Model construction method and device, computer device and computer readable storage medium
By recognizing and matching the attributes of two-dimensional object images, the model building units are automatically selected, solving the problem of low efficiency in manually building two-dimensional models in games and achieving efficient and accurate model building.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NETEASE (HANGZHOU) NETWORK CO LTD
- Filing Date
- 2023-12-12
- Publication Date
- 2026-07-31
AI Technical Summary
When building 2D models in games, existing technologies require players to manually assemble the model components, which is difficult and inefficient, making it hard to build models with good results.
By acquiring two-dimensional object images, image recognition is performed to determine the object attribute information of each pixel, and matching target model building units are selected from the model building unit resources to automatically construct a two-dimensional model.
It improves the efficiency of building 2D models, reduces the manual operation difficulty for game players, and can accurately build 2D models with better results.
Smart Images
Figure CN117679747B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of communication technology, specifically to a model building method, apparatus, computer device, and computer-readable storage medium. Background Technology
[0002] In some games, players can create objects within the game scene. Players can assemble objects using model building blocks provided by the game. The game offers model building blocks of different materials, such as brick and wood. For example, a wall can be built using bricks, and a wall can be built using wood. In these games, players usually need to manually assemble the model building blocks to create a two-dimensional model. This process requires a significant amount of time and is therefore quite difficult and inefficient. Summary of the Invention
[0003] This application provides a model building method, apparatus, computer device, and computer-readable storage medium, which can improve the efficiency and effect of building two-dimensional models.
[0004] This application provides a model building method, including:
[0005] Get a two-dimensional object image containing the target object;
[0006] Image recognition is performed on the two-dimensional object image to obtain object attribute information corresponding to each pixel that constitutes the target object in the two-dimensional object image;
[0007] Based on the object attribute information corresponding to the pixels in the two-dimensional object image, select the resources of the target model constituent unit that match the object attribute information from the resources of the model constituent units of the two-dimensional model;
[0008] Based on the pixel position information and object attribute information in the two-dimensional object image, as well as the resources of the target model constituent units, a two-dimensional model corresponding to the target object is constructed.
[0009] Accordingly, embodiments of this application also provide a model building apparatus, comprising:
[0010] The acquisition unit is used to acquire a two-dimensional object image containing the target object;
[0011] The recognition unit is used to perform image recognition on the two-dimensional object image to obtain object attribute information corresponding to each pixel in the two-dimensional object image that constitutes the target object;
[0012] The selection unit is used to select, based on the object attribute information corresponding to the pixels in the two-dimensional object image, the resources of the model constituent units of the two-dimensional model that match the object attribute information.
[0013] The construction unit is used to construct a two-dimensional model corresponding to the target object based on the position information of pixels and object attribute information in the two-dimensional object image, as well as the resources of the target model constituting unit.
[0014] Accordingly, this application also provides a computer device, including a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to execute any of the model building methods provided in this application.
[0015] Accordingly, embodiments of this application also provide a computer-readable storage medium for storing a computer program, which is loaded by a processor to execute any of the model building methods provided in embodiments of this application.
[0016] This application embodiment obtains a two-dimensional object image containing a target object; performs image recognition on the two-dimensional object image to obtain object attribute information corresponding to each pixel constituting the target object in the two-dimensional object image; selects resources of a target model constituting unit that match the object attribute information from the resources of the model constituting units of the two-dimensional model based on the object attribute information corresponding to the pixels in the two-dimensional object image; and constructs a two-dimensional model corresponding to the target object based on the position information and object attribute information of the pixels in the two-dimensional object image and the resources of the target model constituting unit.
[0017] This application embodiment identifies images containing two-dimensional objects, determines the object attribute information corresponding to each pixel and the target model constituting unit that matches the object attribute information, and then automatically constructs a two-dimensional model based on the target model constituting unit. This can improve the efficiency of constructing two-dimensional models without requiring players to manually build them. Furthermore, it is difficult for players to manually build two-dimensional models and it is difficult to build a good two-dimensional model. However, this application embodiment can accurately construct two-dimensional models based on two-dimensional object images, thus improving the model construction effect. Attached Figure Description
[0018] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0019] Figure 1 This is a flowchart of the model building method provided in the embodiments of this application;
[0020] Figure 2 This is a schematic diagram of the structure of the neural network model provided in the embodiments of this application;
[0021] Figure 3 This is a schematic diagram of the model constituent units provided in the embodiments of this application;
[0022] Figure 4 This is a schematic diagram of the two-dimensional model provided in the embodiments of this application;
[0023] Figure 5 This is a schematic diagram of the model building apparatus provided in an embodiment of this application;
[0024] Figure 6 This is a schematic diagram of the structure of the computer device provided in the embodiments of this application. Detailed Implementation
[0025] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0026] This application provides a model building method, apparatus, computer device, and computer-readable storage medium. The model building apparatus can be integrated into a computer device, which may be a server or a terminal, etc.
[0027] The terminal may include mobile phones, wearable smart devices, tablets, laptops, personal computers (PCs), and in-vehicle computers, etc.
[0028] The server can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.
[0029] The following sections provide detailed descriptions of each example. It should be noted that the order in which the embodiments are described is not intended to limit the preferred order of the embodiments.
[0030] This embodiment will be described from the perspective of a model building device, which can be integrated into a computer device, such as a server or a terminal.
[0031] This application provides a model construction method, such as... Figure 1 As shown, the specific process of this model construction method can be as follows:
[0032] 101. Obtain a two-dimensional object image containing the target object.
[0033] The target object can be an animal, a person, a building, or an item, and the two-dimensional object image can be an image containing the target object.
[0034] For example, some games allow players to create objects within the game scene. Players can assemble objects using model building blocks provided by the game. The game offers model building blocks of different materials, such as brick and wood. A wall can be built using bricks, and a wall can be built using wood. In these games, players typically need to manually assemble these model building blocks to create a 2D model and place it in the game scene. Manually building a 2D model requires players to understand the required materials and structure of the model to assemble the building blocks. This process takes a significant amount of time, making it difficult and inefficient.
[0035] In this embodiment, the terminal device can construct a two-dimensional model corresponding to the target object based on the two-dimensional object containing the target object, eliminating the need for players to manually build it. This improves the efficiency of constructing two-dimensional models, as it is difficult for players to manually build them and produce good results. In contrast, this embodiment can construct exquisite two-dimensional models based on two-dimensional object images, thus enhancing the model construction effect.
[0036] Users can upload a two-dimensional object image containing the target object to the client, so that the client can generate a two-dimensional model of the target object based on the two-dimensional object image. The two-dimensional model is composed of model building units, which can be the basic units that make up the two-dimensional model. For example, the same model building unit can contain at least one brick, or at least one piece of wood, or at least one clump of grass, etc.
[0037] Users can upload images of two-dimensional objects to generate corresponding two-dimensional models, or they can input descriptive text about the target object. The model building method provided in this application can obtain the corresponding two-dimensional object image based on the user-input descriptive text. That is, in one embodiment, the step "obtaining a two-dimensional object image containing the target object" may specifically include:
[0038] Obtain the descriptive text of the target object;
[0039] Semantic understanding is performed on the descriptive text to obtain its semantic feature information;
[0040] Based on the semantic feature information and a preset image set, obtain a two-dimensional object image that semantically matches the descriptive text.
[0041] The descriptive text can be text related to the target object, such as the name, appearance description, or habit description of the target object. The text content of the target object can be determined based on the descriptive text. For example, the descriptive text can be "Corgi", "wooden house", "red brick house" or "vase".
[0042] The preset image set can contain multiple images collected in advance.
[0043] For example, it could involve obtaining the descriptive text of the target object input by the user, performing semantic understanding on the descriptive text to obtain semantic information, and then retrieving an image that matches the semantic information from a preset image set based on the semantic information, which would then be used as a two-dimensional object image containing the target object.
[0044] Specifically, the Clip (Contrastive Language-Image Pre-Training) model can retrieve the corresponding two-dimensional object image based on the descriptive text. The Clip model is a pre-trained model that can process text and images. During the alignment training process in large-scale image and text pairs, it understands the semantic relationship between images and text. After training, the Clip model can retrieve two-dimensional object images that match the descriptive text from the image set.
[0045] Optionally, after retrieving an image that matches the descriptive text from the image set, an image that better matches the descriptive text can be generated as a two-dimensional object image based on the image. That is, in one embodiment, the step "obtaining a two-dimensional object image that semantically matches the descriptive text based on the semantic feature information and the preset image set" may specifically include:
[0046] Based on the semantic feature information, a target image that semantically matches the descriptive text is retrieved from the preset image set;
[0047] The target image is subjected to noise addition processing to obtain a noisy target image;
[0048] Based on the description text and the target image, the noise-added target image is subjected to noise reduction processing to obtain the feature information of the target image;
[0049] The two-dimensional object image is generated based on the feature information.
[0050] The target image is an image from a preset image set, and the two-dimensional object image is an image generated based on the target image.
[0051] For example, specifically, based on the semantic feature information, a target image that semantically matches the descriptive text can be retrieved from a preset image set; in order to generate an image that is more consistent with the descriptive text, the target image can be noise-added, and then noise can be reduced based on the descriptive text and the target image to obtain the feature information of the target image. This feature information can be the representation of the target image in the embedding space, such as latent features. Based on the latent features, the image can be reconstructed to obtain a two-dimensional object image.
[0052] 102. Perform image recognition on the two-dimensional object image to obtain object attribute information corresponding to each pixel constituting the target object in the two-dimensional object image.
[0053] The object's attribute information can include information such as material, color, and structure.
[0054] For example, this could involve image recognition of a two-dimensional object image to determine the object attribute information corresponding to each pixel in the region where the target object is located. A neural network model can be used to extract features from the two-dimensional object, identifying the feature information for each pixel. Then, a classification network can be used to classify the pixels based on this feature information, thus determining the object attribute information corresponding to each pixel.
[0055] In one embodiment, the structure of the neural network model can be as follows: Figure 2 As shown, the neural network model can include convolutional layers, activation layers, pooling layers, flattening layers, and classification layers. The convolutional layers perform convolution processing on the two-dimensional object image to obtain its image feature information. The activation layers process this image feature information to enhance its representational power. The pooling layers then perform pooling processing, and the flattening layers process the image feature information into a one-dimensional vector. Finally, the classification layer predicts the object attribute information corresponding to pixels in the two-dimensional object image. Figure 2 As shown, each circle represents a pixel, and the numbers inside the circles represent object attribute information. "1" and "2" indicate different object attribute information.
[0056] In one embodiment, the object attribute information may be material information. The material information corresponding to each pixel may indicate that the target object is of the material corresponding to the material information at the pixel position, such as a turning head material or a wood material.
[0057] 103. Based on the object attribute information corresponding to the pixels in the two-dimensional object image, select the resources of the target model constituent unit that match the object attribute information from the resources of the model constituent units of the two-dimensional model.
[0058] For example, there can be preset model building units for building models. Each type of object attribute information can correspond to a model building unit. Based on the object attribute information, at least one target model building unit that matches the object attribute information can be selected from the preset model building units.
[0059] The model constituent units can include unit materials capable of displaying attribute effects indicated by object attribute information, such as bricks preset in the game. The resources for the model constituent units can include images containing the model constituent units, or color information corresponding to the pixels of a model constituent unit. For example, the resource for the model constituent unit corresponding to wood material could be image data corresponding to a unit of wood, or the color of each pixel required to display a unit of wood. The model constituent unit can be a pixel unit, that is, a unit composed of small squares. The model constituent units corresponding to different object attribute information can be, for example... Figure 3 As shown, the two-dimensional model constructed based on model building units is a pixel model, which is also composed of small squares. The two-dimensional model obtained based on pixel units can be as follows: Figure 4 As shown.
[0060] For example, the object attribute information corresponding to pixels in a two-dimensional object image may include brick material and wood material. Then, the model composition unit corresponding to the brick material and the model composition unit corresponding to the wood material can be obtained.
[0061] 104. Based on the pixel position information and object attribute information in the two-dimensional object image, and the resources of the target model constituent units, construct a two-dimensional model corresponding to the target object.
[0062] Two-dimensional models describe the outline of a target object using basic geometric shapes such as points, line segments, and polygons. Compared to three-dimensional models, two-dimensional models do not have depth information.
[0063] For example, the position of the target model constituent unit can be determined based on the position information of pixels and the object attribute information in a two-dimensional image. Then, the target model constituent unit can be mapped to the corresponding position to present a visual effect that matches the object attribute information, that is, it has attributes indicated by the object attribute information.
[0064] For example, the object attribute information is brick material information, and the corresponding target model constituent unit can be a resource that can present the effect of bricks, such as bricks preset in the game.
[0065] If displaying a target model unit requires one pixel, placing the target model unit at the position of the pixel whose object attribute information matches that target model unit will result in a two-dimensional object model.
[0066] If the number of pixels required to display a target model constituent unit is greater than one pixel, the pixels can be divided into multiple pixel sets based on the position of the pixels in the two-dimensional object image and the corresponding object attribute information. Each pixel set corresponds to replacing one target model constituent unit. That is, in one embodiment, the step "constructing a two-dimensional model corresponding to the target object based on the position information and object attribute information of the pixels in the two-dimensional object image and the resources of the target model constituent unit" can specifically include:
[0067] Based on the position information of each pixel in the two-dimensional object image and the corresponding object attribute information, the pixels of the two-dimensional object image are divided to obtain a pixel set, wherein each pixel set corresponds to a target model constitutive unit.
[0068] Based on the resources of the target model constituent units, the target model constituent units are mapped to the positions of the pixel set to obtain the two-dimensional model corresponding to the target object.
[0069] Each pixel set contains object attribute information of pixels that corresponds to a target model constituent unit.
[0070] For each target model constituent unit, the pixels in the two-dimensional object model that match the object attribute information of the target model constituent unit can be divided according to the number of pixels required to display the target model constituent unit and the distribution of pixels, to obtain at least one element set. The position corresponding to each element set can indicate the placement position of the target model constituent unit.
[0071] For example, if displaying a target model unit requires a 2×2 pixel matrix, it can be divided according to the pixels that match the target model based on the object attribute information to obtain at least one 2×2 pixel matrix. Each pixel matrix can indicate the placement position of a target model unit.
[0072] By mapping the constituent units of the target model to the positions of the corresponding set of elements in the two-dimensional object image, a two-dimensional model corresponding to the target object can be constructed.
[0073] When partitioning the element set, if the number of pixels corresponding to a target model unit in the object attribute information of the two-dimensional object image cannot be completely partitioned into the element set, and the number of pixels corresponding to a target model unit in the object attribute information of the two-dimensional object image cannot be divided evenly by the number of pixels required to display the target model unit, the excess pixels can be partitioned into the same element set with adjacent elements. This is equivalent to the two element sets sharing a portion of pixels. In the constructed two-dimensional model, the two target model units corresponding to these two element sets partially overlap.
[0074] Optionally, the pixels of the two-dimensional object image can be divided first according to the pixel matrix required by the constituent units of the target model. For pixels that cannot meet the pixel matrix, the pixel matrix is adjusted for division. The division is continuously performed on the pixels corresponding to the object attribute information and the constituent units of the target model until each pixel is assigned to a pixel set. In one embodiment, the pixel set includes a first pixel set and a second pixel set. The step "dividing the pixels of the two-dimensional object image according to the position information of each pixel in the two-dimensional object image and the corresponding object attribute information to obtain a pixel set" may include:
[0075] For each target model constituent unit, based on the size of the pixel matrix required to display the target model constituent unit, the object attribute information and the target pixels corresponding to the target model constituent unit are divided to obtain the first pixel set;
[0076] For undivided target pixels, the size of the pixel matrix is continuously adjusted, and the target pixels are divided based on the adjusted pixel matrix size until each target pixel is divided into the second pixel set.
[0077] The pixel matrix size can be the size of the pixel matrix required to display a target model's constituent units, i.e., the number of pixels in each row and the number of pixels in each column.
[0078] In this context, the target pixel is the pixel in a two-dimensional object image that corresponds to the object attribute information and the constituent unit of the target model.
[0079] Since different object attribute information corresponds to different target model constituent units, for each target model constituent unit, the target pixels can be divided according to the pixel matrix size to obtain a first pixel set with the same size as the pixel matrix required for that target model constituent unit. For target pixels that cannot be divided, the size of the pixel matrix is adjusted, for example, by reducing the number of rows or columns of the pixel matrix, to divide the undivided target pixels and obtain at least a partial second pixel set. For pixels that are not divided into the first pixel set and not divided into the second pixel set, the size of the pixel matrix is adjusted again and divided, and this process is iterated until every target pixel is divided into a pixel set.
[0080] Each adjustment results in a pixel matrix containing fewer pixels than the previous adjustment. In other words, the element set can include pixels less than or equal to those required to display the target model building units, and the number of elements in the set is equal to the number of target building units required.
[0081] For example, if the pixel matrix required to construct the display model unit is 2×2, then the element set can include a 2×2 pixel matrix, or it can include a 2×1 pixel matrix, a 1×2 pixel matrix, or a 1×1 pixel matrix. The partitioning result can be as follows: Figure 2 As shown.
[0082] For an element set whose number of pixels is less than the number of pixels required to display a model unit, the portion of the model unit that exceeds the element set can be cropped based on the adjusted pixel matrix. The number of pixels required for the cropped model unit is the same as the number of pixels contained in the element set.
[0083] Optionally, a target model constituent unit may include head and tail components and intermediate components. The intermediate components are scalable to adjust the size of the target model constituent unit. That is, in one embodiment, the resources of the target model constituent unit include scalable intermediate component resources and head and tail component resources. The step "mapping the target model constituent unit to the location of the pixel set according to the resources of the target model constituent unit to obtain the two-dimensional model corresponding to the target object" includes:
[0084] For each pixel set in the two-dimensional object image, based on the first and last component resources, the first and last components in the target model constitutive unit are mapped to the positions of the pixels at both ends of the pixel set;
[0085] The number of intermediate components to be mapped from the pixel set is determined based on the number of unmapped pixels in the pixel set and the number of pixels required to display an intermediate component.
[0086] Based on the number of intermediate components and the resources of the intermediate components, the intermediate components in the target model constituent unit are mapped to the positions of the unmapped pixels in the pixel set to obtain the two-dimensional model corresponding to the target object.
[0087] The first and last components can include a first component and a last component. The first component, the middle component, and the last component can form a target model unit. By controlling the number of middle components, the size of the target model unit can be adjusted, that is, the middle components are scalable.
[0088] In one embodiment, the pixels in the two-dimensional object image corresponding to the object attribute information and the target model constituent unit can be divided according to the pixel matrix size required to display a minimum target model constituent unit to obtain a first pixel set. Then, for pixels that are not divided into the first pixel set, if the adjacent pixels are divided into the first pixel set, the pixel is added to the first pixel set, and the number of pixels contained in the first pixel set increases. For adjacent pixels that are not divided into the first pixel set, the pixel matrix size can be adjusted to divide again, and this process is iterated until each pixel is divided into a pixel set.
[0089] For a pixel set whose number of pixels is greater than the number of pixels required to display a minimum target model unit, the head and tail components of the target model unit can be mapped to the pixel positions at both ends of the pixel set, respectively. Based on the number of pixels in the intermediate components and the number of unmapped pixels in the pixel set, the required number of intermediate components is determined, and the intermediate components are mapped into the pixel set.
[0090] In some game applications, players need to collect target model building blocks required to construct a 2D model. Only after collecting enough target model building blocks can the corresponding 2D model be constructed. During the 2D model construction process, it can be determined whether the number of target model building blocks held by the player is sufficient. Specifically, in one embodiment, the step "constructing the 2D model corresponding to the target object based on the pixel position information and object attribute information in the 2D object image, and the target model building blocks" may include:
[0091] Based on the object attribute information corresponding to each pixel and the number of pixels required to display the target model constituent units, determine the number of target model constituent units required to construct the two-dimensional model;
[0092] Obtain the number of the target model constituent units currently held by the account;
[0093] If the number of target model constituent units held by the current account is greater than or equal to the number of target model constituent units required to construct the two-dimensional model, then the two-dimensional model corresponding to the target object is constructed based on the position information of each pixel in the two-dimensional object image and the resources of the target model constituent units.
[0094] The current account can be the account currently logged into the client.
[0095] For example, for each type of target model constituent unit, the required number of each target model constituent unit is determined based on the number of pixels required to display a target model constituent unit, the object attribute information in the two-dimensional object image, and the number of pixels that match the target model constituent unit.
[0096] Get the number of each type of target model constituent unit held by the current account. If the number of each type of target model constituent unit held by the current account is greater than or equal to the number of each type of target model constituent unit required to build the two-dimensional object model, then build the two-dimensional object model; otherwise, do not build it.
[0097] As can be seen from the above, the embodiments of this application obtain a two-dimensional object image containing a target object; perform image recognition on the two-dimensional object image to obtain object attribute information corresponding to each pixel constituting the target object in the two-dimensional object image; select resources of a target model constituting unit that match the object attribute information from the resources of the model constituting units of the two-dimensional model according to the object attribute information corresponding to the pixels in the two-dimensional object image; and construct a two-dimensional model corresponding to the target object according to the position information and object attribute information of the pixels in the two-dimensional object image and the resources of the target model constituting unit.
[0098] This application embodiment identifies images containing two-dimensional objects, determines the object attribute information corresponding to each pixel and the target model constituting unit that matches the object attribute information, and then automatically constructs a two-dimensional model based on the target model constituting unit. This can improve the efficiency of constructing two-dimensional models without requiring players to manually build them. Furthermore, it is difficult for players to manually build two-dimensional models and it is difficult to build a good two-dimensional model. However, this application embodiment can accurately construct two-dimensional models based on two-dimensional object images, thus improving the model construction effect.
[0099] To facilitate better implementation of the model building method provided in the embodiments of this application, a model building apparatus is also provided in one embodiment. The meanings of the terms used are the same as in the model building method described above, and specific implementation details can be found in the description of the method embodiments.
[0100] The model building device can be integrated into a computer device, such as... Figure 5As shown, the model building device may include: an acquisition unit 301, an identification unit 302, a selection unit 303, and a construction unit 304, as detailed below:
[0101] (1) Acquisition unit 301 is used to acquire a two-dimensional object image containing the target object.
[0102] In one embodiment, the acquisition unit 301 may include:
[0103] The text acquisition subunit is used to: acquire descriptive text about the target object mentioned above;
[0104] The semantic understanding subunit is used to perform semantic understanding on the above-mentioned descriptive text to obtain the semantic feature information of the above-mentioned descriptive text;
[0105] The image acquisition subunit is used to acquire a two-dimensional object image that semantically matches the above-described text, based on the aforementioned semantic feature information and a preset image set.
[0106] In one embodiment, the image acquisition subunit may include:
[0107] The retrieval module is used to retrieve target images that semantically match the descriptive text from the preset image set based on the aforementioned semantic feature information.
[0108] The noise-adding module is used to add noise to the target image to obtain a noisy target image;
[0109] The noise reduction module is used to perform noise reduction processing on the noisy target image based on the above description text and the above target image to obtain the feature information of the target image;
[0110] The generation module is used to generate the above-mentioned two-dimensional object image based on the above-mentioned feature information.
[0111] (2) The recognition unit 302 is used to perform image recognition on the above two-dimensional object image to obtain object attribute information corresponding to each pixel constituting the target object in the above two-dimensional object image.
[0112] (3) Selection unit 303 is used to select the resources of the target model constituting unit that match the object attribute information from the resources of the model constituting unit of the two-dimensional model, based on the object attribute information corresponding to the pixels in the two-dimensional object image.
[0113] (4) Construction unit 304 is used to construct a two-dimensional model corresponding to the target object based on the position information of pixels and object attribute information in the two-dimensional object image and the resources of the target model constituting unit.
[0114] In one embodiment, the building unit 304 may include:
[0115] The quantity determination subunit is used to determine the number of target model constituent units required to construct the above two-dimensional model based on the object attribute information corresponding to each pixel and the number of pixels required to display the target model constituent units.
[0116] The quantity acquisition sub-unit is used to acquire the number of the aforementioned target model constituent units held by the current account;
[0117] The model building subunit is used to construct the two-dimensional model corresponding to the target object based on the position information of each pixel in the two-dimensional object image and the resources of the target model building units if the number of target model building units held by the current account is greater than or equal to the number of target model building units required to construct the two-dimensional model.
[0118] In one embodiment, the building unit 304 may include:
[0119] The sub-unit is used to divide the pixels of the two-dimensional object image according to the position information of each pixel in the two-dimensional object image and the corresponding object attribute information, so as to obtain a pixel set, wherein each pixel set corresponds to a target model constituting unit.
[0120] The mapping subunit is used to map the target model constituent unit to the location of the pixel set according to the resources of the target model constituent unit, so as to obtain the two-dimensional model corresponding to the target object.
[0121] In one embodiment, the pixel set includes a first pixel set and a second pixel set, and the sub-unit division may include:
[0122] The pixel division module is used to divide the object attribute information and the target pixels corresponding to the target model constituent units according to the size of the pixel matrix required to display the target model constituent units for each target model constituent unit, so as to obtain the first pixel set.
[0123] The adjustment module is used to continuously adjust the size of the pixel matrix for undivided target pixels, and divide the target pixels based on the adjusted pixel matrix size until each target pixel is divided into the second pixel set.
[0124] In one embodiment, the resources of the target model constitutive unit include scalable intermediate component resources and beginning and end component resources, and the mapping subunit may include:
[0125] The first component mapping module is used to map the first and last components in the target model constitutive unit to the positions of the pixels at both ends of the pixel set for each pixel set in the above two-dimensional object image, according to the first and last component resources.
[0126] The component quantity module is used to determine the number of intermediate components to be mapped from the pixel set based on the number of unmapped pixels in the pixel set and the number of pixels required to display an intermediate component.
[0127] The second component mapping module is used to map the intermediate components in the target model constitutive unit to the positions of other pixels in the pixel set that have not been mapped, based on the number of intermediate components and the intermediate component resources, so as to obtain the two-dimensional model corresponding to the target object.
[0128] As can be seen from the above, the model building apparatus of this application embodiment acquires a two-dimensional object image containing a target object through the acquisition unit 301; the recognition unit 302 performs image recognition on the two-dimensional object image to obtain object attribute information corresponding to each pixel constituting the target object in the two-dimensional object image; the selection unit 303 selects resources of a target model constituting unit that match the object attribute information from the resources of the model constituting units of the two-dimensional model according to the object attribute information corresponding to the pixels in the two-dimensional object image; and the construction unit 304 constructs a two-dimensional model corresponding to the target object according to the position information and object attribute information of the pixels in the two-dimensional object image and the resources of the target model constituting unit.
[0129] This application embodiment identifies images containing two-dimensional objects, determines the object attribute information corresponding to each pixel and the target model constituting unit that matches the object attribute information, and then automatically constructs a two-dimensional model based on the target model constituting unit. This can improve the efficiency of constructing two-dimensional models without requiring players to manually build them. Furthermore, it is difficult for players to manually build two-dimensional models and it is difficult to build a good two-dimensional model. However, this application embodiment can accurately construct two-dimensional models based on two-dimensional object images, thus improving the model construction effect.
[0130] Accordingly, embodiments of this application also provide a computer device, which can be a terminal. For example... Figure 6 As shown, Figure 6 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. The computer device 500 includes a processor 501 with one or more processing cores, a memory 502 with one or more computer-readable storage media, and a computer program stored on the memory 502 and executable on the processor. The processor 501 and the memory 502 are electrically connected. Those skilled in the art will understand that the computer device structure shown in the figure does not constitute a limitation on the computer device, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.
[0131] The processor 501 is the control center of the computer device 500. It connects various parts of the computer device 500 through various interfaces and lines. By running or loading software programs and / or modules stored in the memory 502, and calling data stored in the memory 502, it performs various functions of the computer device 500 and processes data, thereby monitoring the computer device 500 as a whole.
[0132] In this embodiment, the processor 501 in the computer device 500 loads the instructions corresponding to the processes of one or more applications into the memory 502 according to the following steps, and the processor 501 runs the applications stored in the memory 502 to achieve various functions:
[0133] Get a two-dimensional object image containing the target object;
[0134] Image recognition is performed on the above two-dimensional object image to obtain object attribute information corresponding to each pixel that constitutes the above target object in the above two-dimensional object image;
[0135] Based on the object attribute information corresponding to the pixels in the two-dimensional object image, select the resources of the target model constituent unit that match the object attribute information from the resources of the model constituent units of the two-dimensional model.
[0136] Based on the pixel position information and object attribute information in the above two-dimensional object image, as well as the resources of the above target model constituent units, a two-dimensional model corresponding to the above target object is constructed.
[0137] In one embodiment, the step "constructing a two-dimensional model corresponding to the target object based on the pixel position information and object attribute information in the two-dimensional object image, and the resources of the target model constituent units" may include:
[0138] Based on the object attribute information corresponding to each pixel and the number of pixels required to display the target model constituent units, determine the number of target model constituent units required to construct the two-dimensional model.
[0139] Obtain the number of the aforementioned target model constituent units currently held by the account;
[0140] If the number of target model constituent units held by the current account is greater than or equal to the number of target model constituent units required to construct the two-dimensional model, then the two-dimensional model corresponding to the target object is constructed based on the position information of each pixel in the two-dimensional object image and the resources of the target model constituent units.
[0141] In one embodiment, the step "acquiring a two-dimensional object image containing the target object" may include:
[0142] Obtain the descriptive text for the aforementioned target object;
[0143] Semantic understanding is performed on the above descriptive text to obtain its semantic feature information;
[0144] Based on the semantic feature information and the preset image set, obtain a two-dimensional object image that semantically matches the above descriptive text.
[0145] In one embodiment, the step "obtaining a two-dimensional object image that semantically matches the above-described text based on the above-mentioned semantic feature information and a preset image set" may include:
[0146] Based on the aforementioned semantic feature information, a target image that semantically matches the aforementioned descriptive text is retrieved from the aforementioned preset image set;
[0147] The target image above is subjected to noise processing to obtain a noisy target image;
[0148] Based on the above description text and the above target image, the above noise-added target image is subjected to noise reduction processing to obtain the feature information of the target image;
[0149] The above-mentioned two-dimensional object image is generated based on the aforementioned feature information.
[0150] In one embodiment, the step "constructing a two-dimensional model corresponding to the target object based on the pixel position information and object attribute information in the two-dimensional object image, and the resources of the target model constituent units" may include:
[0151] Based on the position information of each pixel in the above two-dimensional object image and the corresponding object attribute information, the pixels of the above two-dimensional object image are divided to obtain a pixel set, wherein each pixel set corresponds to a target model constitutive unit.
[0152] Based on the resources of the target model constituent units, the target model constituent units are mapped to the positions of the pixel set to obtain the two-dimensional model corresponding to the target object.
[0153] In one embodiment, the pixel set includes a first pixel set and a second pixel set. The step of "dividing the pixels of the two-dimensional object image according to the position information of each pixel in the above-mentioned two-dimensional object image and the corresponding object attribute information to obtain a pixel set" may include:
[0154] For each target model constituent unit, based on the size of the pixel matrix required to display the target model constituent unit, the object attribute information and the target pixels corresponding to the target model constituent unit are divided to obtain the first pixel set;
[0155] For undivided target pixels, the size of the pixel matrix is continuously adjusted, and the target pixels are divided based on the adjusted pixel matrix size until each target pixel is divided into the second pixel set.
[0156] In one embodiment, the resources of the target model constituent unit include scalable intermediate component resources and beginning and end component resources. The step "mapping the target model constituent unit to the location of the pixel set according to the resources of the target model constituent unit, to obtain the two-dimensional model corresponding to the target object" may include:
[0157] For each pixel set in the above two-dimensional object image, based on the first and last component resources, the first and last components in the above target model constitutive unit are mapped to the positions of the pixels at both ends of the above pixel set;
[0158] Based on the number of unmapped pixels in the aforementioned pixel set and the number of pixels required to display an intermediate component, determine the number of intermediate components that will be mapped from the aforementioned pixel set.
[0159] Based on the number of intermediate components and the resources of the intermediate components, the intermediate components in the target model constitutive unit are mapped to the positions of other pixels in the pixel set that are not mapped, thereby obtaining the two-dimensional model corresponding to the target object.
[0160] As can be seen from the above, the computer device in this application acquires a two-dimensional object image containing a target object; performs image recognition on the two-dimensional object image to obtain object attribute information corresponding to each pixel constituting the target object in the two-dimensional object image; selects resources of a target model constituting unit that match the object attribute information from the resources of the model constituting units of the two-dimensional model based on the object attribute information corresponding to the pixels in the two-dimensional object image; and constructs a two-dimensional model corresponding to the target object based on the position information and object attribute information of the pixels in the two-dimensional object image and the resources of the target model constituting unit.
[0161] This application embodiment identifies images containing two-dimensional objects, determines the object attribute information corresponding to each pixel and the target model constituting unit that matches the object attribute information, and then automatically constructs a two-dimensional model based on the target model constituting unit. This can improve the efficiency of constructing two-dimensional models without requiring players to manually build them. Furthermore, it is difficult for players to manually build two-dimensional models and it is difficult to build a good two-dimensional model. However, this application embodiment can accurately construct two-dimensional models based on two-dimensional object images, thus improving the model construction effect.
[0162] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.
[0163] Optional, such as Figure 6 As shown, the computer device 500 also includes: a touch screen display 503, a radio frequency circuit 504, an audio circuit 505, an input unit 506, and a power supply 507. The processor 501 is electrically connected to the touch screen display 503, the radio frequency circuit 504, the audio circuit 505, the input unit 506, and the power supply 507. Those skilled in the art will understand that... Figure 6 The computer device structure shown does not constitute a limitation on the computer device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.
[0164] The touch display screen 503 can be used to display a graphical user interface (GUI) and receive operation commands generated by the user interacting with the GUI. The touch display screen 503 may include a display panel and a touch panel. The display panel can be used to display information input by the user or information provided to the user, as well as various graphical user interfaces of the computer device. These graphical user interfaces can be composed of graphics, text, icons, video, and any combination thereof. Optionally, the display panel can be configured using a liquid crystal display (LCD), organic light-emitting diode (OLED), or other similar technologies. The touch panel can be used to collect touch operations performed by the user on or near it (such as operations performed by the user using a finger, stylus, or any suitable object or accessory on or near the touch panel), generate corresponding operation commands, and execute the corresponding program according to the operation commands. Optionally, the touch panel may include two parts: a touch detection device and a touch controller. The touch detection device detects the user's touch location and the signal generated by the touch operation, transmitting the signal to the touch controller. The touch controller receives touch information from the touch detection device, converts it into touch point coordinates, and sends it to the processor 501. It can also receive and execute commands from the processor 501. The touch panel can cover the display panel. When the touch panel detects a touch operation on or near it, it transmits the information to the processor 501 to determine the type of touch event. Subsequently, the processor 501 provides corresponding visual output on the display panel based on the type of touch event. In this embodiment, the touch panel and the display panel can be integrated into the touch display screen 503 to achieve input and output functions. However, in some embodiments, the touch panel and the touch display screen 503 can be implemented as two independent components to achieve input and output functions. That is, the touch display screen 503 can also be used as part of the input unit 506 to achieve input functions.
[0165] The radio frequency circuit 504 can be used to transmit and receive radio frequency signals to establish wireless communication with network devices or other computer devices, and to transmit and receive signals with network devices or other computer devices.
[0166] Audio circuitry 505 can be used to provide an audio interface between a user and a computer device via a speaker and a microphone. Audio circuitry 505 converts received audio data into electrical signals, transmits them to the speaker, and the speaker converts them into sound signals for output. Conversely, the microphone converts collected sound signals into electrical signals, which are then received by audio circuitry 505, converted back into audio data, and output to processor 501 for processing. The audio data is then transmitted via radio frequency circuitry 504 to, for example, another computer device, or output to memory 502 for further processing. Audio circuitry 505 may also include an earphone jack to facilitate communication between peripheral headphones and the computer device.
[0167] The input unit 506 can be used to receive input numbers, characters, or user characteristic information (such as fingerprints, iris, facial information, etc.), and to generate keyboard, mouse, joystick, optical, or trackball signal inputs related to user settings and function control.
[0168] Power supply 507 is used to supply power to various components of computer device 500. Optionally, power supply 507 can be logically connected to processor 501 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. Power supply 507 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.
[0169] although Figure 6 As not shown in the diagram, the computer device 500 may also include a camera, sensors, a wireless fidelity module, a Bluetooth module, etc., which will not be described in detail here.
[0170] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.
[0171] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.
[0172] Therefore, embodiments of this application provide a computer-readable storage medium storing a plurality of computer programs that can be loaded by a processor to execute the steps in any of the model building methods provided in embodiments of this application. For example, the computer program can execute the following steps:
[0173] Get a two-dimensional object image containing the target object;
[0174] Image recognition is performed on the above two-dimensional object image to obtain object attribute information corresponding to each pixel that constitutes the above target object in the above two-dimensional object image;
[0175] Based on the object attribute information corresponding to the pixels in the two-dimensional object image, select the resources of the target model constituent unit that match the object attribute information from the resources of the model constituent units of the two-dimensional model.
[0176] Based on the pixel position information and object attribute information in the above two-dimensional object image, as well as the resources of the above target model constituent units, a two-dimensional model corresponding to the above target object is constructed.
[0177] In one embodiment, the step "constructing a two-dimensional model corresponding to the target object based on the pixel position information and object attribute information in the two-dimensional object image, and the resources of the target model constituent units" may include:
[0178] Based on the object attribute information corresponding to each pixel and the number of pixels required to display the target model constituent units, determine the number of target model constituent units required to construct the two-dimensional model.
[0179] Obtain the number of the aforementioned target model constituent units currently held by the account;
[0180] If the number of target model constituent units held by the current account is greater than or equal to the number of target model constituent units required to construct the two-dimensional model, then the two-dimensional model corresponding to the target object is constructed based on the position information of each pixel in the two-dimensional object image and the target model constituent units.
[0181] In one embodiment, the step "acquiring a two-dimensional object image containing the target object" may include:
[0182] Obtain the descriptive text for the aforementioned target object;
[0183] Semantic understanding is performed on the above descriptive text to obtain its semantic feature information;
[0184] Based on the semantic feature information and the preset image set, obtain a two-dimensional object image that semantically matches the above descriptive text.
[0185] In one embodiment, the step "obtaining a two-dimensional object image that semantically matches the above-described text based on the above-mentioned semantic feature information and a preset image set" may include:
[0186] Based on the aforementioned semantic feature information, a target image that semantically matches the aforementioned descriptive text is retrieved from the aforementioned preset image set;
[0187] The target image above is subjected to noise processing to obtain a noisy target image;
[0188] Based on the above description text and the above target image, the above noise-added target image is subjected to noise reduction processing to obtain the feature information of the target image;
[0189] The above-mentioned two-dimensional object image is generated based on the aforementioned feature information.
[0190] In one embodiment, the step "constructing a two-dimensional model corresponding to the target object based on the pixel position information and object attribute information in the two-dimensional object image, and the resources of the target model constituent units" may include:
[0191] Based on the position information of each pixel in the above two-dimensional object image and the corresponding object attribute information, the pixels of the above two-dimensional object image are divided to obtain a pixel set, wherein each pixel set corresponds to a target model constitutive unit.
[0192] Based on the resources of the target model constituent units, the target model constituent units are mapped to the positions of the pixel set to obtain the two-dimensional model corresponding to the target object.
[0193] In one embodiment, the pixel set includes a first pixel set and a second pixel set. The step of "dividing the pixels of the two-dimensional object image according to the position information of each pixel in the above-mentioned two-dimensional object image and the corresponding object attribute information to obtain a pixel set" may include:
[0194] For each target model constituent unit, based on the size of the pixel matrix required to display the target model constituent unit, the object attribute information and the target pixels corresponding to the target model constituent unit are divided to obtain the first pixel set;
[0195] For undivided target pixels, the size of the pixel matrix is continuously adjusted, and the target pixels are divided based on the adjusted pixel matrix size until each target pixel is divided into the second pixel set.
[0196] In one embodiment, the resources of the target model constituent unit include scalable intermediate component resources and beginning and end component resources. The step "mapping the target model constituent unit to the location of the pixel set according to the resources of the target model constituent unit, to obtain the two-dimensional model corresponding to the target object" may include:
[0197] For each pixel set in the above two-dimensional object image, based on the first and last component resources, the first and last components in the above target model constitutive unit are mapped to the positions of the pixels at both ends of the above pixel set;
[0198] Based on the number of unmapped pixels in the aforementioned pixel set and the number of pixels required to display an intermediate component, determine the number of intermediate components that will be mapped from the aforementioned pixel set.
[0199] Based on the number of intermediate components and the resources of the intermediate components, the intermediate components in the target model constitutive unit are mapped to the positions of other pixels in the pixel set that are not mapped, thereby obtaining the two-dimensional model corresponding to the target object.
[0200] As can be seen from the above, the computer device in this application acquires a two-dimensional object image containing a target object; performs image recognition on the two-dimensional object image to obtain object attribute information corresponding to each pixel constituting the target object in the two-dimensional object image; selects resources of a target model constituting unit that match the object attribute information from the resources of the model constituting units of the two-dimensional model based on the object attribute information corresponding to the pixels in the two-dimensional object image; and constructs a two-dimensional model corresponding to the target object based on the position information and object attribute information of the pixels in the two-dimensional object image and the resources of the target model constituting unit.
[0201] This application embodiment identifies images containing two-dimensional objects, determines the object attribute information corresponding to each pixel and the target model constituting unit that matches the object attribute information, and then automatically constructs a two-dimensional model based on the target model constituting unit. This can improve the efficiency of constructing two-dimensional models without requiring players to manually build them. Furthermore, it is difficult for players to manually build two-dimensional models and it is difficult to build a good two-dimensional model. However, this application embodiment can accurately construct two-dimensional models based on two-dimensional object images, thus improving the model construction effect.
[0202] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.
[0203] The storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.
[0204] The above provides a detailed description of a model building method, apparatus, computer device, and computer storage medium provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, those skilled in the art will recognize that there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A model construction method, characterized in that, include: Get a two-dimensional object image containing the target object; Image recognition is performed on the two-dimensional object image to obtain object attribute information corresponding to each pixel that constitutes the target object in the two-dimensional object image; Based on the object attribute information corresponding to the pixels in the two-dimensional object image, select the resources of the target model constituent unit that match the object attribute information from the resources of the model constituent units of the two-dimensional model; Based on the pixel position information and object attribute information in the two-dimensional object image, and the resources of the target model constituent units, a two-dimensional model corresponding to the target object is constructed. The step of constructing a two-dimensional model corresponding to the target object based on the pixel position information and object attribute information in the two-dimensional object image, as well as the resources of the target model constituent units, includes: Based on the position information of each pixel in the two-dimensional object image and the corresponding object attribute information, the pixels of the two-dimensional object image are divided to obtain a pixel set, wherein each pixel set corresponds to a target model constitutive unit. Based on the resources of the target model constituent units, the target model constituent units are mapped to the positions of the pixel set to obtain the two-dimensional model corresponding to the target object; When the pixel set includes a first pixel set and a second pixel set, the step of dividing the pixels of the two-dimensional object image into a pixel set based on the position information of each pixel in the two-dimensional object image and the corresponding object attribute information includes: For each target model constituent unit, based on the size of the pixel matrix required to display the target model constituent unit, the object attribute information and the target pixels corresponding to the target model constituent unit are divided to obtain the first pixel set; For undivided target pixels, the size of the pixel matrix is continuously adjusted, and the target pixels are divided based on the adjusted pixel matrix size until each target pixel is divided into the second pixel set.
2. The method according to claim 1, characterized in that, The resources of the target model constituent units include scalable intermediate component resources and beginning and end component resources. The step of mapping the target model constituent units to the locations of the pixel set based on the resources of the target model constituent units to obtain the two-dimensional model corresponding to the target object includes: For each pixel set in the two-dimensional object image, based on the first and last component resources, the first and last components in the target model constitutive unit are mapped to the positions of the pixels at both ends of the pixel set; The number of intermediate components to be mapped from the pixel set is determined based on the number of unmapped pixels in the pixel set and the number of pixels required to display an intermediate component. Based on the number of intermediate components and the resources of the intermediate components, the intermediate components in the target model constituent unit are mapped to the positions of the unmapped pixels in the pixel set to obtain the two-dimensional model corresponding to the target object.
3. The method according to claim 1, characterized in that, The step of constructing a two-dimensional model corresponding to the target object based on the pixel position information and object attribute information in the two-dimensional object image, as well as the resources of the target model constituent units, includes: Based on the object attribute information corresponding to each pixel and the number of pixels required to display the target model constituent units, determine the number of target model constituent units required to construct the two-dimensional model; Obtain the number of the target model constituent units currently held by the account; If the number of target model constituent units held by the current account is greater than or equal to the number of target model constituent units required to construct the two-dimensional model, then the two-dimensional model corresponding to the target object is constructed based on the position information of each pixel in the two-dimensional object image and the resources of the target model constituent units.
4. The method according to any one of claims 1-3, characterized in that, The step of obtaining a two-dimensional object image containing the target object includes: Obtain the descriptive text of the target object; Semantic understanding is performed on the descriptive text to obtain its semantic feature information; Based on the semantic feature information and a preset image set, obtain a two-dimensional object image that semantically matches the descriptive text.
5. The method according to claim 4, characterized in that, The step of obtaining a two-dimensional object image that semantically matches the descriptive text based on the semantic feature information and a preset image set includes: Based on the semantic feature information, a target image that semantically matches the descriptive text is retrieved from the preset image set; The target image is subjected to noise addition processing to obtain a noisy target image; Based on the description text and the target image, the noise-added target image is subjected to noise reduction processing to obtain the feature information of the target image; The two-dimensional object image is generated based on the feature information.
6. A model building apparatus, characterized in that, include: The acquisition unit is used to acquire a two-dimensional object image containing the target object; The recognition unit is used to perform image recognition on the two-dimensional object image to obtain object attribute information corresponding to each pixel in the two-dimensional object image that constitutes the target object; The selection unit is used to select, based on the object attribute information corresponding to the pixels in the two-dimensional object image, the resources of the model constituent units of the two-dimensional model that match the object attribute information. The construction unit is used to construct a two-dimensional model corresponding to the target object based on the position information of pixels and object attribute information in the two-dimensional object image, as well as the resources of the target model constituting unit; The construction unit is further configured to divide the pixels of the two-dimensional object image according to the position information of each pixel in the two-dimensional object image and the corresponding object attribute information to obtain a pixel set, wherein each pixel set corresponds to a target model constituting unit; and to map the target model constituting unit to the position of the pixel set according to the resources of the target model constituting unit to obtain the two-dimensional model corresponding to the target object. When the pixel set includes a first pixel set and a second pixel set, the step of dividing the pixels of the two-dimensional object image into a pixel set based on the position information of each pixel in the two-dimensional object image and the corresponding object attribute information includes: For each target model constituent unit, the object attribute information and the target pixels corresponding to the target model constituent unit are divided according to the pixel matrix size required to display the target model constituent unit to obtain a first pixel set; for undivided target pixels, the pixel matrix size is continuously adjusted, and the target pixels are divided based on the adjusted pixel matrix size until each target pixel is divided into a second pixel set.
7. A computer device, characterized in that, It includes a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to perform the model building method according to any one of claims 1 to 5.
8. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program, which is loaded by a processor to perform the model building method according to any one of claims 1 to 5.