on the text to be searched, and determine the role subject and the corresponding role behavior from the first word unit and the second word unit.

[0063] Step S304: Associating and combining the role subject and the role behavior to obtain role description information.

[0064] For example, the text to be retrieved is "athlete A is passing the baton forward, athlete B is preparing to take the baton." Through step S303, "athlete A" and the corresponding role behavior "passing the baton", and "athlete B" and the corresponding role behavior "taking the baton" can be obtained. The aforementioned role subjects and role behaviors are associated and combined to obtain the simplified role description information "athlete A passes the baton, athlete B takes the baton". The role description information can be the key feature information of the role object in the text to be retrieved.

[0065] Step S305: Calculate the similarity between the description text corresponding to each animation model in the set database and the character description information.

[0066] Among them, in one embodiment, the description text and the role description information can be converted into corresponding text vectors based on a pre-trained word embedding model, such as Word2Vec, FastText or BERT, and the similarity result can be obtained by calculating the cosine similarity, Euclidean distance, etc. between the two text vectors. In one embodiment, the similarity between the description text and the role description information can be determined based on a pre-trained deep learning model, such as a convolutional neural network, a recurrent neural network or a Transformer. Of course, the aforementioned similarity calculation is only an exemplary description, and the developer can make adaptive adjustments based on the computing resource configuration and computing accuracy requirements of the actual application scenario, which is not limited in this application. In addition, the higher the similarity, the more the animation model conforms to the user's model description.

[0067] Step S306: Filter out a first target animation model whose similarity is within a preset range from the animation models.

[0068] Among them, by screening out the first target animation model whose similarity is within the preset range, the user can be provided with search results that meet the set similarity conditions. The larger the preset range, the more first target animation models there are. Relatively speaking, the covered similarity value range will be larger, and the requirement for similarity will be lower. Developers can make adaptive adjustments to the preset range for the retrieval accuracy requirements according to the actual application scenario, and this application does not limit this.

[0069] Step S307: Combine the first target animation models to obtain a first animation model set.

[0070] Step S308: Load the first animation model set into the application interface to display the search results corresponding to the text to be searched to the user.

[0071] As mentioned above, by calculating the similarity between the character description information and the description text, the degree of match between the animation model in the database and the text to be retrieved can be effectively determined. By screening out the first target animation model whose similarity is within the preset range, the animation model with a high correlation with the user's needs can be reasonably extracted, and the model can be effectively retrieved in accordance with the user's description.

[0072] Figure 4 A flowchart of an animation model retrieval method including a process of loading a first animation model set in batches is provided in an embodiment of the present application. Figure 4 As shown, the animation model retrieval method specifically includes the following steps:

[0073] Step S401: Acquire the text to be searched received through the application interface.

[0074] Step S402: segment the text to be searched to obtain multiple word units, and perform part-of-speech tagging on each word unit to obtain a first word unit marked as a noun and a second word unit marked as a verb.

[0075] Step S403: perform syntactic analysis on the text to be retrieved, and determine the role subject and the corresponding role behavior from the first word unit and the second word unit.

[0076] Step S404: Associating and combining the role subject and the role behavior to obtain role description information.

[0077] Step S405: Calculate the similarity between the description text corresponding to each animation model in the set database and the character description information.

[0078] Step S406: Filter out a first target animation model whose similarity is within a preset range from the animation models.

[0079] Step S407: Associating and combining the first target animation model with the corresponding similarity to obtain a first animation model set.

[0080] The first target animation model may be associated and combined with the corresponding similarity, and each first target animation model may be marked with the corresponding similarity.

[0081] Step S408: Divide the first animation models in the first animation model set into batches of a first preset number from high to low according to similarity to obtain a plurality of first subsets.

[0082] Among them, since there may be multiple first animation models in the first animation model set, by sorting them from high to low according to the similarity and dividing them into batches according to the first preset number, multiple first subsets with similarity from high to low can be divided. For example, there are 12 first animation models in the first animation model set. Then if the first preset number is 4, they can be divided into 3 batches from high to low according to the similarity.

[0083] Step S409: Load the multiple first subsets into the application interface in pages from high to low according to the similarity, so as to display the search results corresponding to the text to be searched to the user.

[0084] In the application interface, a first animation model in the first subset can be displayed on each page, and the closer the page number is, the higher the similarity is. The user can view and select the search results through interactive operations.

[0085] As described above, by loading the first animation model set in batches, the slow loading problem caused by loading the entire first animation model set at one time is avoided. By displaying the paginated results from high to low similarity, the user is guaranteed to have a smooth experience of quickly viewing the search results, effectively improving the user selection efficiency and optimizing the user experience.

[0086] Figure 5 A flowchart of another animation model retrieval method including a process of retrieving and determining a first target animation model provided by an embodiment of the present application. Each animation model stored in the database is provided with a plurality of first description codes, and the first description codes can be information used to characterize different aspects of the animation model, such as character information, action information, object information, scene information, etc. Figure 5 As shown, the animation model retrieval method specifically includes the following steps:

[0087] Step S501: Acquire the text to be searched received through the application interface.

[0088] Step S502: segment the text to be searched to obtain a plurality of word units, and perform part-of-speech tagging on each word unit to obtain a first word unit marked as a noun and a second word unit marked as a verb.

[0089] Step S503: perform syntactic analysis on the text to be retrieved, and determine the role subject and the corresponding role behavior from the first word unit and the second word unit.

[0090] Step S504: Generate a plurality of second description codes associated with the character subject and the character behavior according to a preset coding rule.

[0091] Among them, the preset coding rule can be used to perform multi-level coding on the relevant information of the character subject and the character behavior to obtain the second description code. The preset coding rule can be set with different information dimensions, different classification levels and different code representations. For example, in terms of character information, character-athlete-running athlete can be set, and the corresponding description code can be A1-a01-001; in terms of action information, action-sports-running can be set, and the corresponding description code can be B1-b01-002; in terms of item information, item-car-off-road vehicle can be set, and the corresponding description code can be C1-C01-003. Therefore, in combination with the above example, taking the text to be retrieved containing "the athlete holding the baton while running" as an example, at least the corresponding second description code generated includes A1-a01-001 and B1-b01-002. Of course, the coding rules of the aforementioned description codes are only exemplary, and developers can make adaptive adjustments to the coding rules according to the needs of actual application scenarios, which are not limited in this application.

[0092] Step S505: Filter out the second target animation model whose corresponding first description code includes the second description code from the set database.

[0093] If the first description code set for the animation model includes all the second description codes generated in step S504, the animation model is likely to meet the user's needs, and the animation model can be regarded as the second target animation model.

[0094] Step S506: Combine the second target animation models to obtain a first animation model set.

[0095] Step S507: Load the first animation model set into the application interface to display the search results corresponding to the text to be searched to the user.

[0096] As mentioned above, by judging whether the first description code set for the animation model contains the second description code generated based on the character subject and the character behavior, the matching degree between the animation model in the database and the text to be retrieved can be effectively judged, and the first animation model set that meets the user's needs can be accurately screened out, thereby improving the user's model selection efficiency.

[0097] Figure 6A flowchart of another animation model retrieval method including a process of loading a first animation model set in batches provided by an embodiment of the present application. Figure 6 As shown, the animation model retrieval method specifically includes the following steps:

[0098] Step S601: Acquire the text to be searched received through the application interface.

[0099] Step S602: segment the text to be searched to obtain a plurality of word units, and perform part-of-speech tagging on each word unit to obtain a first word unit marked as a noun and a second word unit marked as a verb.

[0100] Step S603: perform syntactic analysis on the text to be retrieved, and determine the role subject and the corresponding role behavior from the first word unit and the second word unit.

[0101] Step S604: Generate multiple second description codes associated with the character subject and the character behavior according to a preset coding rule.

[0102] Step S605: Filter out the second target animation model whose corresponding first description code includes the second description code from the set database.

[0103] Step S606: Associating and combining the second target animation model with the number of corresponding first description codes to obtain a first animation model set.

[0104] The second target animation model may be associated and combined with the number of corresponding first description codes, and each second target animation model may be marked with the number of corresponding first description codes.

[0105] Step S607: Divide the first animation models in the first animation model set into batches of a second preset number from high to low according to the corresponding description code quantities to obtain a plurality of second subsets.

[0106] Among them, since the number of first animation models in the first animation model set may be multiple, by sorting from high to low according to the corresponding description code number and batching according to the second preset number, multiple second subsets with corresponding description code numbers from high to low can be divided. For example, there are 15 second animation models in the second animation model set. Then if the second preset number is 5, it can be divided into 3 batches from high to low according to the corresponding description code number. Among them, the more description codes there are, the more information the second animation model contains, and it can be displayed to the user first.

[0107] Step S608: Load the plurality of second subsets into the application interface in pages from high to low according to the number of description codes.

[0108] Among them, the application interface can display a second animation model in the second subset per page, the earlier the page number, the higher the number of description codes, and the more information of the corresponding second animation model is presented. Users can view and select search results through interactive operations.

[0109] As described above, by loading the first animation model set in batches, the slow loading problem caused by loading the entire first animation model set at one time is avoided. By displaying the paginated results from high to low according to the number of description codes, animation models with rich information can be displayed first. The greater the possibility of covering the animation elements required by the user, the more effectively the user selection efficiency is improved, and the user experience is optimized.

[0110] Figure 7 A flowchart of an animation model retrieval method including a process of retrieving and generating a second animation model set is provided in an embodiment of the present application. Figure 7 As shown, the animation model retrieval method specifically includes the following steps:

[0111] Step S701: Acquire the text to be searched received through the application interface.

[0112] Step S702: segment the text to be searched to obtain multiple word units, and perform part-of-speech tagging on each word unit to obtain a first word unit marked as a noun and a second word unit marked as a verb.

[0113] Step S703: perform syntactic analysis on the text to be searched, and determine the role subject and the corresponding role behavior from the first word unit and the second word unit.

[0114] Step S704: Generate multiple second description codes associated with the character subject and the character behavior according to a preset coding rule.

[0115] Step S705: Filter out the second target animation model whose corresponding first description code includes the second description code from the set database.

[0116] Step S706: Combine the second target animation models to obtain a first animation model set.

[0117] Step S707: Load the first animation model set into the application interface to display the search results corresponding to the text to be searched to the user.

[0118] Step S708: when the first animation model set is empty, randomly combine the plurality of second description codes according to a third preset number to obtain a plurality of reference groups, wherein the third preset number is smaller than the number of the second description codes.

[0119] The first animation model set is empty, indicating that there is no animation model in the database that completely covers the second description code. Therefore, the second description code can be randomly combined according to the third preset number to obtain multiple reference groups, and a secondary search is performed based on the reference groups. 1 、M 2 and M 3 , the third preset number is 2, and 3 reference groups can be obtained, namely M 1 +M 2 、M 2 +M 3 and M 1 +M 3 The third preset number can be obtained by subtracting 1 from the number of the second description codes to ensure that the key description information of the text to be retrieved is retained to the greatest extent.

[0120] Step S709: retrieve the animation model in the database based on the reference group to obtain a second animation model set.

[0121] Step S710: Load the second animation model set into the application interface.

[0122] As described above, when it is impossible to screen out an animation model that completely covers the second description code, multiple reference values ​​are obtained by randomly combining part of the second description codes. While retaining the key description information of the text to be retrieved to the greatest extent, the user is provided with search results with greater relevance, thereby reducing the user's adverse experience of being unable to obtain search results.

[0123] Figure 8 This is a structural block diagram of an animation model retrieval device provided in an embodiment of the present application. The device is configured to execute the animation model retrieval method provided in the above embodiment and has the corresponding functional modules and beneficial effects of the execution method. Figure 8 As shown, the device specifically includes:

[0124] The text acquisition module 101 is configured to acquire the text to be retrieved received through the application interface;

[0125] The text segmentation and tagging module 102 is configured to segment the text to be searched to obtain a plurality of word units, and tag each word unit with a part of speech to obtain a first word unit marked as a noun and a second word unit marked as a verb;

[0126] The role information determination module 103 is configured to perform syntactic analysis on the text to be retrieved, and determine the role subject and the corresponding role behavior from the first word unit and the second word unit;

[0127] Animation model retrieval module 104 is configured to retrieve animation models based on the character subject and the character behavior in a set database to obtain a first animation model set;

[0128] The model result loading module 105 is configured to load the first animation model set into the application interface to display the search results corresponding to the text to be searched to the user.

[0129] In the above, the text to be retrieved is obtained through the application interface; the text to be retrieved is segmented to obtain multiple word units, and each word unit is marked with part of speech to obtain a first word unit marked as a noun and a second word unit marked as a verb; the text to be retrieved is syntactically analyzed to determine the character subject and the corresponding character behavior from the first word unit and the second word unit; the animation model is retrieved based on the character subject and the character behavior in the set database to obtain a first animation model set; the first animation model set is loaded into the application interface to display the search results corresponding to the text to be retrieved to the user. In the above scheme, by providing the user with an application interface to receive the text to be retrieved, it is convenient for the user to provide more relevant description information of the target animation model; by segmenting and syntactically analyzing the text to be retrieved to obtain the character subject and the character behavior, the key description information of the text to be retrieved can be accurately captured, which is conducive to accurately retrieving the animation model set that meets the user's description, improving the model retrieval efficiency, shortening the time for the user to select the target animation model, and improving the fluency of the user's art production.

[0130] In a possible embodiment, the role subject includes a first role subject and a second role subject, and the role behavior includes a first role behavior corresponding to the first role subject and a second role behavior corresponding to the second role subject;

[0131] Correspondingly, the animation model retrieval module 104 is further configured as follows:

[0132] Retrieving the animation model in a set database based on the first character subject and the first character behavior to obtain a first candidate model set;

[0133] Retrieving the animation model in the set database based on the second character subject and the second character behavior to obtain a second candidate model set;

[0134] The first candidate model set and the second candidate model set are intersected to extract the first animation model set.

[0135] In a possible embodiment, each animation model stored in the database is provided with a description text;

[0136] Correspondingly, the animation model retrieval module 104 is further configured as follows:

[0137] Associating and combining the role subject and the role behavior to obtain role description information;

[0138] Calculate the similarity between the description text corresponding to each animation model in the set database and the character description information;

[0139] Filter out a first target animation model from the animation models whose similarity is within a preset range;

[0140] The first target animation models are combined to obtain a first animation model set.

[0141] In a possible embodiment, the animation model retrieval module 104 is further configured to:

[0142] Associating and combining the first target animation model with the corresponding similarity to obtain a first animation model set;

[0143] Correspondingly, the model result loading module 105 is also configured as follows:

[0144] Divide the first animation models in the first animation model set into a first preset number of batches according to similarity from high to low to obtain a plurality of first subsets;

[0145] The multiple first subsets are loaded into the application interface in pages from high to low according to the similarity.

[0146] In a possible embodiment, each animation model stored in the database is provided with a plurality of first description codes;

[0147] Correspondingly, the animation model retrieval module 104 is further configured as follows:

[0148] Generating a plurality of second description codes associated with the role subject and the role behavior according to a preset coding rule;

[0149] Filter out the second target animation model whose corresponding first description code includes the second description code from the set database;

[0150] The second target animation models are combined to obtain a first animation model set.

[0151] In a possible embodiment, the animation model retrieval module 104 is further configured to:

[0152] Associating and combining the second target animation model with the number of corresponding first description codes to obtain a first animation model set;

[0153] Correspondingly, the model result loading module 105 is also configured as follows:

[0154] Divide the first animation models in the first animation model set into batches of a second preset number from high to low according to the corresponding description code quantity to obtain a plurality of second subsets;

[0155] The plurality of second subsets are loaded into the application interface in pages from high to low according to the number of description codes.

[0156] In a possible embodiment, it further includes a secondary retrieval loading module configured as follows:

[0157] In the case that the first animation model set is empty, a plurality of second description codes are randomly combined according to a third preset number to obtain a plurality of reference groups, where the third preset number is less than the number of the second description codes;

[0158] Retrieving the animation model in the database based on the reference group to obtain a second animation model set;

[0159] The second animation model set is loaded into the application interface.

[0160] Fig. 9 A schematic diagram of the structure of an animation model retrieval device provided in an embodiment of the present application, such as Fig. 9 As shown, the device includes a processor 201, a memory 202, an input device 203 and an output device 204; the number of processors 201 in the device can be one or more. Fig. 9 A processor 201 is taken as an example; the processor 201, memory 202, input device 203 and output device 204 in the device can be connected by a bus or other means. Fig. 9 The example of connecting through a bus is taken. The memory 202, as a computer-readable storage medium, can be configured to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the animation model retrieval method in the embodiment of the present application. The processor 201 executes various functional applications and data processing of the device by running the software programs, instructions and modules stored in the memory 202, that is, implements the above-mentioned animation model retrieval method. The input device 203 can be configured to receive input digital or character information, and generate key signal input related to user settings and function control of the device. The output device 204 may include a display device such as a display screen.

[0161] The animation model retrieval device provided above can be used to execute the animation model retrieval method provided in any of the above embodiments, and has corresponding functions and beneficial effects.

[0162] An embodiment of the present application also provides a non-volatile storage medium containing computer executable instructions, which, when executed by a computer processor, is configured to execute an animation model retrieval method described in the above embodiment, which includes: obtaining a text to be retrieved received through an application interface; performing word segmentation on the text to be retrieved to obtain multiple word units, and performing part-of-speech tagging on each word unit to obtain a first word unit marked as a noun and a second word unit marked as a verb; performing syntactic analysis on the text to be retrieved, and determining a character subject and a corresponding character behavior from the first word unit and the second word unit; retrieving animation models based on the character subject and the character behavior in a set database to obtain a first animation model set; loading the first animation model set into the application interface to display the search results corresponding to the text to be retrieved to the user.

[0163] Storage medium - any of various types of memory devices or storage devices. The term "storage medium" is intended to include: installation media, such as CD-ROM, floppy disk or tape device; computer system memory or random access memory, such as DRAM, DDR RAM, SRAM, EDO RAM, Rambus RAM, etc.; non-volatile memory, such as flash memory, magnetic media, optical storage; registers or other similar types of memory elements, etc. Storage media may also include other types of memory or combinations thereof. In addition, the storage medium may be located in the first computer system in which the program is executed, or may be located in a different second computer system, which is connected to the first computer system via a network (such as the Internet). The second computer system can provide program instructions to the first computer for execution. The term "storage medium" may include two or more storage media residing in different locations (e.g., in different computer systems connected via a network). The storage medium may store program instructions (e.g., embodied as a computer program) that can be executed by one or more processors.

[0164] Of course, the storage medium containing computer executable instructions provided in the embodiment of the present application is not limited to the above animation model retrieval method, and the computer executable instructions can also execute related operations in the animation model retrieval method provided in any embodiment of the present application.

[0165] It should be noted that the numbering of each step in this scheme is only used to describe the overall design framework of this scheme, and does not represent the necessary order relationship between the steps. On the basis that the overall implementation process conforms to the overall design framework of this scheme, it belongs to the protection scope of this scheme, and the order of precedence in the form of text when describing is not an exclusive limitation on the specific implementation process of this scheme. It should be understood by those skilled in the art that the embodiments of the present application can be provided as methods, systems, or computer program products. In a typical configuration, a computing device includes one or more processors (CPU), an input / output interface, a network interface, and a memory. The memory may include non-permanent memory in a computer-readable medium, a random access memory (RAM) and / or a non-volatile memory in the form of a read-only memory (ROM) or a flash memory (fl ash RAM). The memory is an example of a computer-readable medium.

[0166] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0167] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. An animation model retrieval method, characterized in that: include: Obtaining the text to be retrieved received through the application interface; Performing word segmentation on the text to be searched to obtain a plurality of word units, and performing part-of-speech tagging on each of the word units to obtain a first word unit marked as a noun and a second word unit marked as a verb; Performing syntactic analysis on the text to be retrieved, determining a role subject and a corresponding role behavior from the first word unit and the second word unit, wherein the role subject includes a first role subject and a second role subject, and the role behavior includes a first role behavior corresponding to the first role subject and a second role behavior corresponding to the second role subject; Retrieving the animation model based on the character subject and the character behavior in a set database to obtain a first animation model set, which includes retrieving the animation model based on the first character subject and the first character behavior in the set database to obtain a first candidate model set, retrieving the animation model based on the second character subject and the second character behavior in the set database to obtain a second candidate model set, and performing intersection of the first candidate model set and the second candidate model set to extract a first animation model set; The first animation model set is loaded into the application interface to display the search results corresponding to the text to be searched to the user.

2. The animation model retrieval method according to claim 1, characterized in that: Each animation model stored in the database is provided with a description text; Correspondingly, the animation model is retrieved in the set database based on the character subject and the character behavior to obtain a first animation model set, including: Associating and combining the role subject and the role behavior to obtain role description information; Calculate the similarity between the description text corresponding to each animation model in the set database and the character description information; Filtering out a first target animation model whose similarity is within a preset range from the animation models; The first target animation models are combined to obtain a first animation model set.

3. The animation model retrieval method according to claim 2, characterized in that: The step of combining the first target animation models to obtain a first animation model set includes: Associating and combining the first target animation model with the corresponding similarity to obtain a first animation model set; Accordingly, the step of loading the first animation model set into the application interface includes: Divide the first animation models in the first animation model set into a first preset number of batches according to similarity from high to low to obtain a plurality of first subsets; The multiple first subsets are loaded into the application interface in pages from high to low according to the similarity.

4. The animation model retrieval method according to claim 1, characterized in that: Each animation model stored in the database is provided with a plurality of first description codes; Correspondingly, the animation model is retrieved in the set database based on the character subject and the character behavior to obtain a first animation model set, including: Generating a plurality of second description codes associated with the role subject and the role behavior according to a preset coding rule; Filtering out from a set database a second target animation model whose corresponding first description code includes the second description code; The second target animation models are combined to obtain a first animation model set.

5. The animation model retrieval method according to claim 4, characterized in that: The step of combining the second target animation models to obtain a first animation model set includes: Associating and combining the second target animation model with the number of corresponding first description codes to obtain a first animation model set; Accordingly, the step of loading the first animation model set into the application interface includes: Divide the first animation models in the first animation model set into batches of a second preset number from high to low according to the corresponding description code numbers to obtain a plurality of second subsets; The plurality of second subsets are loaded into the application interface in pages from high to low according to the number of description codes.

6. The animation model retrieval method according to claim 4, characterized in that: Also includes: In the case that the first animation model set is empty, the plurality of second description codes are randomly combined according to a third preset number to obtain a plurality of reference groups, wherein the third preset number is less than the number of the second description codes; Retrieving animation models in the database based on the reference group to obtain a second animation model set; The second animation model set is loaded into the application interface.

7. An animation model retrieval device, characterized in that: include: A text acquisition module, configured to acquire the text to be retrieved received through the application interface; A text segmentation and tagging module is configured to perform segmentation processing on the text to be searched to obtain a plurality of word units, and to perform part-of-speech tagging on each of the word units to obtain a first word unit marked as a noun and a second word unit marked as a verb; a role information determination module configured to perform syntactic analysis on the text to be retrieved, and determine a role subject and a corresponding role behavior from the first word unit and the second word unit, wherein the role subject includes a first role subject and a second role subject, and the role behavior includes a first role behavior corresponding to the first role subject, and a second role behavior corresponding to the second role subject; An animation model retrieval module is configured to retrieve animation models based on the character subject and the character behavior in a set database to obtain a first animation model set, including retrieving animation models based on the first character subject and the first character behavior in the set database to obtain a first candidate model set, retrieving animation models based on the second character subject and the second character behavior in the set database to obtain a second candidate model set, and performing intersection of the first candidate model set and the second candidate model set to extract a first animation model set; The model result loading module is configured to load the first animation model set into the application interface to display the search results corresponding to the text to be searched to the user.

8. An animation model retrieval device, comprising: one or more processors; The storage device is configured to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the animation model retrieval method described in any one of claims 1 to 6.

9. A non-volatile storage medium storing computer executable instructions, wherein the computer executable instructions are configured to execute the animation model retrieval method according to any one of claims 1 to 6 when executed by a computer processor.

Images