CN118037981A

CN118037981A - Method for generating three-dimensional world topography based on artificial intelligence

Info

Publication number: CN118037981A
Application number: CN202410438109.7A
Authority: CN
Inventors: 盛江; 姜长嵩; 姜良仁
Original assignee: Lungfish Culture Technology Dalian Co ltd
Current assignee: Lungfish Culture Technology Dalian Co ltd
Priority date: 2024-04-12
Filing date: 2024-04-12
Publication date: 2024-05-14
Anticipated expiration: 2044-04-12
Also published as: CN118037981B

Abstract

The embodiment of the invention provides a method for generating three-dimensional world topography based on artificial intelligence, and relates to the technical field of artificial intelligence. The method comprises the following steps: generating a terrain height map and converting the terrain height map to obtain a terrain height grid; generating candidate maps of different surface types, and determining a target map in the candidate maps according to the height value of each region in the terrain height grid; rendering the target map to a corresponding area to generate a target terrain grid; generating different types of biological grids, and distributing target biological grids for each region in the terrain elevation grid; generating a basic path, and attaching the basic path to the terrain height grid to obtain a road grid; generating a building community, determining the distribution position of the building community, and generating a building community grid according to the distribution position; and combining the target terrain grid, the target biological grid, the road grid and the building community grid to generate a three-dimensional scene. Compared with the prior art, the three-dimensional scene can be generated by applying the scheme provided by the embodiment of the invention.

Description

Method for generating three-dimensional world topography based on artificial intelligence

Technical Field

The invention relates to the technical field of artificial intelligence, in particular to a method for generating three-dimensional world topography based on artificial intelligence.

Background

With the improvement of the living standard of people, games or other virtual platforms gradually become an important component of life entertainment. In order to improve the scene effect of the virtual scene, the producer usually makes the virtual scene as a 3D (three-dimensional) scene, that is, converts the presentation effect of the virtual scene from a two-dimensional plane effect to a three-dimensional stereoscopic effect, thereby improving the reality of the three-dimensional scene as much as possible so as to improve the immersive experience effect of the user in the three-dimensional scene.

Based on this, how to generate a three-dimensional scene becomes a technical problem to be solved currently.

Disclosure of Invention

An object of an embodiment of the invention is to provide a method for generating three-dimensional world topography based on artificial intelligence so as to generate a three-dimensional scene. The specific technical scheme is as follows:

in a first aspect, an embodiment of the present invention provides a method for generating a three-dimensional world based on artificial intelligence, where the method includes:

generating a terrain height map of a scene to be generated, and converting the terrain height map into a three-dimensional terrain height grid;

Generating candidate maps with different surface types, and determining a target map matched with the surface type of each region in the terrain height grid according to the height value of the region in the candidate maps;

rendering the target map to a corresponding area, and generating a target terrain grid of the scene to be generated;

Generating different types of biological grids, and distributing target biological grids for each region in the terrain elevation grid;

Generating a basic path of the scene to be generated, and adjusting the basic path to be attached to the terrain height grid to obtain a road grid of the scene to be generated;

generating a building community, determining the distribution position of the building community based on the road grids, and generating the building community grids of the scene to be generated according to the determined distribution position;

And combining the target terrain grid, the target biological grid corresponding to each region, the road grid and the building community grid to generate the three-dimensional scene of the scene to be generated.

In an embodiment of the present invention, the generating a terrain elevation map of a scene to be generated includes:

and randomly generating a plurality of random numbers, wherein the random numbers are respectively used as terrain height values corresponding to all points in the two-dimensional image, and the two-dimensional image added with the terrain height values is used as the terrain height map of the scene to be generated.

In one embodiment of the present invention, the taking the two-dimensional image added with the terrain height value as the terrain height map of the scene to be generated includes:

smoothing the topographic elevation value in the two-dimensional image;

and taking the smoothed two-dimensional image as the topographic elevation map of the scene to be generated.

In an embodiment of the present invention, the generating the base path of the to-be-generated scene includes:

Determining an area contained in the scene to be generated, and dividing the area into subareas;

and determining the boundary line between adjacent subareas as the basic path of the scene to be generated.

In one embodiment of the present invention, the building community includes buildings, and each building in the building community is generated by:

selecting a plurality of building units from a preset building unit group to be combined into an initial building structure;

and carrying out structure rationalization treatment on the initial building structure to generate the building in the building community.

In one embodiment of the present invention, the allocating a target biological grid to each region in the terrain elevation grid includes:

Triangularizing the terrain height grid to obtain a triangular grid for representing the terrain surface of the scene to be generated;

Different terrain characteristic values are randomly configured for each triangular grid;

And determining the target biological grids corresponding to the triangular grids based on the corresponding relation between the preset terrain characteristic values and the types of the biological grids.

In one embodiment of the present invention, the generating candidate maps of different surface types includes:

And inputting the prompt words representing different surface types into a pre-trained map generation model to obtain candidate maps corresponding to the surface types represented by the prompt words, wherein the map generation model is a generated artificial intelligent model.

In a second aspect, an embodiment of the present invention provides an apparatus for generating three-dimensional world topography based on artificial intelligence, the apparatus comprising:

The height grid generation module is used for generating a terrain height map of a scene to be generated and converting the terrain height map into a three-dimensional terrain height grid;

The target mapping determining module is used for generating candidate mapping of different surface types, and determining target mapping matched with the surface type of each region in the terrain height grid according to the height value of the region in the candidate mapping;

The map rendering module is used for rendering the target map to the corresponding area and generating the target terrain grid of the scene to be generated;

the biological grid distribution module is used for generating different types of biological grids and distributing a target biological grid for each region in the terrain height grid;

the road grid generation module is used for generating a basic path of the scene to be generated, and adjusting the basic path to be attached to the terrain height grid to obtain a road grid of the scene to be generated;

The building community generation module is used for generating a building community, determining the distribution position of the building community based on the road grids, and generating the building community grids of the scene to be generated according to the determined distribution position;

and the three-dimensional scene rendering module is used for combining the target terrain grids, the target biological grids corresponding to the areas, the road grids and the building community grids to generate the three-dimensional scene of the scene to be generated.

In one embodiment of the present invention, the altitude grid generation module includes:

The terrain height random generation sub-module is used for randomly generating a plurality of random numbers, respectively serving as terrain height values corresponding to all points in the two-dimensional image, and taking the two-dimensional image added with the terrain height values as the terrain height map of the scene to be generated.

In one embodiment of the present invention, the terrain height random generation submodule is specifically configured to:

smoothing the topographic elevation value in the two-dimensional image;

In one embodiment of the present invention, the road grid generating module is specifically configured to:

In one embodiment of the present invention, the building community includes buildings, and each building in the building community is generated by the following modules:

The unit combination module is used for selecting a plurality of building units from a preset building unit group to be combined into an initial building structure;

and the building generation module is used for carrying out structure rationalization treatment on the initial building structure to generate buildings in the building community.

In one embodiment of the present invention, the biological grid allocation module is specifically configured to:

In one embodiment of the present invention, the target map determining module is specifically configured to:

In a third aspect, an embodiment of the present invention provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, where the processor, the communication interface, and the memory complete communication with each other through the communication bus;

A memory for storing a computer program;

a processor for implementing the method steps of any of the first aspects when executing a program stored on a memory.

In an eighth aspect, embodiments of the present invention provide a computer readable storage medium having a computer program stored therein, the computer program implementing the method steps of any of the first aspects when executed by a processor.

The embodiment of the invention has the beneficial effects that:

In the above, when the scheme provided by the embodiment of the invention is applied, when a three-dimensional scene is generated, firstly, a topographic elevation map of the scene to be generated is generated, and in order to facilitate the generation of the three-dimensional scene, the topographic elevation map is converted into a three-dimensional topographic elevation grid. Secondly, because the earth surface types corresponding to different terrains are different, in order to improve the authenticity of the generated three-dimensional scene, candidate maps with different earth surface types can be generated, and for each region in the terrain height grid, a target map matched with the earth surface type of the region is determined in the candidate maps according to the height value of the region. Then, the determined target map can be rendered to the corresponding area, so that the target terrain grid of the scene to be generated is generated. Then, different types of biological grids are generated, and a target biological grid is allocated to each region in the terrain elevation grid. In order to improve the rationality of the generated three-dimensional scene in road division, after the basic path of the scene to be generated is generated, the basic path and the terrain height grid can be adjusted to be attached to each other, so that the road grid of the scene to be generated is obtained. Then, after the building group is generated, the distribution position of the generated building group is determined based on the road grid, and the building group grid of the scene to be generated is generated according to the determined distribution position. Thus, the three-dimensional scene of the scene to be generated can be generated by combining the generated target terrain mesh, target biological mesh corresponding to each region, the road mesh, and the building community mesh.

Based on the scheme provided by the embodiment of the invention, when the three-dimensional scene is generated, the electronic equipment can automatically generate various data about the scene to be generated and directly render and generate the three-dimensional scene by utilizing the generated data. In other words, in the scheme provided by the embodiment of the invention, the generation process of the three-dimensional scene is dominated by the electronic equipment, namely, the generation process of the three-dimensional scene does not need human intervention, so that the automation of the generation process of the three-dimensional scene is realized, the manufacturing period of the three-dimensional scene is reduced, and the generation efficiency of the three-dimensional scene is further improved.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below.

FIG. 1 is a flow chart of a first method for generating three-dimensional world topography based on artificial intelligence according to an embodiment of the present invention;

FIG. 2 is a flow chart of a second method for generating three-dimensional world topography based on artificial intelligence according to an embodiment of the present invention;

FIG. 3 is a flow chart of a third method for generating three-dimensional world topography based on artificial intelligence according to an embodiment of the present invention;

fig. 4 is a schematic diagram illustrating generation of a road grid according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of a portion of a quad-tree according to an embodiment of the present invention;

FIG. 6 is a schematic flow chart of a building generation method according to an embodiment of the present invention;

FIG. 7 is a schematic flow chart of an embodiment of the present invention;

FIG. 8 is a schematic structural diagram of an apparatus for generating three-dimensional world topography based on artificial intelligence according to an embodiment of the present invention;

Fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by the person skilled in the art based on the present invention are included in the scope of protection of the present invention.

With the improvement of the living standard of people, games or other virtual platforms gradually become an important component of life entertainment. In order to improve the scene effect of the virtual scene, a producer usually produces the virtual scene as a 3D scene, that is, converts the presentation effect of the virtual scene from a two-dimensional plane effect to a three-dimensional stereoscopic effect, thereby improving the reality of the produced three-dimensional scene as much as possible and improving the immersive experience effect of the user in the three-dimensional scene. Based on this, how to generate a three-dimensional scene becomes a technical problem to be solved currently.

In order to solve the technical problems, the embodiment of the invention provides a method for generating three-dimensional world topography based on artificial intelligence.

The method can be suitable for various application scenes for generating the three-dimensional scene. For example, a three-dimensional scene in a game is generated, a three-dimensional scene in a virtual reality is generated, and the like. The method can be applied to various electronic devices such as notebook computers, tablet computers and desktop computers, and is hereinafter referred to as electronic device. Based on this, the embodiment of the present invention does not limit the application scenario and execution subject of the method.

The method for generating the three-dimensional world topography based on the artificial intelligence according to the embodiment of the invention is specifically described below with reference to the accompanying drawings.

Referring to fig. 1, a flow chart of a method for generating three-dimensional world topography based on artificial intelligence according to an embodiment of the present invention is provided, where the method includes the following steps S101 to S107.

Step S101: and generating a terrain height map of the scene to be generated, and converting the terrain height map into a three-dimensional terrain height grid.

When generating a three-dimensional scene, the electronic device may first generate a terrain elevation map of the scene to be generated, thereby determining a terrain elevation value corresponding to each point in the terrain elevation map in two-dimensional space. However, since the scene to be generated is a three-dimensional scene, after the generation of the terrain elevation map of the scene to be generated, the terrain elevation map can be directly converted into a three-dimensional terrain elevation grid.

Since the terrain elevation map is a two-dimensional image, in order to intuitively embody the terrain elevation value of each point in the terrain elevation map, different pixel values may be used for characterization for different terrain elevation values.

Based on this, in one embodiment of the present invention, a correspondence relation between a terrain height value and a pixel value may be preset, so that after determining the terrain height value corresponding to each point in the terrain height map, the pixel value corresponding to each point may be determined based on the correspondence relation between the terrain height value and the pixel value.

For example, when the pixel values are RGB values, different terrain height values are characterized by different colors. For another example, when the pixel value is a gray value, different terrain height values are represented by gray values of different levels.

In addition, in order to improve the generation efficiency of the three-dimensional scene, the size of the terrain height map of the scene to be generated is the same as the size of the terrain height grid. In this way, each point in the terrain elevation map can be in one-to-one correspondence with each point of the terrain elevation grid on the two-dimensional plane, thereby simplifying the subsequent generation process of the three-dimensional scene.

In one embodiment of the present invention, the terrain elevation grid may characterize elevation data, slope data, and latitude and longitude data for each point.

The elevation data is a height value corresponding to the point, the gradient data can be calculated based on the height value of the point and a point in a preset range nearby, and the longitude and latitude data are represented by the abscissa and the ordinate of the position of the point.

Step S102: candidate maps of different surface types are generated, and target maps matched with the surface types of the areas are determined in the candidate maps according to the height values of the areas aiming at each area in the terrain height grid.

For the terrain, as the altitude value increases, the altitude corresponding to the altitude value increases, and accordingly, the terrain area of the point corresponding to the altitude value changes. For example, an area with an altitude greater than 3800 meters is a glacier area, and an area with an altitude between 200 and 500 meters may be a hilly area. However, the types of ground surfaces corresponding to different terrain areas are different, for example, glaciers are the types of ground surfaces in glaciers and grasslands are the types of ground surfaces in hilly areas.

Therefore, in the embodiment of the present invention, the surface types corresponding to the different terrain areas are presented by using the different types of maps, that is, in the embodiment of the present invention, the surface types of the areas corresponding to the different height values are different, that is, the maps of the areas corresponding to the different height values are different. For example, the map used to characterize glacier regions having a height value greater than 3800 meters may be a glacier map; the map used to characterize hilly areas with height values of 200 to 500 meters may be a grass map.

Specifically, when generating a target terrain mesh of a scene to be generated, the electronic device may generate candidate maps of different surface types.

The surface type can be wetland, rock, snow mountain, sand and the like. The embodiment of the present invention is not particularly limited in this regard. Correspondingly, if the earth surface type is wetland, the correspondingly generated candidate map is a wetland vegetation map; the earth surface type is rock, and then the generated candidate chartlet rock chartlet is correspondingly generated; the surface type is snow mountain, and the corresponding generated candidate map is snow map; and if the surface type is sand, the corresponding generated candidate map is a brown sand map.

In one embodiment of the present invention, multiple maps of different surface types may be obtained in advance as the map to be fused, so that when generating the candidate map, any two maps to be fused are directly fused to obtain the candidate map.

In order to improve the rationality of the generated candidate map, in one embodiment of the present invention, at least two map to be fused of the same surface type may be fused when generating the candidate map, and a new map of the same surface type as the above map to be fused may be generated as the candidate map of the surface type.

Thus, after generating the candidate maps with different surface types, the electronic device can determine, for each region in the terrain height grid, a target map matched with the surface type of the region in the candidate maps according to the height value of the region.

For example, if the terrain elevation value characterizes the region as a wetland region, the surface type of the region being wetland, a wetland vegetation map may be determined as a target map that matches the surface type of the region; if the terrain height value represents that the area is a plain area and the surface type of the area is rock, determining a rock map as a target map matched with the surface type of the area; if the terrain elevation value characterizes the area as a beach area and the surface type of the area is sand, a brown sand map may be determined as a target map that matches the surface type of the area.

During the formation of the earth's surface in a region, the type of earth's surface in different regions may also be affected by the slope at which the region is located. For example, a slope of 55 degrees to 90 degrees, where the area is a vertical wall, there is no significant vegetation growing in the area, and the grassland vegetation map cannot be taken as the target map for the area; the slope is 0 degrees to 0.5 degrees, the area is plain, a great amount of vegetation can be grown in the area, and the grassland vegetation map can be used as a target map of the area.

Based on this, in one embodiment of the present invention, for each region in the terrain elevation grid, a plurality of candidate maps matching with the elevation value of the region are first screened out from the candidate maps according to the elevation value of the region, and then, a target map matching with the earth surface type of the region is determined from the screened plurality of candidate maps according to the gradient of the region. Thus, the target map of the region is determined in combination with the height value and the gradient of the region to improve the accuracy of the determined target map.

The gradient of the region can be calculated based on the height values of the region and the region in a preset range nearby.

In addition, in the formation process of the regional surface, different surface types can be influenced by longitude and latitude of the region. For example, the surface type of the ice bank is different from that of the tropical rain forest bank, and thus, a map matching the surface type of the ice bank cannot be used as a target map of the tropical rain forest bank.

Based on this, in one embodiment of the present invention, for each region in the terrain elevation grid, first, a plurality of candidate maps matching with the longitude and latitude of the region are screened out from the candidate maps according to the longitude and latitude of the region, and then, a target map matching with the earth surface type of the region is determined from the screened plurality of candidate maps according to the elevation value of the region, so that the influence of the longitude and latitude and the elevation value on the earth surface type is comprehensively considered to improve the accuracy of the determined target map.

Step S103: and rendering the target map to the corresponding area, and generating a target terrain grid of the scene to be generated.

Specifically, after determining, for each region in the terrain elevation grid, a target map matching with the surface type of the region in the candidate map according to the elevation value of the region, the electronic device may render the determined target map to the corresponding region, thereby generating the target terrain grid of the scene to be generated.

In the embodiment of the invention, the generated terrain height grid is a terrain height white model, namely a white three-dimensional scene model, in the three-dimensional scene generation process, so that after the determined target map is rendered to the terrain height grid, a colored three-dimensional scene model, namely the target terrain grid of the scene to be generated, can be obtained.

For example, if the terrain elevation value characterizes the region as a wetland region, rendering a wetland vegetation map to the region; if the terrain height value represents that the area is a plain area, rendering grassland or forest map to the area; if the terrain elevation value characterizes the area as a beach area, a brown sand map is rendered to the area.

The process of rendering the determined target map to the corresponding region is a process of attaching the target map to the programmed material related to the three-dimensional modeling, and attaching the attached programmed material to the region corresponding to the terrain elevation grid.

Step S104: different types of biological meshes are generated, and a target biological mesh is allocated to each region in the terrain elevation mesh.

In particular, in generating a three-dimensional scene, living beings in the scene are also one of the indispensable scene elements. And, as the height of the terrain increases, the type of living organism in the area corresponding to the height of the terrain is different. For example, saussurea involucrata grows in glacier areas with a terrain height value greater than 3800 meters, and coconut trees grow in beach areas with a terrain height value of about 150 meters. Therefore, the areas corresponding to different terrain height values can be assigned different biological types.

Based on this, the electronic device may generate different types of biological grids, such that for each region in the terrain elevation grid, a target biological grid is assigned to that region. The types of biological lattices may include animals and plants, and the embodiments of the present invention are not particularly limited.

In one embodiment of the present invention, to increase the efficiency of generating the target biological grid, the electronic device may determine a surface type of each region in the terrain elevation grid, so that the target biological grid matching the surface type of the region is directly allocated based on the surface type of the region.

For example, if the surface type of the area is wetland, biological grids of lotus and reed are allocated to the area; if the surface type of the area is plain, a biological grid of shrubs is distributed to the area; if the surface type of the area is sandy, the area is allocated with biological grids of aloe.

For another example, if the surface type of the area is ocean, a biological grid of fish is allocated to the area; if the surface type of the area is wetland, the biological grid of the geranium is distributed to the area.

Step S105: and generating a basic path of the scene to be generated, and adjusting the basic path to be attached to the terrain height grid to obtain the road grid of the scene to be generated.

In the process of generating a three-dimensional scene, the path of the scene to be generated is an important factor for determining how the biological communities and the building communities are distributed in the scene. Specifically, when generating the three-dimensional scene, the electronic device may generate the base path of the to-be-generated scene, and in order to improve the authenticity and rationality of the three-dimensional scene, adjust the base path to be attached to the terrain height grid, thereby obtaining the road grid of the to-be-generated scene.

For example, a steep cliff may not have a path, a lake may not have a path, and the like.

Step S106: and generating a building community, determining the distribution position of the building community based on the road grid, and generating a building community grid of the scene to be generated according to the determined distribution position.

In order to increase the diversity of the scene to be generated, the electronic device may generate building communities for distribution in the scene to be generated, and determine distribution positions of the building communities based on the generated road grids, thereby generating building community grids of the scene to be generated according to the determined distribution positions.

In order to improve the rationality and the authenticity of the generated building community grid, in one embodiment of the invention, the distribution position of the building community is determined on the generated road grid according to the building community distribution rule. The building community distribution rule may be that the building cannot be distributed on the road, the building community can only be distributed on the flat land, and the like, and the embodiment of the invention is not particularly limited.

In addition, the distribution position of the building community is determined based on the road grid, and the road grid is generated by attaching the basic path to the terrain height grid, so that the building community grid of the scene to be generated according to the determined distribution position is also attached to the terrain height grid, and further, the display effect of the generated scene to be generated can be improved.

Step S107: and generating a three-dimensional scene of the scene to be generated by combining the target terrain grid, the target biological grid, the road grid and the building community grid corresponding to each region.

After the electronic equipment generates the target terrain grids, the target biological grids corresponding to the areas, the road grids and the building community grids of the scene to be generated, the target terrain grids, the target biological grids corresponding to the areas, the road grids and the building community grids can be combined to directly render and generate the three-dimensional scene of the scene to be generated, so that the automation of the generation process of the three-dimensional scene is realized, the manufacturing period of the three-dimensional scene is reduced, and the generation efficiency of the three-dimensional scene is further improved.

In addition, the target terrain mesh of the scene to be generated is generated based on the candidate map and the terrain elevation mesh, the target biological mesh corresponding to each region is generated based on the target terrain mesh, the road mesh is generated by attaching a base path to the terrain elevation mesh, the distribution position of the building community in the building community mesh is determined based on the road mesh, and in summary, it can be determined that the generated target biological mesh, road mesh and building community mesh are matched with the terrain elevation mesh of the scene to be generated, thereby improving the presentation effect of the three-dimensional scene of the generated scene to be generated.

In one embodiment of the present invention, in order to observe the generation process of the three-dimensional scene in real time, the generation process may be output to a display for presentation. It is reasonable that the display and the electronic device may be disposed in the same device or may be disposed in different devices. The embodiment of the present invention is not particularly limited in this regard.

Referring to fig. 2, a flow chart of a second method for generating three-dimensional world topography based on artificial intelligence is provided in an embodiment of the present invention. In comparison with the embodiment shown in fig. 1 described above, the aforementioned step S101 may be implemented by the following step S101A.

Step S101A: and randomly generating a plurality of random numbers, respectively serving as terrain height values corresponding to all points in the two-dimensional image, taking the two-dimensional image added with the terrain height values as a terrain height map of a scene to be generated, and converting the terrain height map into a three-dimensional terrain height grid.

Specifically, when generating a topographic elevation map of a scene to be generated, in order to improve the randomness and the authenticity of the generated topographic elevation map, firstly, randomly generating a plurality of random numbers which are respectively used as topographic elevation values corresponding to various points in a two-dimensional image, and adding each topographic elevation value to the corresponding point, thereby taking the two-dimensional image added with the topographic elevation values as the topographic elevation map of the scene to be generated, so that the generated topographic elevation map can simulate the height variation of natural topography, and then, converting the generated topographic elevation map into a three-dimensional topographic elevation grid so as to improve the randomness and the authenticity of the generated topographic elevation grid.

In one embodiment of the invention, a plurality of noise data are randomly generated by using a noise wave generation algorithm, the noise data are respectively used as the topographic elevation values corresponding to all points in the two-dimensional image, and the two-dimensional image added with the topographic elevation values is used as the topographic elevation map of the scene to be generated. The noise data is the random number provided by the embodiment of the invention.

The noise generation algorithm may be a berlin noise generation algorithm, a simplex noise generation algorithm, or the like. The embodiment of the present invention is not particularly limited in this regard.

In some cases, the randomness of the random numbers may be too high, which may cause the terrain condition represented by the generated terrain height value to not conform to the actual situation in the real world, so that the terrain planning of the generated three-dimensional scene is not reasonable. For example, among three points adjacent to each other in sequence in the topographic elevation map, the topographic elevation value of the first point is 100 meters, the topographic elevation value of the second point is 0 meters, and the topographic elevation value of the third point is 1000 meters. This is clearly not justified.

Therefore, in order to improve the rationality of the generated three-dimensional scene, the authenticity of the generated three-dimensional scene is improved. In one embodiment of the present invention, in the step S101A, the step of using the two-dimensional image with the added terrain height value as the terrain height map of the scene to be generated may include the following steps A1-A2:

step A1: and smoothing the terrain height value in the two-dimensional image.

Step A2: and taking the smoothed two-dimensional image as a topographic elevation map of the scene to be generated.

Specifically, after generating and obtaining the topographic elevation values corresponding to the points in the two-dimensional image of the scene to be generated, the electronic device may perform smoothing processing on the topographic elevation values in the two-dimensional image, so that the smoothed two-dimensional image is used as the topographic elevation map of the scene to be generated, so as to improve the rationality and the authenticity of the generated three-dimensional scene.

The fractal algorithm may be used to smooth the terrain height value in the two-dimensional image, which is not specifically limited in the embodiment of the present invention.

In some cases, in order to increase the efficiency of the generation of the desired candidate map, the desired surface type of candidate map may be directly generated using a map generation model.

In one embodiment of the present invention, in the step S102, the step of generating candidate maps of different surface types includes the following step B:

And (B) step (B): and inputting the prompt words representing different surface types into a pre-trained map generation model to obtain candidate maps corresponding to the surface types represented by the prompt words.

The map generation model is a generation type artificial intelligent model.

Specifically, a pre-trained map generation model may be utilized to generate candidate maps of different surface types. The map generation model is a generated artificial intelligent model, namely, the prompt word is input into the model, and the model can generate a required map.

Therefore, the electronic device can input the prompt words representing different surface types into the map generation model, and the map generation model generates the maps corresponding to the surface types represented by the prompt words as candidate maps.

The prompting words may be beach, desert, wetland, snow mountain, cliff, etc., which is not particularly limited in the embodiment of the present invention.

In addition, when the map generation model is utilized, the maps generated by the same prompt word at different input moments can be different, so that the map type of the generated three-dimensional scene can be expanded by utilizing the map generation model to generate the maps so as to enrich the diversity of the maps, and therefore, the diversity of the generated three-dimensional scene is improved.

In addition, in order to further improve the efficiency of generating the map and thus save the time for generating the map of the required surface type, the prompt words may also include words for characterizing the surface details, such as the surface color and the surface texture. The surface color may be white snowfield, blue glacier, brown rock, brown sand, etc., and the surface texture may be a zebra texture, a stripe texture, etc., which is not particularly limited in the embodiment of the present invention.

In order to improve the accuracy of the generated target biological grid, in one embodiment of the present invention, the step of assigning a target biological grid to each region in the terrain elevation grid in the step S104 may include the following steps C1-C3:

step C1: triangularizing the terrain height grid to obtain a triangular grid for representing the terrain surface of the scene to be generated;

step C2: different terrain characteristic values are randomly configured for each triangular grid;

Step C3: and determining the target biological grids corresponding to the triangular grids based on the corresponding relation between the preset terrain characteristic values and the types of the biological grids.

Specifically, after generating different types of biological grids, the electronic device may perform triangulation processing on the terrain elevation grid, thereby obtaining a triangular grid for characterizing the terrain surface of the scene to be generated. In addition, in order to improve the randomness and texture effect of the generated three-dimensional scene, the electronic device can randomly configure different terrain characteristic values for each triangular mesh.

In one embodiment of the present invention, the terrain elevation grid is triangulated using a predetermined triangulating algorithm. The preset triangulation algorithm may be a Thiessen polygon algorithm, a triangle subdivision algorithm, or a nearest neighbor search algorithm, which are all reasonable, and the embodiment of the present invention is not limited specifically.

In one embodiment of the invention, a plurality of random numbers are randomly generated by using a noise wave diagram, and the generated random numbers are used as terrain features and are randomly configured to each triangular grid.

In order to improve the authenticity and rationality of the generated three-dimensional scene, a correspondence relationship between the terrain feature value and the type of the biological grid may be pre-established, wherein in the correspondence relationship, different terrain features correspond to different types of biological grids, for example, the terrain feature value is 100, and the type of the corresponding biological grid may be coco tree; the terrain feature value is 102 and the corresponding biological mesh may be of the palm tree type.

In this way, after different terrain characteristic values are randomly configured for each triangular grid, the target biological grid corresponding to each triangular grid can be determined based on the corresponding relation between the preset terrain characteristic values and the types of the biological grids, so that the distribution of living beings in the three-dimensional scene is realized.

And, the correspondence between the terrain characteristic value and the type of the biological grid, which is established in advance, can improve the rationality and accuracy of biological distribution, thereby improving the accuracy of the generated target biological grid.

Symbiotic relationships between organisms under consideration, for example, sea anemones and clown fishes, crocodiles and toothpick birds, etc. Thus, there may be multiple types of biological meshes in the same triangular mesh. Based on this, in one embodiment of the present invention, a plurality of terrain feature intervals may be set in advance for the terrain feature values, so that a correspondence relationship between the terrain feature intervals and the types of biological meshes is established, that is, the same terrain feature interval may correspond to a plurality of types of biological meshes.

In this way, after different terrain feature values are randomly arranged for each triangular mesh, a target section to which the triangular mesh belongs is determined for each triangular mesh, and thus each biological mesh corresponding to the target section is determined as each target biological mesh of the triangular mesh according to the correspondence between the terrain feature section and the type of biological mesh.

In the embodiment of the invention, when a plurality of types of biological grids exist in the same triangular grid, the types and the numbers of the target biological grids of the triangular grid can be distributed according to a preset biological type distribution rule.

In an embodiment of the present invention, the preset biological type allocation rule may be a random allocation. For example, sea anemones and clown fishes are included in the same triangular mesh, and since the two biological types belong to a symbiotic relationship, the respective numbers of the two biological meshes can be randomly allocated.

In an embodiment of the present invention, the preset biological type allocation rule may be an allocation according to a preset weight coefficient of each type of biological grid, where the smaller the weight coefficient is, the smaller the number of corresponding biological grids is. For example, the same triangular mesh includes zebra and leopard, and the two biological types belong to the relationship between predators and predators, so that the weight coefficient of the leopard can be preset to be 1, and the weight coefficient of the zebra is preset to be 5, namely, when the respective numbers of the two biological meshes are allocated, the two biological meshes are allocated according to the number ratio of 1:5.

The number relationship between the weight coefficient and the type of the biological grid described in the above embodiment is only an example, and the number relationship between the weight coefficient and the type of the biological grid in the embodiment of the present invention is not limited.

Referring to fig. 3, a flow chart of a third method for generating three-dimensional world topography based on artificial intelligence according to an embodiment of the present invention is provided. Compared with the embodiment shown in fig. 1, the step S105 generates the base path of the scene to be generated, and adjusts the base path to be attached to the terrain height grid, so as to obtain the road grid of the scene to be generated, which may be implemented in the following steps S105A to S105B.

Step S105A: determining a region contained in a scene to be generated, and dividing the region into subareas;

Step S105B: and determining boundary lines between adjacent subareas as basic paths of the scene to be generated, and adjusting the basic paths to be attached to the terrain height grids to obtain the road grids of the scene to be generated.

Specifically, when the electronic device generates the base path, the region included in the to-be-generated scene may be determined first, so that the region is divided into a plurality of sub-regions. In this way, the boundary line between adjacent subareas can be determined as the basic path of the scene to be generated, and the basic path is adjusted to be attached to the terrain height grid in order to improve the authenticity and rationality of the three-dimensional scene, so that the road grid of the scene to be generated is obtained.

For example, referring to fig. 4, a schematic diagram of generating a road grid according to an embodiment of the present invention is shown, and fig. 5 is a schematic diagram of a portion of a quadtree according to an embodiment of the present invention.

As shown in fig. 4, the region 401 included in the scene to be generated is divided by using an L-system algorithm, each node in the divided region may be used as a sub-region, 4 sub-regions a, b, c and d are obtained, boundary lines between adjacent sub-regions are determined as paths of the scene to be generated, and the dividing process is recorded in a quadtree, thereby obtaining an initial path of the scene to be generated.

Then, for each sub-region, the sub-region a is divided twice by using the L-system algorithm, and the sub-region a is further divided twice by using the L-system algorithm as an example, and each node in the divided sub-region is used as a sub-region in the sub-region, so as to obtain 4 sub-regions a1, a2, a3 and a4, and boundary lines between adjacent sub-regions are determined as paths of a scene to be generated.

After that, the above-mentioned dividing process is recorded in the quadtree, after the quadtree is updated, the step of performing region division by using the L-system algorithm is returned, so that for the node containing the path, it is further decomposed into smaller nodes, so as to improve the accuracy of the generated path.

Repeating the steps until reaching the preset stopping condition, and obtaining the basic path of the scene to be generated.

The preset stopping condition may be that the update times reach the preset update times, the number of the divided sub-areas reaches the preset area number, and the area of the divided minimum sub-area is smaller than the area of the preset area. The embodiment of the present invention is not particularly limited in this regard.

In order to improve the rationality of the generated basic path and the authenticity of the generated three-dimensional scene, in one embodiment of the invention, the boundary line between adjacent subareas is determined as the basic path of the scene to be generated according to a preset path determination rule.

The preset path determining rule may be: the bending degree of the path cannot be larger than 180 degrees, and the number of the intersecting points existing on the same boundary line is the number of preset intersecting points. The embodiment of the present invention is not particularly limited in this regard.

In one embodiment of the present invention, in order to improve the generation efficiency of the base path, after determining the region included in the scene to be generated, the region may be divided into sub-regions by using a preset dividing line, and the boundary line between adjacent sub-regions is determined as the base path of the scene to be generated, thereby implementing the generation of the base path.

Referring to fig. 6, a flow chart of a building generation method provided by an embodiment of the present invention is shown in fig. 6, in an embodiment of the present invention, a building community includes buildings, and each building in the building community may be generated through the following steps S601 to S602:

Step S601: selecting a plurality of building units from a preset building unit group to be combined into an initial building structure;

step S602: and (3) performing structure rationalization treatment on the initial building structure to generate buildings in the building community.

Specifically, in order to improve the generation efficiency of the three-dimensional scene, a preset building unit group for generating a building may be established in advance. The preset building unit group may include building units such as wall units, roof units, window units, and door units, which are not particularly limited in this embodiment of the present invention.

In addition, in order to enrich the shapes of the building units, in one embodiment of the present invention, each building unit included in the preset building unit group may include at least one building type. For example, the building types of the roof unit may include flat roof and pitched roof, and the like.

Thus, when the electronic device generates each building in the building community, a plurality of building units can be selected from the preset building unit groups to be combined into an initial building structure.

Because the initial building structures are randomly combined, the resulting initial building structure may not be as rational as possible, e.g., roof units are not attached to wall units, door units are not provided on the first floor of the overall building, the building is not provided with window units, etc. Therefore, after the initial building structure is obtained, the initial building structure can be subjected to the structure rationalization treatment, thereby obtaining a structure rationalized building as a building in the building community.

In one embodiment of the present invention, the initial building structure is structurally rationalized using a wave function collapse algorithm, so that the buildings in the building community are generated according to the physical rules and the probability distribution corresponding to each building unit. Therefore, the generated building has rich shapes and structures, and the authenticity and the consistency of physical properties of the building are improved.

The physical rules may include gravity, material properties, structural stability, and the like. The embodiment of the present invention is not particularly limited in this regard.

In one embodiment of the present invention, in order to enhance the presentation effect of the generated building, after the building is generated, a corresponding texture map may be added to each building, thereby increasing the reality of the visual effect of the building. The texture map may be color, material, texture, etc. of the building unit. The embodiment of the present invention is not particularly limited in this regard.

In one embodiment of the invention, in order to improve the building generation efficiency, a preset building group is pre-established, and the preset building group contains independent buildings of different types and multiple building styles, so that the building is directly obtained from the preset building group when the building is generated, thereby improving the building generation efficiency and further improving the generation efficiency of the three-dimensional scene.

In order to facilitate understanding of the method for generating three-dimensional world topography based on artificial intelligence provided by the embodiment of the invention, a specific description will be given below using an embodiment.

Referring to fig. 7, a schematic flow chart of a specific embodiment of the present invention is provided.

As shown in fig. 7, when generating a 3D scene of a product, firstly, a berlin noise wave and a fractal algorithm may be used to generate a height map, where the 3D scene of the product is a three-dimensional scene of the scene to be generated in the embodiment of the present invention, and the height map is a topographic height map in the embodiment of the present invention.

Then, grid data with the length, width and height consistent with the height map are generated by sampling the height map, and a basic terrain grid white model is generated. The grid data is the terrain height grid in the embodiment of the invention, and the basic terrain grid white mode is the terrain height white mode in the embodiment of the invention.

Then, the prompting words such as beach, desert, wetland, snow mountain, cliff and the like are input into a pre-training AI (ARTIFICIAL INTELLIGENCE ) mapping model, so that the candidate mapping of the surface type represented by the prompting words is generated by using the pre-training AI mapping model. The pre-trained AI mapping model is a mapping generation model in the embodiment of the invention.

And acquiring gradient, height and longitude and latitude from the grid data, and selecting an automatic terrain material matched with the earth surface type of each area in the grid data based on the gradient, the height and the longitude and latitude. The gradient, the height and the longitude and latitude are the elevation data, the gradient data and the longitude and latitude data of each point represented by the terrain elevation grid in the embodiment of the invention; the automatic terrain material is the target map in the embodiment of the invention.

Thus, the generated automatic terrain material and the basic terrain grid white model can be fused to generate the basic terrain grid. The basic terrain grid is the target terrain grid in the embodiment of the invention.

Then, after the basic terrain height data is obtained from the grid data, basic path data is generated by using an L-Systerm algorithm and a quadtree, and the basic path data is adjusted by the basic path data and the basic terrain height data, so that an adjusted path of the attached terrain is obtained, and the path of the attached terrain is used as a road grid, so that the road grid is generated. The basic terrain height data are terrain height grids in the embodiment of the invention, and the basic path data are basic paths in the embodiment of the invention.

And then generating a grid building by utilizing a wave function collapse algorithm, and fusing the obtained path attached to the terrain with the grid building to obtain a building community distribution grid generated around the road. The grid building is a building community in the embodiment of the invention, and the building community distribution grid generated around the road is the building community grid in the embodiment of the invention.

And (3) fusing the gradient, the height and the longitude and latitude obtained from the grid data by using a Thiessen polygon algorithm and a noise wave map to generate community map data, thereby generating a vegetation grid. The community map data is the target biological grid in the embodiment of the invention, and the vegetation grid is the target biological grid corresponding to each area in the embodiment of the invention.

And finally, fusing the basic terrain grids, the road grids, the building community distribution grids generated around the roads and the vegetation grids to obtain the 3D scene of the product, thereby realizing the automatic generation of the three-dimensional scene.

The basic terrain grids, the road grids, the building community distribution grids generated around the roads and the vegetation grids are generated based on the grid data, namely, the basic terrain grids, the road grids, the building community distribution grids generated around the roads and the vegetation grids are attached to the grid data, so that the generated 3D scene of the product is more real and reasonable.

Corresponding to the method for generating the three-dimensional world topography based on the artificial intelligence provided by the embodiment of the invention, the embodiment of the invention also provides a device for generating the three-dimensional world topography based on the artificial intelligence.

Fig. 8 is a schematic structural diagram of an apparatus for generating three-dimensional world topography based on artificial intelligence according to an embodiment of the present invention, and as shown in fig. 8, the apparatus may include the following modules:

The altitude grid generation module 801 is configured to generate a terrain altitude map of a scene to be generated, and convert the terrain altitude map into a three-dimensional terrain altitude grid;

A target map determining module 802, configured to generate candidate maps of different surface types, and determine, for each region in the terrain elevation grid, a target map matching the surface type of the region according to the elevation value of the region in the candidate maps;

The map rendering module 803 is configured to render a target map to a corresponding area, and generate the target terrain mesh of the scene to be generated;

a biological grid allocation module 804, configured to generate different types of biological grids, and allocate, for each region in the terrain elevation grid, a target biological grid for the region;

The road grid generating module 805 is configured to generate a base path of the to-be-generated scene, and adjust the base path to be attached to the terrain height grid to obtain a road grid of the to-be-generated scene;

The building community generation module 806 is configured to generate a building community, determine a distribution position of the building community based on the road grid, and generate a building community grid of the scene to be generated according to the determined distribution position;

the three-dimensional scene rendering module 807 is configured to combine the target terrain mesh, the target biological mesh corresponding to each region, the road mesh, and the building community mesh to generate the three-dimensional scene of the scene to be generated.

In one embodiment of the present invention, the altitude grid generation module 801 includes:

In the scheme provided by the embodiment of the invention, the random numbers generated randomly are respectively used as the terrain height values corresponding to the points in the two-dimensional image, so that the generated terrain height map can simulate the height change of natural terrain, and the randomness and the authenticity of the generated terrain height grid are further improved.

In one embodiment of the present invention, the terrain highly random generation sub-module is specifically configured to:

smoothing the topographic elevation value in the two-dimensional image;

In the above, in the scheme provided by the embodiment of the invention, in order to avoid that the terrain condition represented by the generated terrain height value is not in line with the actual condition in the real world due to the too high randomness of the random number, the terrain planning of the generated three-dimensional scene is not reasonable. After the terrain height value in the two-dimensional image is obtained, the terrain height value can be subjected to smoothing treatment, so that the terrain represented by the generated terrain height map accords with the actual situation in the real world, and the rationality and the authenticity of the generated three-dimensional scene are improved.

In one embodiment of the present invention, the road grid generating module 805 is specifically configured to:

In the scheme provided by the embodiment of the invention, after the region included in the scene to be generated is divided into the subareas, the boundary line between the adjacent subareas can be used as the basic path of the scene to be generated, so that the generated basic path is more attached to the actual situation, and the rationality and the planning performance of the generated basic path are improved.

As can be seen from the above, in the solution provided in the embodiment of the present invention, since the initial building structures are randomly combined, the obtained initial building structures may not meet the normal requirements, and therefore, after the initial building structures are obtained, the initial building structures may be subjected to structural rationalization, so that the generated building meets the actual situation, and further, the authenticity and rationality of the generated three-dimensional scene are improved.

In one embodiment of the present invention, the biological grid allocation module 804 is specifically configured to:

In the above, in the solution provided by the embodiment of the present invention, in order to improve randomness and texture effects of the generated three-dimensional scene, the electronic device may randomly configure different terrain feature values for each triangular mesh, and in order to improve authenticity and rationality of the generated three-dimensional scene, determine, based on the corresponding relationship between the preset terrain feature values and types of biological meshes, the target biological mesh corresponding to each triangular mesh, thereby implementing distribution of living things in the three-dimensional scene, and further improving authenticity and rationality of the generated three-dimensional scene.

In one embodiment of the present invention, the target map determining module 802 is specifically configured to:

In the above, in the solution provided by the embodiment of the present invention, in order to improve the efficiency of generating the required candidate maps, the electronic device may input the prompt words representing different surface types into the map generation model, and directly generate, as the candidate maps, the maps corresponding to the surface types represented by the prompt words by using the map generation model. Moreover, when the map generation model is utilized, the maps generated by the same prompt word at different input moments can be different, so that the map type of the generated three-dimensional scene can be expanded by utilizing the map generation model to generate the maps so as to enrich the diversity of the maps, and the diversity of the generated three-dimensional scene is improved.

Corresponding to the method for generating three-dimensional world topography based on artificial intelligence provided by the embodiment of the invention, the embodiment of the invention also provides an electronic device, as shown in fig. 9, comprising a processor 901, a communication interface 902, a memory 903 and a communication bus 904, wherein the processor 901, the communication interface 902 and the memory 903 complete communication with each other through the communication bus 904,

A memory 903 for storing a computer program;

the processor 901 is configured to implement any of the methods for three-dimensional world topography generated based on artificial intelligence provided in the embodiments of the present invention described above when executing the program stored in the memory 903.

When the electronic equipment provided by the embodiment of the invention is used for generating three-dimensional world topography, firstly, a topography height map of a scene to be generated is generated, and in order to facilitate the generation of the three-dimensional scene, the topography height map is converted into a three-dimensional topography height grid. Secondly, because the earth surface types corresponding to different terrains are different, in order to improve the authenticity of the generated three-dimensional scene, candidate maps with different earth surface types can be generated, and for each region in the terrain height grid, a target map matched with the earth surface type of the region is determined in the candidate maps according to the height value of the region. Then, the determined target map can be rendered to the corresponding area, so that the target terrain grid of the scene to be generated is generated. Then, different types of biological grids are generated, and a target biological grid is allocated to each region in the terrain elevation grid. In order to improve the rationality of the generated three-dimensional scene in road division, after the basic path of the scene to be generated is generated, the basic path and the terrain height grid can be adjusted to be attached to each other, so that the road grid of the scene to be generated is obtained. Then, after the building group is generated, the distribution position of the generated building group is determined based on the road grid, and the building group grid of the scene to be generated is generated according to the determined distribution position. Thus, the three-dimensional scene of the scene to be generated can be generated by combining the generated target terrain mesh, target biological mesh corresponding to each region, the road mesh, and the building community mesh.

The communication bus mentioned by the above terminal may be a peripheral component interconnect standard (PERIPHERAL COMPONENT INTERCONNECT, abbreviated as PCI) bus or an extended industry standard architecture (Extended Industry Standard Architecture, abbreviated as EISA) bus, etc. The communication bus may be classified as an address bus, a data bus, a control bus, or the like. For ease of illustration, the figures are shown with only one bold line, but not with only one bus or one type of bus.

The communication interface is used for communication between the terminal and other devices.

The memory may include random access memory (Random Access Memory, RAM) or may include non-volatile memory (non-volatile memory), such as at least one disk memory. Optionally, the memory may also be at least one memory device located remotely from the aforementioned processor.

The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, abbreviated as CPU), a network processor (Network Processor, abbreviated as NP), etc.; but may also be a digital signal Processor (DIGITAL SIGNAL Processor, DSP), application Specific Integrated Circuit (ASIC), field-Programmable gate array (FPGA) or other Programmable logic device, discrete gate or transistor logic device, discrete hardware components.

In yet another embodiment of the present invention, a computer readable storage medium is provided, in which a computer program is stored, which when executed by a processor implements the method of any of the above embodiments based on artificial intelligence generated three-dimensional world topography.

When the computer program stored in the computer readable storage medium provided by the embodiment of the invention is applied to generate three-dimensional world topography, firstly, a topography height map of a scene to be generated is generated, and in order to facilitate the generation of the three-dimensional scene, the topography height map is converted into a three-dimensional topography height grid. Secondly, because the earth surface types corresponding to different terrains are different, in order to improve the authenticity of the generated three-dimensional scene, candidate maps with different earth surface types can be generated, and for each region in the terrain height grid, a target map matched with the earth surface type of the region is determined in the candidate maps according to the height value of the region. Then, the determined target map can be rendered to the corresponding area, so that the target terrain grid of the scene to be generated is generated. Then, different types of biological grids are generated, and a target biological grid is allocated to each region in the terrain elevation grid. In order to improve the rationality of the generated three-dimensional scene in road division, after the basic path of the scene to be generated is generated, the basic path and the terrain height grid can be adjusted to be attached to each other, so that the road grid of the scene to be generated is obtained. Then, after the building group is generated, the distribution position of the generated building group is determined based on the road grid, and the building group grid of the scene to be generated is generated according to the determined distribution position. Thus, the three-dimensional scene of the scene to be generated can be generated by combining the generated target terrain mesh, target biological mesh corresponding to each region, the road mesh, and the building community mesh.

In yet another embodiment of the present invention, there is also provided a computer program product containing instructions that, when run on a computer, cause the computer to perform the method of any of the above embodiments based on artificial intelligence generated three-dimensional world topography.

When the computer program product provided by the embodiment of the invention is applied to generate three-dimensional world terrains, firstly, a terrain height map of a scene to be generated is generated, and in order to facilitate the generation of the three-dimensional scene, the terrain height map is converted into a three-dimensional terrain height grid. Secondly, because the earth surface types corresponding to different terrains are different, in order to improve the authenticity of the generated three-dimensional scene, candidate maps with different earth surface types can be generated, and for each region in the terrain height grid, a target map matched with the earth surface type of the region is determined in the candidate maps according to the height value of the region. Then, the determined target map can be rendered to the corresponding area, so that the target terrain grid of the scene to be generated is generated. Then, different types of biological grids are generated, and a target biological grid is allocated to each region in the terrain elevation grid. In order to improve the rationality of the generated three-dimensional scene in road division, after the basic path of the scene to be generated is generated, the basic path and the terrain height grid can be adjusted to be attached to each other, so that the road grid of the scene to be generated is obtained. Then, after the building group is generated, the distribution position of the generated building group is determined based on the road grid, and the building group grid of the scene to be generated is generated according to the determined distribution position. Thus, the three-dimensional scene of the scene to be generated can be generated by combining the generated target terrain mesh, target biological mesh corresponding to each region, the road mesh, and the building community mesh.

In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, produces a flow or function in accordance with embodiments of the present invention, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another, for example, by wired (e.g., coaxial cable, optical fiber, digital Subscriber Line (DSL)), or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk Solid STATE DISK (SSD)), etc.

It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

In this specification, each embodiment is described in a related manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for the apparatus embodiments, the electronic device embodiments, the computer-readable storage medium embodiments, and the computer program product embodiments, the description is relatively simple, as relevant to the description of the method embodiments in part, since they are substantially similar to the method embodiments.

The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A method of three-dimensional world topography generated based on artificial intelligence, the method comprising:

Generating a building community, determining the distribution position of the building community based on the road grid, and generating the building community grid of the scene to be generated according to the determined distribution position;

Combining the target terrain grids, the target biological grids corresponding to each region, the road grids and the building community grids to generate a three-dimensional scene of the scene to be generated;

the generating the topographic elevation map of the scene to be generated comprises the following steps:

Randomly generating a plurality of random numbers by using a noise wave generation algorithm, respectively serving as terrain height values corresponding to all points in a two-dimensional image of a scene to be generated, and adding each terrain height value to a corresponding point;

taking the two-dimensional image added with the terrain height value as a terrain height map of the scene to be generated; wherein the two-dimensional image added with the terrain height value represents a height change of a natural terrain of the scene to be generated;

the generating candidate maps of different surface types includes:

2. The method according to claim 1, wherein said adding the two-dimensional image of the terrain height value as a terrain height map of the scene to be generated comprises:

smoothing the topographic elevation value in the two-dimensional image;

3. The method of claim 1, wherein the generating the base path of the scene to be generated comprises:

and determining boundary lines between adjacent subareas as a basic path of the scene to be generated.

4. The method of claim 1, wherein the building community comprises buildings, each building in the building community being generated by:

Selecting a plurality of building units from a preset building unit group to be combined into an initial building structure; wherein the preset building unit group comprises at least two building units of a wall unit, a roof unit, a window unit and a door unit;

And carrying out structure rationalization treatment on the initial building structure to generate the buildings in the building community.

5. The method of claim 1, wherein said assigning a target biological grid to each region in said terrain elevation grid comprises:

Performing triangulation processing on the terrain height grid to obtain a triangular grid used for representing the terrain surface of the scene to be generated;

6. An apparatus for generating three-dimensional world topography based on artificial intelligence, the apparatus comprising:

The system comprises a height grid generation module, a height grid generation module and a display module, wherein the height grid generation module is used for generating a terrain height map of a scene to be generated and converting the terrain height map into a three-dimensional terrain height grid;

The building community generation module is used for generating a building community, determining the distribution position of the building community based on the road grid, and generating the building community grid of the scene to be generated according to the determined distribution position;

The three-dimensional scene rendering module is used for generating a three-dimensional scene of the scene to be generated by combining the target terrain grid, the target biological grid corresponding to each region, the road grid and the building community grid;

wherein, high grid generation module includes:

The terrain height random generation sub-module is used for randomly generating a plurality of random numbers by utilizing a noise wave generation algorithm, respectively serving as terrain height values corresponding to all points in a two-dimensional image of a scene to be generated, and adding each terrain height value to a corresponding point; taking the two-dimensional image added with the terrain height value as a terrain height map of the scene to be generated; wherein the two-dimensional image added with the terrain height value represents a height change of a natural terrain of the scene to be generated;

The target map determining module is specifically configured to:

7. The apparatus of claim 6, wherein the terrain highly random generation sub-module is configured to:

smoothing the topographic elevation value in the two-dimensional image;

8. The apparatus of claim 6, wherein the road grid generation module is specifically configured to:

9. The apparatus of claim 6, wherein the building community comprises buildings, each building in the building community being generated by:

The unit combination module is used for selecting a plurality of building units from a preset building unit group to be combined into an initial building structure; wherein the preset building unit group comprises at least two building units of a wall unit, a roof unit, a window unit and a door unit;

10. The apparatus of claim 6, wherein the biological mesh distribution module is specifically configured to:

11. The electronic equipment is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus;

A memory for storing a computer program;

A processor for implementing the method of any of claims 1-5 when executing a program stored on a memory.

12. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program which, when executed by a processor, implements the method of any of claims 1-5.