A method, device, storage medium and equipment for generating a virtual image

Abstract

The present application discloses a method, device, storage medium and equipment for generating a virtual image, the method comprising: first obtaining target data; wherein the target data includes target text, target image or target video in the teaching content; then extracting the target subject words in the target data, and obtaining the feature description content corresponding to the target subject words; then, generating a virtual image corresponding to the target subject words according to the target subject words and their corresponding feature description content, and then displaying the virtual image to students to improve the students' learning experience. It can be seen that since the present application can generate a specific virtual image based on the teaching content during the teaching process and display it to students for viewing, it can improve students' cognition of specific characters or objects in the teaching content, thereby improving students' learning experience and reducing the workload of teachers.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, device, storage medium and equipment for generating a virtual image. Background Art

[0002] With the rapid development of the Internet industry, artificial intelligence has led to more and more applications of the "virtual world". From animation to live broadcasting to the education of young children, all involve the application of "virtual images".

[0003] At present, in the field of early childhood education, due to the limitation of early childhood children's knowledge reserve and insufficient cognitive reserve of external things, when they encounter specific characters or objects (such as the Forbidden City and the Great Wall) in the specific process of teaching, early childhood children cannot imagine or depict the images of these specific characters or objects, which is not conducive to their cognitive learning. In this regard, in order to express the cognition of specific characters or objects to early childhood children more vividly, teachers usually select images of specific characters and objects from existing film and television works or comic works to show them to children, so as to assist children in establishing their own images of specific characters and objects so that children can learn better. However, for teachers, the workload of searching is relatively large, and there are many characters and objects involved in the course, which is not conducive to teachers finding all the corresponding materials one by one to show them to students. Usually, they can only complete the search and display of some of the main characters or objects, and cannot search and display all of them. This situation greatly restricts teachers' teaching and students' learning, resulting in poor learning effects for students. Summary of the invention

[0004] The main purpose of the embodiments of the present application is to provide a method, device, storage medium and equipment for generating a virtual image, which can generate a specific virtual image based on the teaching content and display it to students for viewing, so as to improve students' understanding of specific characters or objects in the teaching content, thereby improving students' learning experience and reducing teachers' workload.

[0005] The present application embodiment provides a method for generating a virtual image, including:

[0006] Acquire target data; the target data includes target text, target image or target video in the teaching content;

[0007] Extracting target keywords from the target data and obtaining feature description content corresponding to the target keywords;

[0008] Generating a virtual image corresponding to the target keyword according to the target keyword and its corresponding characteristic description content;

[0009] The virtual image is displayed to students to enhance the students' learning experience.

[0010] In a possible implementation, extracting the target keyword in the target data and obtaining the feature description content corresponding to the target keyword includes:

[0011] Extracting a target keyword from the target data, and obtaining description content about the target keyword from the target data as feature description content corresponding to the target keyword;

[0012] And / or, extracting a target keyword from the target data, and matching description contents related to the target keyword from a pre-constructed database as feature description contents corresponding to the target keyword.

[0013] In a possible implementation, generating a virtual image corresponding to the target keyword according to the target keyword and its corresponding characteristic expression content includes:

[0014] The target keyword and its corresponding feature description content are input into a pre-built virtual image generation model to generate a virtual image corresponding to the target keyword.

[0015] In a possible implementation, the step of constructing the virtual image generation model includes:

[0016] Acquire training data, and extract training keywords from the training data;

[0017] Obtaining feature description content corresponding to the training subject word;

[0018] The initial virtual image generation model is trained according to the training keywords and the corresponding feature description content as well as the virtual image generation labels corresponding to the training keywords to generate the virtual image generation model.

[0019] In a possible implementation, the method further includes:

[0020] Acquire verification data, and extract verification keywords from the verification data;

[0021] Acquire the feature description content corresponding to the verification subject word; and input the verification subject word and its corresponding feature description content into the virtual image generation model to obtain the virtual image generation result of the verification subject word;

[0022] When the similarity between the virtual image generation result of the verification theme word and the virtual image marking result corresponding to the verification theme word does not meet the preset threshold, the verification theme word is used again as the training theme word to update the virtual image generation model.

[0023] In a possible implementation, generating a virtual image corresponding to the target keyword according to the target keyword and its corresponding characteristic expression content includes:

[0024] Matching the characteristic description content corresponding to the target keyword with the description content related to each keyword in the pre-built database to determine the candidate virtual image corresponding to the candidate keyword that meets the preset conditions;

[0025] The candidate virtual images are processed for commonality, so as to generate a virtual image corresponding to the target keyword according to the processing result.

[0026] In a possible implementation, presenting the virtual image to students to improve the students' learning experience includes:

[0027] During the teaching process, when the teacher explains the target keyword, the virtual image is displayed at a preset page position to show the students, so as to improve the students' learning experience;

[0028] Alternatively, during the teaching process, when the teacher has finished explaining the target data, the virtual image is displayed at a preset page position to show the students so as to improve the students' learning experience;

[0029] Alternatively, during the teaching process, when the teacher has finished explaining the target data, the virtual image and its preset scene in the target data are displayed together at a preset page position to show the students, so as to improve the students' learning experience.

[0030] The embodiment of the present application also provides a device for generating a virtual image, including:

[0031] A first acquisition unit is used to acquire target data; the target data includes target text, target image or target video in the teaching content;

[0032] A second acquisition unit is used to extract the target keyword in the target data and acquire the feature description content corresponding to the target keyword;

[0033] A generating unit, configured to generate a virtual image corresponding to the target keyword according to the target keyword and its corresponding characteristic expression content;

[0034] The display unit is used to display the virtual image to students to improve the students' learning experience.

[0035] In a possible implementation manner, the second obtaining unit is specifically configured to:

[0036] Extracting a target keyword from the target data, and obtaining description content about the target keyword from the target data as feature description content corresponding to the target keyword;

[0037] And / or, extracting a target keyword from the target data, and matching description contents related to the target keyword from a pre-constructed database as feature description contents corresponding to the target keyword.

[0038] In a possible implementation, the generating unit is specifically used for:

[0039] The target keyword and its corresponding feature description content are input into a pre-built virtual image generation model to generate a virtual image corresponding to the target keyword.

[0040] In a possible implementation manner, the device further includes:

[0041] A third acquisition unit is used to acquire training data and extract training keywords from the training data;

[0042] A fourth acquisition unit is used to acquire feature description content corresponding to the training subject word;

[0043] The training unit is used to train the initial virtual image generation model according to the training keywords and the corresponding feature description content and the virtual image generation labels corresponding to the training keywords to generate the virtual image generation model.

[0044] In a possible implementation manner, the device further includes:

[0045] A fifth acquisition unit, used to acquire verification data and extract verification keywords from the verification data;

[0046] A sixth acquisition unit is used to acquire the feature description content corresponding to the verification subject word; and input the verification subject word and its corresponding feature description content into the virtual image generation model to obtain the virtual image generation result of the verification subject word;

[0047] The updating unit is used to update the virtual image generation model by reusing the verification theme as the training theme when the similarity between the virtual image generation result of the verification theme and the virtual image marking result corresponding to the verification theme does not meet a preset threshold.

[0048] In a possible implementation, the generating unit includes:

[0049] A matching subunit is used to match the characteristic description content corresponding to the target keyword with the description content related to each keyword in the pre-built database, and determine the candidate virtual image corresponding to the candidate keyword that meets the preset conditions;

[0050] The generating subunit is used to perform a collection and commonality processing on the candidate virtual images, so as to generate a virtual image corresponding to the target keyword according to the processing result.

[0051] In a possible implementation, the display unit is specifically used for:

[0052] During the teaching process, when the teacher explains the target keyword, the virtual image is displayed at a preset page position to show the students, so as to improve the students' learning experience;

[0053] Alternatively, during the teaching process, when the teacher has finished explaining the target data, the virtual image is displayed at a preset page position to show the students so as to improve the students' learning experience;

[0054] Alternatively, during the teaching process, when the teacher has finished explaining the target data, the virtual image and its preset scene in the target data are displayed together at a preset page position to show the students, so as to improve the students' learning experience.

[0055] The embodiment of the present application also provides a device for generating a virtual image, including: a processor, a memory, and a system bus;

[0056] The processor and the memory are connected via the system bus;

[0057] The memory is used to store one or more programs, and the one or more programs include instructions, which, when executed by the processor, enable the processor to execute any one of the implementations of the above-mentioned virtual image generation method.

[0058] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, and when the instructions are executed on a terminal device, the terminal device executes any one of the implementations of the above-mentioned virtual image generation method.

[0059] The embodiment of the present application further provides a computer program product. When the computer program product is run on a terminal device, the terminal device executes any one of the implementations of the above-mentioned virtual image generation method.

[0060] The embodiment of the present application provides a method, device, storage medium and equipment for generating a virtual image, first obtaining target data; wherein the target data includes target text, target image or target video in the teaching content; then extracting the target subject words in the target data, and obtaining the feature description content corresponding to the target subject words; then, generating a virtual image corresponding to the target subject words according to the target subject words and their corresponding feature description content, and then displaying the virtual image to students to improve the students' learning experience. It can be seen that since the embodiment of the present application can generate a specific virtual image based on the teaching content during the teaching process and display it to students for viewing, it can improve students' cognition of specific characters or objects in the teaching content, thereby improving students' learning experience and reducing the workload of teachers. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0062] Figure 1 A schematic diagram of a flow chart of a method for generating a virtual image provided in an embodiment of the present application;

[0063] Figure 2 A schematic diagram of the composition of a virtual image generation device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0064] At present, in the field of early childhood education, due to the limitation of early childhood knowledge reserves and insufficient cognitive reserves of external things, when encountering specific people or objects (such as the Forbidden City and the Great Wall) in the specific teaching process, early childhood children are unable to imagine or depict the images of these specific people or objects, which is not conducive to the cognitive learning of early childhood children. For example, when teachers tell the story of "Little Red Riding Hood" to early childhood children, children are usually particularly interested in the characters in the story, such as Little Red Riding Hood and the Big Bad Wolf, but due to the limitation of children's knowledge reserves and insufficient cognitive reserves of external objects, it is difficult for them to outline these characters in their minds, which is not conducive to the cognitive learning of these stories.

[0065] In this regard, the existing solution is for teachers to select images of specific characters and objects from existing film and television works or comics and show them to children to help them build their own images of specific characters and objects so that they can learn better. However, this solution requires a lot of searching work for teachers, and there are many characters and objects involved in the course, which is not conducive to teachers finding all the corresponding materials one by one to show to students. Usually, only some of the main characters or objects can be found and displayed, and all of them cannot be found and displayed. This situation greatly restricts teachers' teaching and students' learning, resulting in poor learning effects for students.

[0066] In order to solve the above defects, the present application provides a method for generating a virtual image, firstly obtaining target data; wherein the target data includes target text, target image or target video in the teaching content; then extracting the target subject words in the target data, and obtaining the feature description content corresponding to the target subject words; then, generating a virtual image corresponding to the target subject words according to the target subject words and their corresponding feature description content, and then displaying the virtual image to students to improve the students' learning experience. It can be seen that since the embodiment of the present application can generate a specific virtual image based on the teaching content during the teaching process and display it to students for viewing, it can improve students' cognition of specific characters or objects in the teaching content, thereby improving students' learning experience and reducing the workload of teachers.

[0067] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0068] First embodiment

[0069] See also Figure 1 , is a flow chart of a method for generating a virtual image provided by this embodiment, the method comprising the following steps:

[0070] S101: Acquire target data; wherein the target data includes target text, target image or target video in the teaching content.

[0071] In this embodiment, the teaching content for which a virtual image needs to be generated can be defined as target data, and any text, image or video in the teaching content (such as text, image or video played through a projection device) can be defined as target text, target image or target video respectively. This embodiment does not limit the language type of the target text, for example, the target text can be Chinese text or English text, etc. Moreover, this embodiment does not limit the length of the target text, for example, the target text can be a sentence text or a chapter text, etc. This embodiment also does not limit the type of the target image, for example, the target image can be a color image composed of the three primary colors of red (G), green (G), and blue (B), or a grayscale image, etc.

[0072] It should be noted that this application will be introduced later using the target data as the target text in the teaching content as an example. The process of generating a virtual image for other types of teaching content data such as target data such as target images or target videos can be implemented by reference and will not be described in detail.

[0073] S102: Extract target keywords from target data, and obtain feature description content corresponding to the target keywords.

[0074] In this embodiment, after the target data (i.e., the target text) is acquired through step S101, in order to extract the entity words (defined here as target subject words) in the target text to generate its corresponding virtual image, the target text can be first segmented to obtain the various words contained in the target text, and then the entity words (such as words representing entities such as people or objects) therein are extracted as target subject words, and then the relevant content describing the entity corresponding to the target subject word is extracted from the target text as the feature description content corresponding to the target subject word, so as to execute the subsequent step S103.

[0075] Among them, when the target text is a sentence text, the target text can be segmented using existing or future word segmentation methods to obtain the various words in the target text, and then the entity words are extracted from them as the target subject words. For example, assuming that the target text is "Everyone calls her Little Red Riding Hood", after word segmentation, four words "everyone, call, her, Little Red Riding Hood" can be obtained, and then the entity word "Little Red Riding Hood" representing the character can be used as the target subject word, and after obtaining the corresponding feature expression content, the subsequent step S103 can be continued.

[0076] Alternatively, if the target text is a paragraph text, it is necessary to first segment the target text into sentences to obtain the sentence texts of the target text, then segment the sentence texts using the word segmentation method to obtain the words in the target text, and then extract the entity words from them as the target subject words.

[0077] In a possible implementation method of an embodiment of the present application, the implementation process of this step S102 may specifically include: extracting the target subject words in the target data, and obtaining the description content about the target subject words from the target data as the feature expression content corresponding to the target subject words; and / or, extracting the target subject words in the target data, and matching the description content related to the target subject words from a pre-constructed database as the feature expression content corresponding to the target subject words.

[0078] Specifically, in this implementation, while extracting the target keyword in the target text, the image description content of the entity corresponding to the target keyword can also be obtained from the target text as the feature description content corresponding to the target keyword.

[0079] For example: suppose the target text is a story about Little Red Riding Hood, and the original text of the story is "Once upon a time there was a lovely little girl. Everyone who saw her liked her, but the one who liked her the most was her grandmother (or translated as "grandmother"), who gave her whatever she wanted. Once, the grandmother gave the little girl a little red hat made of velvet, which fit her head perfectly. From then on, the girl was no longer willing to wear any other hat, so everyone called her "Little Red Riding Hood"... ". After processing the target text, the target keyword "Little Red Riding Hood" representing the main character can be extracted, and the characteristic expression content for "Little Red Riding Hood" can also be extracted from the original text of the story, which is "a lovely little girl wearing a velvet little red hat" to execute the subsequent step S103.

[0080] Alternatively, when there is no content in the target text that vividly describes the entity corresponding to the target keyword, after extracting the target keyword from the target text, the descriptive content related to the target keyword can be matched from a pre-constructed database, and the matched descriptive content can be used as the feature expression content corresponding to the target keyword.

[0081] For example: suppose the target word extracted from the target text is "Great Wall", but the image description content for "Great Wall" is not extracted in the target text. At this time, the image description content for "Great Wall" can be extracted from the pre-constructed database as the feature expression content corresponding to the target keyword "Great Wall" to execute the subsequent step S103.

[0082] The database is constructed using a large amount of Internet data and manually annotated data, and includes but is not limited to a large number of common entity image descriptions and other information. The specific construction method is consistent with the existing database construction method and will not be repeated here.

[0083] Alternatively, while extracting the target subject terms from the target text, not only can the figurative description content of the entity corresponding to the target subject term be obtained from the target text, but also the descriptive content related to the target subject term can be matched from a pre-constructed database, and these matched descriptive contents can be used together with the figurative description content of the entity corresponding to the target subject term previously obtained from the target text as the feature expression content corresponding to the target subject term.

[0084] S103: Generate a virtual image corresponding to the target keyword according to the target keyword and its corresponding characteristic expression content.

[0085] In this embodiment, after extracting the target keywords in the target data through step S102 and obtaining the characteristic description content corresponding to the target keywords, a virtual image corresponding to the target keywords can be further generated according to the target keywords and their corresponding characteristic description content to execute the subsequent step S104.

[0086] Specifically, an optional implementation method is that after extracting the target keywords from the target data and obtaining the feature description content corresponding to the target keywords, the target keywords and their corresponding feature description content can be further input into a pre-built virtual image generation model to generate a virtual image corresponding to the target keywords.

[0087] The virtual image may be a 2D plane image or a 3D anthropomorphic image, a 3D cartoon image, a 3D animal image, a 3D real image, or the like.

[0088] Next, this embodiment will introduce the construction process of the virtual image generation model, which includes the following steps A1-A3:

[0089] Step A1: Obtain training data and extract training keywords from the training data.

[0090] In this embodiment, in order to construct the virtual image generation model, a lot of preparatory work needs to be done in advance. First, a large number of texts belonging to different categories (such as history, geography, etc.) need to be collected as sample texts to form training data. For example, 100 texts describing different characters, buildings and other types of entities can be collected in advance, and each of the collected texts can be used as a sample text, and the training keywords in each sample text are extracted. Then, the virtual images corresponding to the training keywords in these sample texts are manually annotated in advance to train the virtual image generation model.

[0091] Step A2: Obtain the feature description content corresponding to the training keywords.

[0092] In this embodiment, after the training data is obtained through step A1 and the training keywords in the training data are extracted, they cannot be directly used to train and generate a virtual image generation model. Instead, it is necessary to extract the feature description content corresponding to the training keywords from the training data or a pre-built database, and then use the training keywords and their corresponding feature description content to train a virtual image generation model.

[0093] Step A3: Train the initial virtual image generation model according to the training keywords and the corresponding feature description content and the virtual image generation labels corresponding to the training keywords to generate a virtual image generation model.

[0094] In this embodiment, after obtaining the feature description content corresponding to the training keywords through step A2, the initial virtual image generation model can be further trained according to the training keywords and the corresponding feature description content and the virtual image labeling results corresponding to the training keywords, thereby generating a virtual image model.

[0095] The initial virtual image generation model can be selected as any deep learning model, such as a convolutional neural network (CNN) or a recurrent neural network (RNN). The specific model structure is not limited in this application. After the initial virtual image generation model is determined, a sample text data can be extracted from the training data in turn, and multiple rounds of model training can be performed until the training end condition is met. At this time, the virtual image generation model is generated.

[0096] Through the above embodiment, the training data can be used to train and generate a virtual image generation model, and further, the generated virtual image generation model can be verified using the verification data. The specific verification process may include the following steps B1-B3:

[0097] Step B1: Acquire verification data, and extract verification keywords from the verification data.

[0098] In this embodiment, in order to verify the virtual image generation model, it is first necessary to obtain verification data, wherein the verification data refers to text data that can be used to verify the virtual image generation model. After obtaining the verification data and extracting the verification keywords in the verification data, step B2 can be continued.

[0099] Step B2: Obtain the feature description content corresponding to the verification subject word; and input the verification subject word and its corresponding feature description content into the virtual image generation model to obtain the virtual image generation result of the verification subject word.

[0100] In this embodiment, after the verification data is obtained through step B1 and the verification keywords in the verification data are extracted, they cannot be directly used to verify the virtual image generation model. Instead, it is necessary to extract the feature description content corresponding to the verification keywords from the verification data or a pre-built database, and then input the verification keywords and their corresponding feature description content into the virtual image generation model to obtain the virtual image generation result of the verification keywords, and then continue to execute step B3.

[0101] Step B3: When the similarity between the virtual image generation result of the verification keyword and the virtual image labeling result corresponding to the verification keyword does not meet the preset threshold, the verification keyword is used again as the training keyword to update the virtual image generation model.

[0102] In this embodiment, after obtaining the virtual image generation result of the verification keyword through step B2, when the similarity between the virtual image and the manual annotation result corresponding to the verification keyword does not meet the preset threshold, the verification keyword and its corresponding verification data can be re-used as training keywords and training data to update the virtual image generation model.

[0103] The calculation of the virtual image similarity can be implemented by any image display degree calculation method, which is not limited here. The specific value of the preset threshold can also be set according to the actual situation, which is not limited in the embodiment of the present application. For example, the preset threshold can be set to 0.8.

[0104] Through the above-mentioned embodiments, the verification data can be used to effectively verify the virtual image generation model. When the similarity between the virtual image generation result of the verification keyword and the virtual image marking result corresponding to the verification keyword does not meet the preset threshold, the virtual image generation model can be adjusted and updated in time, which helps to improve the generation precision and accuracy of the model.

[0105] In another possible implementation method of the embodiment of the present application, the implementation process of the above-mentioned step S103 may specifically include: first matching the feature description content corresponding to the target keyword with the description content related to each keyword in a pre-constructed database to determine the candidate virtual image corresponding to the candidate keyword that meets the preset conditions; then performing commonality processing on the candidate virtual images to generate a virtual image corresponding to the target keyword based on the processing results.

[0106] In this implementation, after extracting the target subject words in the target data and obtaining the feature description content corresponding to the target subject words, the entity words similar to the target subject words can be searched in the pre-built database as candidate subject words according to the feature description content corresponding to the target subject words, and then the entity images corresponding to the candidate subject words whose similarity exceeds the preset similarity threshold (i.e., meets the preset conditions) are grouped as subsequent virtual images into an image set, and then all the candidate virtual images in the image set are sorted in order from high to low according to the similarity. Then, the commonalities of all the candidate virtual images in the image set are grouped, and partial images of the final virtual image are generated based on the commonalities. Specifically, all candidate virtual images in the image set can be divided into multiple dimensions, and the same part in each dimension that exceeds a specified threshold (the specific value can be preset according to actual conditions) is taken as a common part, and the common part is taken as the corresponding part of the virtual image corresponding to the target keyword. The images corresponding to other dimensions that do not belong to the common part can be supplemented by the images of the dimensions corresponding to the candidate virtual images with the highest similarity ranking in the image set, so as to form a virtual image corresponding to the target keyword, so as to execute the subsequent step S104.

[0107] For example, the entity corresponding to the target keyword is a person, and three candidate virtual images A, B, and C are obtained, and then the three candidate virtual images are divided into three dimensions of "head", "upper body", and "lower body". It is found that the similarity of the "head" of A, B, and C exceeds the specified threshold of 0.8, and the similarity of the "upper body" of A and B also exceeds the specified threshold of 0.8, but the similarity of the "lower body" of A, B, and C does not exceed the specified threshold of 0.8. Then, the "head" of A, B, or C can be used as the "head" of the virtual image corresponding to the target keyword, and the "upper body" of A or B can be used as the "upper body" of the virtual image corresponding to the target keyword, and then the "lower body" of the candidate virtual image with the highest similarity (such as A) can be selected as the "lower body" of the virtual image corresponding to the target keyword, so that a complete virtual image corresponding to the target keyword can be generated.

[0108] S104: Displaying the virtual image to students to improve their learning experience.

[0109] In this embodiment, after the virtual image corresponding to the target keyword is generated in step S103, the virtual image can be further displayed to the students to improve the students' learning experience.

[0110] Specifically, one optional implementation method is that after generating a virtual image corresponding to the target keyword, when the teacher explains the target keyword during the teaching process, the virtual image can be displayed at a preset page position to show it to students. For example, the virtual image can be popped up and displayed on the PPT page where the target text is located to attract students' attention and strengthen students' cognition of specific people or objects corresponding to the target keyword, thereby improving the learning experience.

[0111] Alternatively, after generating a virtual image corresponding to the target keyword, during the teaching process, after the teacher has finished explaining the entire target text, the virtual image can be displayed at a preset page position and shown to students. For example, after the teacher has finished explaining the entire target text, the virtual image can be displayed on the last page of the PPT page where the target text is located and shown to students for viewing, so as to strengthen students' cognition of specific people or objects corresponding to the target keyword, thereby improving the learning experience.

[0112] Alternatively, after generating a virtual image corresponding to the target keyword, during the teaching process, after the teacher has finished explaining the entire target text, the virtual image and the preset scene in the target data can be displayed together at a preset page position for presentation to students. For example, after the teacher has finished explaining the entire target text, the virtual image and the preset scene in the target data can be displayed on the last page of the PPT page where the target text is located, and displayed to students for viewing, so as to strengthen students' cognition of specific people or objects corresponding to the target keyword, so as to improve the students' learning experience.

[0113] For example: suppose that after the teacher has finished explaining the entire target text "The Emperor's New Clothes", the virtual image "King" and the preset scene of him parading in the street in new clothes can be displayed on the last page of the PPT page where the target text is located, and shown to students for viewing. Among them, the preset scene of "King" parading in the street in new clothes can be the following picture: a street full of people on both sides, and various buildings and different types of shops are distributed on both sides of the street. At the same time, the king is parading in the street in a carriage wearing new clothes.

[0114] In this way, by executing the above steps S101-S104, specific virtual images corresponding to various entity words (such as words representing people or things) appearing in the target text can be generated during the teaching process, and presented to students in two-dimensional or three-dimensional form, thereby improving the vividness of teaching.

[0115] In summary, the present embodiment provides a method for generating a virtual image, firstly obtaining target data; wherein the target data includes target text, target image or target video in the teaching content; then extracting the target subject words in the target data, and obtaining the feature description content corresponding to the target subject words; then, generating a virtual image corresponding to the target subject words according to the target subject words and their corresponding feature description content, and then displaying the virtual image to students to improve their learning experience. It can be seen that since the present embodiment of the application can generate a specific virtual image based on the teaching content during the teaching process and display it to students for viewing, it can improve students' cognition of specific characters or objects in the teaching content, thereby improving students' learning experience and reducing the workload of teachers.

[0116] Second embodiment

[0117] This embodiment will introduce a device for generating a virtual image. For related content, please refer to the above method embodiment.

[0118] See also Figure 2 , is a schematic diagram of the composition of a virtual image generation device provided in this embodiment, the device 200 includes:

[0119] The first acquisition unit 201 is used to acquire target data; the target data includes target text, target image or target video in the teaching content;

[0120] The second acquisition unit 202 is used to extract the target keyword in the target data and acquire the feature description content corresponding to the target keyword;

[0121] A generating unit 203, configured to generate a virtual image corresponding to the target keyword according to the target keyword and its corresponding characteristic expression content;

[0122] The display unit 204 is used to display the virtual image to students to improve the learning experience of the students.

[0123] In an implementation of this embodiment, the second acquiring unit 202 is specifically configured to:

[0124] Extracting a target keyword from the target data, and obtaining description content about the target keyword from the target data as feature description content corresponding to the target keyword;

[0125] And / or, extracting a target keyword from the target data, and matching description contents related to the target keyword from a pre-constructed database as feature description contents corresponding to the target keyword.

[0126] In an implementation of this embodiment, the generating unit 203 is specifically configured to:

[0127] The target keyword and its corresponding feature description content are input into a pre-built virtual image generation model to generate a virtual image corresponding to the target keyword.

[0128] In an implementation of this embodiment, the device further includes:

[0129] A third acquisition unit is used to acquire training data and extract training keywords from the training data;

[0130] A fourth acquisition unit is used to acquire feature description content corresponding to the training subject word;

[0131] The training unit is used to train the initial virtual image generation model according to the training keywords and the corresponding feature description content and the virtual image generation labels corresponding to the training keywords to generate the virtual image generation model.

[0132] In an implementation of this embodiment, the device further includes:

[0133] A fifth acquisition unit, used to acquire verification data and extract verification keywords from the verification data;

[0134] A sixth acquisition unit is used to acquire the feature description content corresponding to the verification subject word; and input the verification subject word and its corresponding feature description content into the virtual image generation model to obtain the virtual image generation result of the verification subject word;

[0135] The updating unit is used to update the virtual image generation model by reusing the verification theme as the training theme when the similarity between the virtual image generation result of the verification theme and the virtual image marking result corresponding to the verification theme does not meet a preset threshold.

[0136] In an implementation of this embodiment, the generating unit 203 includes:

[0137] A matching subunit is used to match the characteristic description content corresponding to the target keyword with the description content related to each keyword in the pre-built database, and determine the candidate virtual image corresponding to the candidate keyword that meets the preset conditions;

[0138] The generating subunit is used to perform a collection and commonality processing on the candidate virtual images, so as to generate a virtual image corresponding to the target keyword according to the processing result.

[0139] In an implementation of this embodiment, the display unit 204 is specifically used for:

[0140] During the teaching process, when the teacher explains the target keyword, the virtual image is displayed at a preset page position to show the students, so as to improve the students' learning experience;

[0141] Alternatively, during the teaching process, when the teacher has finished explaining the target data, the virtual image is displayed at a preset page position to show the students so as to improve the students' learning experience;

[0142] Alternatively, during the teaching process, when the teacher has finished explaining the target data, the virtual image and its preset scene in the target data are displayed together at a preset page position to show the students, so as to improve the students' learning experience.

[0143] Furthermore, the embodiment of the present application also provides a virtual image generation device, including: a processor, a memory, and a system bus;

[0144] The processor and the memory are connected via the system bus;

[0145] The memory is used to store one or more programs, and the one or more programs include instructions, which, when executed by the processor, enable the processor to execute any implementation method of the above-mentioned virtual image generation method.

[0146] Furthermore, an embodiment of the present application also provides a computer-readable storage medium, in which instructions are stored. When the instructions are executed on a terminal device, the terminal device executes any one of the implementation methods of the above-mentioned virtual image generation method.

[0147] Furthermore, an embodiment of the present application also provides a computer program product, which, when executed on a terminal device, enables the terminal device to execute any one of the implementation methods of the above-mentioned virtual image generation method.

[0148] It can be known from the description of the above implementation mode that those skilled in the art can clearly understand that all or part of the steps in the above-mentioned embodiment method can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a storage medium such as ROM / RAM, a disk, an optical disk, etc., including several instructions for enabling a computer device (which can be a personal computer, a server, or a network communication device such as a media gateway, etc.) to execute the methods described in the various embodiments of the present application or certain parts of the embodiments.

[0149] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0150] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article 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, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0151] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for generating a virtual image, characterized in that: include: Acquire target data; the target data includes target text, target image or target video in the teaching content; Extracting target keywords from the target data and obtaining feature description content corresponding to the target keywords; Generating a virtual image corresponding to the target keyword according to the target keyword and its corresponding characteristic description content; Displaying the virtual image to students to enhance the students' learning experience; The step of generating a virtual image corresponding to the target keyword based on the target keyword and its corresponding characteristic expression content includes: Inputting the target keyword and its corresponding feature expression content into a pre-built virtual image generation model to generate a virtual image corresponding to the target keyword; The construction method of the virtual image generation model includes: Acquire training data, and extract training keywords from the training data; Obtaining feature description content corresponding to the training subject word; Training an initial virtual image generation model according to the training keywords and the corresponding feature description content and the virtual image generation labels corresponding to the training keywords to generate the virtual image generation model; Alternatively, generating a virtual image corresponding to the target keyword based on the target keyword and its corresponding characteristic expression content includes: Matching the characteristic description content corresponding to the target keyword with the description content related to each keyword in the pre-built database to determine the candidate virtual image corresponding to the candidate keyword that meets the preset conditions; The candidate virtual images are processed for commonality, so as to generate a virtual image corresponding to the target keyword according to the processing result.

2. The method according to claim 1, characterized in that The extracting the target subject words in the target data and obtaining the feature description content corresponding to the target subject words includes: Extracting a target keyword from the target data, and obtaining description content about the target keyword from the target data as feature description content corresponding to the target keyword; And / or, extracting a target keyword from the target data, and matching description contents related to the target keyword from a pre-constructed database as feature description contents corresponding to the target keyword.

3. The method according to claim 1, characterized in that The method further comprises: Acquire verification data, and extract verification keywords from the verification data; Acquire the feature description content corresponding to the verification subject word; and input the verification subject word and its corresponding feature description content into the virtual image generation model to obtain the virtual image generation result of the verification subject word; When the similarity between the virtual image generation result of the verification theme word and the virtual image marking result corresponding to the verification theme word does not meet the preset threshold, the verification theme word is used again as the training theme word to update the virtual image generation model.

4. The method according to claim 1, characterized in that: The presenting the virtual image to the students to improve the students' learning experience includes: During the teaching process, when the teacher explains the target keyword, the virtual image is displayed at a preset page position to show the students, so as to improve the students' learning experience; Alternatively, during the teaching process, when the teacher has finished explaining the target data, the virtual image is displayed at a preset page position to show the students so as to improve the students' learning experience; Alternatively, during the teaching process, when the teacher has finished explaining the target data, the virtual image and its preset scene in the target data are displayed together at a preset page position to show the students, so as to improve the students' learning experience.

5. A device for generating a virtual image, characterized in that: include: A first acquisition unit is used to acquire target data; the target data includes target text, target image or target video in the teaching content; A second acquisition unit is used to extract the target keyword in the target data and acquire the feature description content corresponding to the target keyword; A generating unit, configured to generate a virtual image corresponding to the target keyword according to the target keyword and its corresponding characteristic expression content; A display unit, used to display the virtual image to students to improve the learning experience of the students; The generating unit is specifically used for: Inputting the target keyword and its corresponding feature description content into a pre-built virtual image generation model to generate a virtual image corresponding to the target keyword; The device also includes: A third acquisition unit is used to acquire training data and extract training keywords from the training data; A fourth acquisition unit is used to acquire feature description content corresponding to the training subject word; A training unit, configured to train an initial virtual image generation model according to the training keywords and corresponding feature description contents and virtual image generation labels corresponding to the training keywords, so as to generate the virtual image generation model; Alternatively, the generating unit comprises: A matching subunit is used to match the characteristic description content corresponding to the target keyword with the description content related to each keyword in the pre-built database, and determine the candidate virtual image corresponding to the candidate keyword that meets the preset conditions; The generating subunit is used to perform a collection and commonality processing on the candidate virtual images, so as to generate a virtual image corresponding to the target keyword according to the processing result.

6. A device for generating a virtual image, characterized in that: include: Processor, memory, system bus; The processor and the memory are connected via the system bus; The memory is used to store one or more programs, wherein the one or more programs include instructions, and when the instructions are executed by the processor, the processor executes the method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a terminal device, the terminal device executes the method according to any one of claims 1 to 4.

Images