Methods, apparatuses, devices, media, and program products for generating three-dimensional representations

Abstract

Methods, apparatuses, devices, media, and program products for generating a three-dimensional representation are provided. The method includes receiving a first indication, the first indication being used to describe an object; generating an image based on the first indication, the image including a plurality of graphical elements, the plurality of graphical elements respectively representing the object; and generating a three-dimensional representation of the object based on the image, the three-dimensional representation including a plurality of planes arranged in a spatial relationship, the image being respectively mapped to the plurality of planes.

Description

Technical Field

[0001] The examples in this article generally relate to the field of computer science, and in particular to methods, apparatuses, devices, media, and program products for generating three-dimensional representations. Background Technology

[0002] In a 3D scene, the goal is to automatically generate a 3D representation of the object's spatial appearance based on the object's description information. Summary of the Invention

[0003] In a first aspect of this document, a method for generating a three-dimensional representation of an object is provided. The method includes: receiving a first instruction for describing the object; generating an image based on the first instruction, the image including a plurality of graphical elements, each graphical element representing the object; and generating a three-dimensional representation of the object based on the image, the three-dimensional representation including a plurality of planes arranged according to spatial relationships, the image being mapped to the plurality of planes respectively.

[0004] In a second aspect of this document, an apparatus for generating a three-dimensional representation of an object is provided. The apparatus includes: an indication receiving module configured to receive a first indication describing an object; an image generating module configured to generate an image based on the first indication, the image including a plurality of graphic elements, each graphic element representing the object; and a representation generating module configured to generate a three-dimensional representation of the object based on the image, the three-dimensional representation including a plurality of planes arranged according to spatial relationships, the image being mapped to the plurality of planes respectively.

[0005] In a third aspect of this document, an electronic device is provided. The electronic device includes at least one processor; and at least one memory coupled to the at least one processor and storing instructions for execution by the at least one processor, the instructions causing the electronic device to perform the method of the first aspect when executed by the at least one processor.

[0006] In a fourth aspect of this document, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program that can be executed by a processor to implement the method of the first aspect.

[0007] In a fifth aspect of this document, a computer program product is provided. This computer program product is tangibly stored in a computer storage medium and includes computer-executable instructions that, when executed by a device, cause the device to perform the method of the first aspect.

[0008] In this way, an image comprising multiple graphic elements representing the object can be generated based on a first indication used to describe the object, and the image can be mapped onto multiple planes arranged according to spatial relationships to form a three-dimensional representation of the object, thereby using multiple planes to carry the image content to present the spatial appearance of the object.

[0009] It should be understood that the content described in this section is not intended to limit the key or important features of the examples in this article, nor is it intended to restrict the scope of the solution. Other features will become readily apparent from the following description. Attached Figure Description

[0010] The above and other features, advantages, and aspects of the various examples herein will become more apparent when taken in conjunction with the accompanying drawings and the following detailed description. In the accompanying drawings, the same or similar reference numerals denote the same or similar elements, wherein: Figure 1 The diagram shows example environments for some scenarios; Figure 2A The flowcharts show example processes for generating 3D representations of objects in some scenarios; Figure 2B Example interfaces are shown for some scenarios where a three-dimensional representation of an object is generated by an intelligent system; Figure 3 Schematic diagrams of example 3D meshes and example 2D layouts for some scenarios are shown; Figure 4 A schematic diagram showing example images corresponding to different forms of an object is provided; Figure 5 A schematic diagram is shown showing the generation of an image based on a reference image and a layout reference image corresponding to a two-dimensional layout; Figure 6 The diagram illustrates examples of three-dimensional representations of objects in different scenarios; Figure 7 Flowcharts are shown for methods used to generate three-dimensional representations of objects in several scenarios; Figure 8 Block diagrams of apparatuses for generating three-dimensional representations of objects in several scenarios are shown; and Figure 9 Block diagrams are shown for some scenarios of electronic devices used to generate three-dimensional representations of objects. Detailed Implementation

[0011] The examples in this document will now be described in more detail with reference to the accompanying drawings. While some examples are shown in the drawings, it should be understood that solutions can be implemented in various forms and should not be construed as limited to the examples presented herein. Rather, these examples are provided to provide a more thorough and complete understanding of the solutions. It should be understood that the drawings and examples in this document are for illustrative purposes only and are not intended to limit the scope of protection of the solutions.

[0012] It should be noted that the headings of any section / subsection provided herein are not restrictive. Various examples are described throughout this document, and examples of any type may be included under any section / subsection. Furthermore, examples described in any section / subsection may be combined in any way with any other examples described in the same section / subsection and / or different sections / subsections.

[0013] In the description of the examples in this document, the term "including" and similar terms should be understood as open inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "an example" or "the example" should be understood as "at least one example". The term "some examples" should be understood as "at least some examples". Other explicit and implicit definitions may also be included below. The terms "first", "second", etc., may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0014] The examples in this document may involve user data, data acquisition, and / or use. All of these aspects comply with relevant laws, regulations, and rules. In the examples, all data collection, acquisition, processing, manipulation, forwarding, and use are conducted with the user's knowledge and confirmation. Accordingly, when implementing each example, the type, scope of use, and usage scenarios of any data or information that may be involved should be communicated to the user and their authorization obtained through appropriate means, in accordance with relevant laws and regulations. The specific methods of notification and / or authorization can vary depending on the actual situation and application scenario; the scope of the solution is not limited in this regard.

[0015] In this manual and the sample solutions, any processing of personal information will be conducted only under legal grounds (such as obtaining the consent of the data subject or being necessary for the performance of a contract) and will only be carried out within the scope stipulated or agreed upon. A user's refusal to process personal information beyond what is necessary for basic functions will not affect the user's use of basic functions.

[0016] In one context, an intelligent system is an information processing system. This intelligent system uses at least one or more models to process one or more inputs. The intelligent system may include the one or more models as components, or it may use models external to the system for processing. Those skilled in the art will recognize that there are various implementations of intelligent systems in the industry, including but not limited to agents, artificial intelligence (AI) assistants, AI workers, copilot systems, and chatbots. The inputs include data in one or more modalities. The modal data includes, but is not limited to, text, audio, images, video, application-specific data, or scientific computing data. The models include, but are not limited to, one or more combinations of the following: large-scale parametric models, deep learning models, neural network models, machine learning models, statistical models, rule-based models, or other models capable of processing the inputs. The specific implementation of the models can employ various existing or improved models. The intelligent system may involve one or more of the following operations: receiving single-round or multi-round input; recognizing and / or understanding the input; generating answers, decisions, or processing results using one or more models; invoking one or more external tools, interfaces, knowledge bases, or other systems to obtain information or perform operations; processing in a self-decision-making or human-machine collaborative manner, or transferring tasks to human or other systems for processing. Self-decision-making refers to execution without human intervention. Human-machine collaboration refers to partial human participation or confirmation. The above explanation of "intelligent system" is illustrative, and each description is an example, not necessarily necessary, and does not constitute an exhaustive list.

[0017] As mentioned above, in a 3D scene, several image planes with transparency information can be combined to represent objects such as flowers, grass, and shrubs. Multiple planes only need to carry 2D image content to provide multi-directional visual hierarchy. The number of geometric faces is usually significantly less than the number of geometric faces required to directly construct a complete 3D surface. Furthermore, each plane can also present dynamic effects such as swinging or cooperating with scene elements through rendering parameters.

[0018] Related methods can directly generate 3D models of objects with complete solid surfaces. However, the geometric topology and texture coordinate (UV) layout of such outputs may vary with the generated results, making it difficult to stably match the pre-set planar combination structure. Complete solid surfaces may also contain many geometric faces, increasing the rendering overhead when placing a large number of repetitive objects in the scene, and making it difficult to quickly obtain another appearance by replacing images.

[0019] When the generated image is not constrained according to a pre-made two-dimensional layout (2D layout), graphic elements may cross predetermined partitions or deviate from their corresponding positions. Mapping such images to multiple planes may result in stretching, misalignment, incomplete transparent edges, or interference between contents of adjacent partitions. If image generation and 3D assembly are performed separately in different processing stages, additional manual work may be required to correct the correspondences.

[0020] A scheme for generating a three-dimensional representation of an object is proposed. According to the scheme, an electronic device receives a first instruction for describing the object, generates an image based on the first instruction, the image including multiple graphic elements that respectively represent the object, and then generates a three-dimensional representation of the object based on the image. The three-dimensional representation includes multiple planes arranged according to spatial relationships, and the image is mapped to the multiple planes respectively.

[0021] The above scheme converts the initial description of an object into an image comprising multiple graphic elements, which are then mapped onto multiple planes arranged according to spatial relationships, thereby forming a three-dimensional representation capable of presenting the object's appearance from multiple directions. Each graphic element represents the object, and the image content, along with the spatial relationships of the multiple planes, collectively forms the object's spatial appearance, enabling the three-dimensional representation of the object to utilize multiple planes to carry the image content.

[0022] The following describes various examples of this scheme in further detail with reference to the accompanying drawings.

[0023] Figure 1 A schematic diagram of example environment 100 is shown. (e.g.) Figure 1 As shown, example environment 100 may include electronic device 110.

[0024] In example environment 100, electronic device 110 may run application 120 that supports generating 3D representations of objects. Application 120 may be a 3D content generation application, scene editing application, resource processing application, or other suitable application. User 140 may interact with application 120 via electronic device 110 and / or attached devices of electronic device 110.

[0025] exist Figure 1 In environment 100, if application 120 is active, electronic device 110 can use application 120 to present interface 150 for receiving first instructions, displaying candidate results, or presenting a three-dimensional representation of an object.

[0026] In some cases, electronic device 110 communicates with server 130 to provide services to application 120. Electronic device 110 can be any type of mobile terminal, fixed terminal, or portable terminal, including mobile phones, desktop computers, laptop computers, notebook computers, netbook computers, tablet computers, media computers, multimedia tablets, handheld computers, portable gaming terminals, virtual reality (VR) / augmented reality (AR) devices, personal communication system (PCS) devices, personal navigation devices, personal digital assistants (PDAs), audio / video players, digital cameras / camcorders, positioning devices, television receivers, radio receivers, e-book devices, gaming devices, or any combination of the foregoing, including accessories and peripherals of these devices or any combination thereof. In some cases, electronic device 110 can also support any type of user-facing interface (such as "wearable" circuitry).

[0027] Server 130 can be a single physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing 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, and artificial intelligence platforms. Server 130 may include, for example, computing systems / servers such as mainframes, edge computing nodes, computing devices in cloud environments, etc. Server 130 can provide backend services for application 120 in electronic device 110 that supports generating three-dimensional representations of objects.

[0028] A communication connection can be established between server 130 and electronic device 110. This communication connection can be established via wired or wireless means. The communication connection can include, but is not limited to, Bluetooth, mobile network, Universal Serial Bus (USB), and Wireless Fidelity (WiFi) connections. In some cases, server 130 and electronic device 110 can exchange signaling information through their communication connection.

[0029] It should be understood that the structure and function of the various elements in environment 100 are described for illustrative purposes only and do not imply any limitation on the scope of the scheme.

[0030] The examples described herein can be implemented on electronic device 110. It should be noted that the operations described relative to electronic device 110 can be performed by a related application 120 running on electronic device 110. The interface described relative to electronic device 110 can be provided by application 120; for example, electronic device 110 presents the corresponding interface by running application 120. Furthermore, some operations described relative to electronic device 110 may require the assistance of server 130 to complete.

[0031] In some cases, electronic device 110 can independently perform the process of generating a 3D representation of an object. Application 120 can locally generate an image based on a first instruction, then invoke a local assembly tool to generate a 3D representation of the object based on the image and present it via interface 150. In some cases, application 120 can also locally obtain mapping relationships and corresponding spatial constraint information. A mapping relationship refers to the mapping or association between a first category of an object and spatial constraint information. A first category can be mapped to one or more candidate spatial constraint information.

[0032] Spatial constraint information refers to information used to indicate structural constraints on the three-dimensional representation of an object. In some cases, spatial constraint information can include two-dimensional layout and three-dimensional structural information. Two-dimensional layout is used to indicate the arrangement of multiple graphic elements in a two-dimensional plane (i.e., an image). Two-dimensional layout is sometimes also referred to as UV layout, partition boundaries, layout mask, or layout reference image. Three-dimensional structural information is used to indicate the spatial relationships between multiple planes, thereby constraining the three-dimensional structure. In some cases, three-dimensional structural information can, for example, include the form of a three-dimensional mesh to indicate the three-dimensional spatial relationships of multiple planes. Specifically, three-dimensional structural information can indicate at least one of the angles, scaling factors, or spatial positions of the multiple planes.

[0033] An image generation tool refers to a software component, model, or service capable of generating an image based on a first instruction. In some cases, the image generation tool can further generate an image based on a two-dimensional layout. An image refers to two-dimensional data generated according to the first instruction, and may include multiple graphic elements that respectively represent an object. The image may simultaneously include a first channel for indicating color information and a second channel for indicating transparency information, and may be mapped as a whole to multiple planes arranged according to spatial relationships.

[0034] A 3D representation of an object refers to a set of data or renderable resources that can characterize the appearance and spatial structure of an object in three-dimensional space. This may include at least one of the following: multiple planes arranged according to spatial relationships, images mapped to the multiple planes respectively, transparency information, material information, and rendering parameters. In some cases, a 3D mesh can be used to carry multiple planes and their 3D structural information. When the object is a plant, its 3D representation can be called a billboard / cross-quad plant or a billboard plant resource. An assembly tool refers to a software component or service that can map a single image to multiple planes, arrange the multiple planes according to spatial relationships, and form a 3D representation of the object.

[0035] In some cases, server 130 can perform the process of generating a three-dimensional representation of an object. Electronic device 110 can obtain a first instruction via interface 150 and send the first instruction to server 130. Server 130 can generate an image based on the first instruction, generate a three-dimensional representation of the object based on the image, and then send the three-dimensional representation of the object or its access location to electronic device 110. In some cases, server 130 can also obtain spatial constraint information and generate a three-dimensional representation of the object based on the spatial constraint information.

[0036] In some scenarios, electronic device 110 and server 130 can jointly execute the process of generating a three-dimensional representation of an object. For example, electronic device 110 can receive a first instruction and present the result, while server 130 can complete category determination, spatial constraint information selection, and image generation. Electronic device 110 can then use the image returned by server 130 and its local three-dimensional structural information to complete the assembly. As another example, electronic device 110 can select a two-dimensional layout and three-dimensional structural information, and server 130 can generate an image based on the two-dimensional layout and return the image.

[0037] In some cases, electronic device 110 can generate a three-dimensional representation of an object without relying on a mapping between object category and spatial constraint information. For example, electronic device 110 can receive a first instruction to describe the object, invoke an image generation tool to directly generate an image based on the first instruction, and invoke an assembly tool to generate a three-dimensional representation of the object based on the image. The image may include multiple graphic elements that represent the object, and the three-dimensional representation of the object may include multiple planes arranged according to spatial relationships, with the image mapped to each of the multiple planes.

[0038] In one example, the first instruction may include at least one of text describing the object to be generated and a reference image. The image generation tool can determine the object's appearance features based on the first instruction and generate multiple graphic elements in the same image. These multiple graphic elements can respectively represent the object's appearance from different viewing directions, the parts of the object to be presented on different planes, or multiple visible components of the object. Each graphic element can represent all or part of the object's appearance, and multiple graphic elements work together to form the object's spatial appearance.

[0039] In one example, the assembly tool can generate a 3D representation using a pre-defined general plane combination method. Upon receiving an image, the tool can create multiple planes, map the image as a texture resource onto each plane, and set the planes according to pre-defined spatial relationships. For example, multiple planes can be combined in an intersecting plane manner or set to face the viewing direction. This process can be completed during the generation of the 3D representation without requiring obtaining a specific 3D mesh or 2D layout as a separate step before image generation.

[0040] In another example, application 120 can provide one or more plane combination options and receive selection operations for these options. The assembly tool can determine the number of planes and their respective angles, scaling factors, or spatial positions based on the selection operations, and generate a 3D representation based on the image. The selection operations can be received before or after image generation, and the image can still be generated directly based on the first indication without automatically selecting spatial constraint information according to object category.

[0041] When mapping an image to multiple planes, different texture reading methods can be used. In some cases, each plane can read the entire area of ​​the image, allowing the same image to be shared by multiple planes. In other cases, each plane can read the corresponding area of ​​the image using its own texture coordinates. This corresponding area can be determined according to pre-set texture coordinates or, in the case of using a 2D layout, according to the 2D layout itself. Regardless of the method used, multiple graphic elements remain within the same image, eliminating the need to generate multiple separate image files for mapping.

[0042] The above generation method does not require obtaining the first category after receiving the first instruction, nor does it require obtaining spatial constraint information through mapping relationships. The first category, mapping relationships, two-dimensional layout, and three-dimensional structure information can be used as optional refinement information to further constrain the image content or the arrangement of multiple planes. By setting the above information as optional information, the electronic device 110 can still generate an image based solely on the first instruction and form a three-dimensional representation of the image content carried by multiple planes based on that image, thereby completing the assembly of the object's appearance and planar combination structure even in the absence of a specific category mapping.

[0043] The following description of the example will continue with reference to the accompanying drawings.

[0044] Figure 2A A flowchart of an example process 200A for generating a 3D representation of an object based on certain conditions is shown. Process 200A can be implemented at electronic device 110. (See below for reference.) Figure 1 and Figure 2A To describe process 200A.

[0045] At point 201, electronic device 110 receives a first instruction describing an object. The first instruction refers to input information characterizing the object to be generated and its appearance requirements, and may include text, a reference image, or a combination of both. It may also carry at least one of the following: category, color, shape, visual style, density, scene atmosphere, and generation parameters. In this document, an object refers to an entity or entity type that can be represented in three dimensions by combining multiple planes with an image, and may include plants and other objects suitable for presentation using a planar combination structure. A plane refers to a patch or set of patches used to carry image content. Multiple planes may be combined in a cross-quad, a billboard orientation, or other suitable manner. The first instruction may include at least one of text or a reference image. For example, text may describe the category, color, shape, visual style, seasonal characteristics, density, or scene atmosphere of the plant to be generated, and the reference image may characterize the desired object appearance.

[0046] In some cases, application 120 can extract at least one of category information, color information, shape information, and generation parameters from the first instruction to obtain structured instruction information. Structured instruction information refers to information extracted from the first instruction and organized according to predetermined fields, and may include at least one of category information, color information, shape information, and generation parameters. Generation parameters may include style selection, number of candidates, or sampling method. When the fields in the first instruction are insufficient or their meaning is ambiguous, application 120 can present a supplementary request through interface 150 and update the structured instruction information based on the supplementary input.

[0047] At point 202, electronic device 110 obtains a first category based on a first instruction, the first category corresponding to the shape of the object. The first category refers to the category corresponding to the shape of the object. When the object is a plant, the first category may include tall-flowered, medium-flowered, dwarf-flowered, grass, shrub, or other categories divided according to shape and arrangement of segments. When the object is a plant, electronic device 110 can classify the specific plant indicated by the first instruction into tall-flowered, medium-flowered, dwarf-flowered, grass, shrub, or other appropriate categories, so that user 140 does not need to directly specify the spatial relationship of multiple planes.

[0048] Additionally, in some cases, the electronic device 110 can obtain spatial constraint information based on a first category and a mapping relationship. The spatial constraint information indicates structural constraints on the three-dimensional representation of an object, and the mapping relationship includes the mapping between the first category and the spatial constraint information. In some cases, the spatial constraint information may include two-dimensional layout and three-dimensional structural information; the two-dimensional layout indicates the arrangement of multiple graphic elements in the image, and the three-dimensional structural information indicates the spatial relationships between multiple planes. A graphic element refers to a visual content unit in an image used to represent an object. When the object is a plant, the graphic elements may include image content of flowers, stems, leaves, branches, or combinations thereof. Multiple graphic elements can collectively constitute an image without needing to be separated from the image and then mapped. In one example, the three-dimensional structural information can be represented by a three-dimensional mesh (3D Mesh) data structure, which may include multiple planes and their respective geometric data and spatial transformation data. In the example of a patch object, this three-dimensional mesh may also be referred to as a patch mesh. In another example, the three-dimensional structural information can be represented by a data structure containing spatial structural parameters for each plane, such as at least one of the plane's normal vector, center position, rotation angle, and relative offset between the planes.

[0049] In some cases, the mapping relationship may include a mapping between a first category and multiple candidate spatial constraint information. The electronic device 110 can obtain multiple candidate spatial constraint information based on the first category and the mapping relationship. Each candidate spatial constraint information may include mutually matching candidate two-dimensional layouts and candidate three-dimensional structural information. A candidate three-dimensional structural information can be associated with one or more candidate two-dimensional layouts, thereby providing different combinations of planar arrangements and image layouts for the same object shape.

[0050] Electronic device 110 can receive a selection operation, which is used to select spatial constraint information from multiple candidate spatial constraint information. The processing entity can also select at least one candidate spatial constraint information according to a sampling strategy, or generate three-dimensional representations of objects for all or part of the candidate spatial constraint information, and provide multiple results for user 140 to choose from via interface 150. The processing entity refers to a computer program or computing device capable of invoking at least one software tool and organizing processing steps based on input, and may include a process orchestration program, task execution program, electronic device 110, server 130, or a combination thereof. When a first instruction involves multiple object categories, the processing entity can also obtain corresponding spatial constraint information for each object category separately.

[0051] At point 203, electronic device 110 generates an image based on a first instruction and a two-dimensional layout. The image includes multiple graphic elements, each representing an object. For example, a plant image may include multiple graphic elements such as flowers, stems, and leaves, all of which are collectively represented in the plant image. In some cases, the image may include multiple partitions, and the multiple graphic elements may be distributed across these partitions according to a two-dimensional layout.

[0052] In some cases, electronic device 110 can obtain visual information based on a first instruction. Visual information refers to information used to define the visual presentation of an image, and may indicate at least one of the following: plant category, flower color, seasonal characteristics, visual style, density, or scene atmosphere. In some cases, the visual information may indicate the visual style of an image, and an image may be generated based on that visual style. In some cases, electronic device 110 may also generate an image in conjunction with a two-dimensional layout. Thus, the same object shape can correspond to images of different categories, colors, seasonal characteristics, or visual styles.

[0053] In some cases, spatial constraint information includes two-dimensional layout and three-dimensional structural information. Electronic device 110 can obtain layout constraint information based on the two-dimensional layout to constrain the position, extent, or orientation of multiple graphic elements in an image, and generate an image based on a first indication and this layout constraint information. In some cases, the layout constraint information can be represented by a layout mask, a block guide map, or a layout reference image corresponding to the two-dimensional layout. A layout reference image is an image that visually represents partitions, boundaries, or drawable areas in a two-dimensional layout. The layout reference image can be associated with the two-dimensional layout through a mapping relationship and provided as layout constraint information to the image generation tool.

[0054] A 2D layout can indicate multiple partitions in an image. A partition is a region in a 2D layout that corresponds to a plane or set of planes in a 3D mesh. Multiple partitions can have the same or different shapes and sizes. In some cases, a 2D layout can indicate the correspondence between multiple graphic elements and multiple partitions. Spacing regions can be set between multiple partitions to maintain partition boundaries during image sampling and subsequent texture processing. Spacing regions are areas between adjacent partitions that are not used to host target graphic elements; these regions serve as safety margins for partition boundaries to reduce mutual interference between adjacent partition contents during image sampling.

[0055] The processing entity can invoke an image generation tool to generate an image based on a first instruction. In some cases, the image generation tool can also generate an image based on a two-dimensional layout or corresponding layout constraint information. The image generation tool can generate graphic elements matching the object's shape within each partition and output color and transparency information. Color information can be indicated by a first channel, and transparency information can be indicated by a second channel.

[0056] In some cases, the electronic device 110 can generate multiple candidate images based on a first instruction. A candidate image refers to one or more images to be selected generated for the same first instruction. Different candidate images may correspond to different sampling results or different values ​​of generation parameters. In some cases, each candidate image includes multiple candidate graphic elements that respectively represent an object. Different candidate images can be obtained through multiple samplings or through different values ​​of generation parameters. In some cases, the electronic device 110 can also generate multiple candidate images by combining a two-dimensional layout.

[0057] At position 204, the electronic device 110 obtains quality information and determines, based on the quality information, whether the image quality of each of the multiple candidate images meets the quality requirements. Quality information refers to information characterizing whether a candidate image is suitable for mapping to multiple planes, and may indicate at least one of the following: the alignment of multiple candidate graphic elements with multiple partitions in the image, the edge integrity of the multiple candidate graphic elements, and the style consistency among the multiple candidate graphic elements. The electronic device 110 can select an image from the multiple candidate images that meet the quality requirements based on the quality information.

[0058] In response to the first candidate image's image quality failing to meet quality requirements, the electronic device 110 can adjust a first value of a first parameter to obtain a second value, and regenerate the first candidate image based on the second value of the first parameter. The first parameter refers to a parameter that can affect the image generation result, and may include sampling parameters, compositional constraint parameters, visual style parameters, or other appropriate parameters. The first parameter can be adjusted from a first value to a second value to regenerate the candidate image that does not meet the quality requirements. Figure 2A The path from 204 back to 203 represents this regeneration process. Electronic device 110 can repeatedly generate and determine the quality until an image that meets the quality requirements or a preset stopping condition is met.

[0059] At position 205, the processing entity can invoke the assembly tool to map the selected image onto multiple planes based on the image and 3D structural information, and set up the multiple planes according to the spatial relationships indicated by the 3D structural information. The mapping unit is the entire image including multiple graphic elements. Each plane can read the corresponding partition in the entire image through its own texture coordinates without separating the graphic elements from the image.

[0060] The spatial relationship between multiple planes can indicate at least one of the angles, scaling factors, or spatial positions of each plane. The assembly tool can read the spatial transformation data from the pre-made 3D mesh, combine multiple planes according to the spatial relationship, and bind color information, transparency information, and appropriate material parameters to the multiple planes.

[0061] At point 206, electronic device 110 obtains a three-dimensional representation of the object. The three-dimensional representation includes multiple planes arranged according to spatial relationships, with the image mapped to each plane. In some cases, electronic device 110 can generate a three-dimensional representation of the object based on the image and three-dimensional structural information. The object can be a plant, and the generated three-dimensional representation can be an intercalation plant resource usable in a rendering engine. Electronic device 110 can also save the association data between the three-dimensional representation, three-dimensional structural information, two-dimensional layout, and image to support subsequent reuse.

[0062] In some cases, electronic device 110 can receive a second instruction describing another object, generate a second image based on the second instruction, and generate a three-dimensional representation of the other object based on the second image. In other cases, electronic device 110 can generate a second image based on the second instruction and a used two-dimensional layout, and generate a three-dimensional representation of the other object based on the same three-dimensional structural information and the second image. Thus, spatial constraint information can be reused among multiple objects, and a three-dimensional representation with a different appearance can be obtained simply by updating the image.

[0063] In the above manner, process 200A generates an image comprising multiple graphic elements representing the object based on a first instruction used to describe the object, and maps the image onto multiple planes arranged according to spatial relationships, thereby forming a three-dimensional representation of the object. When spatial constraint information is used, the mapping relationship between the object shape and the spatial constraint information can be integrated through category determination, constrained image generation, quality determination, and three-dimensional assembly. Candidate selection and regeneration paths improve output stability, and the reuse of spatial constraint information reduces the resources required for repeated geometry construction, image adjustment, and storage, thereby improving the processing efficiency of the electronic device 110 or server 130 in batch generating three-dimensional representations.

[0064] Figure 2B An example interface 200B is shown, illustrating a 3D representation of an object generated by an intelligent system based on certain scenarios. Example interface 200B can be a specific implementation of interface 150 and may include a dialogue area and a results display area.

[0065] exist Figure 2B In the example scenario shown, user 140 can send a first instruction 210 to the intelligent system through a dialogue area. The first instruction 210 describes the object to be represented in three dimensions. The first instruction may include at least one of text 210 or a reference image 211. For example, user 140 may input text 210 describing the object's category, shape, color, visual style, or other appearance requirements, and may attach a reference image 211 to further indicate the desired object appearance.

[0066] In response to receiving the first instruction, the intelligent system can execute the reference Figure 2AThe process described is 200A. For example, the intelligent system can generate an image comprising multiple graphic elements representing an object based on a first instruction, and generate a three-dimensional representation of the object based on the image. This three-dimensional representation includes multiple planes arranged according to spatial relationships, with the image mapped to each of the multiple planes. In some cases, the intelligent system can also obtain a first category based on the first instruction and obtain spatial constraint information based on the first category and the mapping relationship to generate the three-dimensional representation of the object. The intelligent system can present the generated three-dimensional representation of the object to the user 140 in the result display area 212. In some cases, the user 140 can also view different candidate results, select a candidate result, or re-initiate the generation request through the example interface 200B.

[0067] It should be understood that Figure 2B The example of sending the first instruction via dialogue is merely one exemplary interaction method for receiving a generation request, and is not a necessary interaction method for generating a 3D representation of an object. In other cases, a dialogue area or session history may not be presented. For example, application 120 may present a non-dialogue-based generation interface provided by an image generation engine. User 140 can select an object to be represented in 3D using object selection controls, input controls, menus, or other interactive elements. Application 120 can then generate a first instruction based on user 140's selection and submit a generation request to the intelligent system. The intelligent system can generate a 3D representation of the object based on this first instruction and display the generated result to user 140 through this non-dialogue-based generation interface. Therefore, this document does not limit the specific interface form used for receiving the first instruction and presenting the 3D representation of the object.

[0068] Figure 3 Example 300 is shown, illustrating spatial constraint information based on certain scenarios. For example... Figure 3 As shown, the spatial constraint information in Example 300 includes three-dimensional structural information represented by a three-dimensional mesh 301 and a two-dimensional layout 302. The "planar layout" in the attached figure can be considered as a representation of the two-dimensional layout 302.

[0069] The 3D mesh 301 may include multiple planes and represent the spatial relationships between them. These planes can be arranged in a pre-defined interlacing pattern, allowing image content viewed from different directions to collectively form the spatial appearance of the object. The 3D mesh 301 can store vertex data, texture coordinates, and at least one of the following: angle, scaling factor, and spatial position for each plane.

[0070] In some examples, when the object is a plant, the 3D mesh 301 can be an interpolated mesh that matches the plant's morphology. For example, multiple planes of tallflowers can extend vertically and form a more concentrated layout in the top region, multiple planes of grasses can form a lower and denser layout, and multiple planes of shrubs can form a multi-layered, intersecting layout. The above morphologies are merely open examples.

[0071] The two-dimensional layout 302 can indicate the layout of multiple graphic elements in an image. In some examples, the two-dimensional layout 302 can be a layout obtained by unfolding the texture coordinates corresponding to multiple planes of the three-dimensional mesh 301 onto a single plane. Each partition in the two-dimensional layout 302 corresponds to a plane or a set of planes in the three-dimensional mesh 301, so that the content located in the corresponding partition in the image can be read by the corresponding plane.

[0072] Multiple partitions can be set as rectangles or other suitable shapes, and the number, size, and arrangement of partitions can be determined based on the number of planes in the 3D mesh 301 and the expected composition. Spacing regions can be set between adjacent partitions, which can reduce the color interference between adjacent partitions at different image resolutions or texture levels.

[0073] Electronic device 110 can generate layout constraint information based on two-dimensional layout 302. For example, electronic device 110 can form a layout mask or block guide map from partition boundaries, or obtain a layout reference image corresponding to two-dimensional layout 302. Based on this, the image generation tool generates graphic elements within the drawable areas defined by each partition.

[0074] The first category and spatial constraint information can be associated through a mapping relationship. A first category can be mapped to multiple candidate spatial constraint information, and each candidate spatial constraint information can include mutually matching candidate 3D structural information and candidate 2D layout 302. The 3D structural information can be represented by a 3D mesh 301. A candidate 3D structural information can be associated with one or more candidate 2D layouts 302. When adding a new object shape, the corresponding spatial constraint information can be added to the mapping relationship without changing the overall processing order of image generation and assembly.

[0075] When mapping an image onto multiple planes, the assembly tool can provide the entire image as a texture resource to each of the planes. In cases where the three-dimensional structural information is represented by a three-dimensional mesh 301, each plane can select the corresponding partition of the image using the corresponding texture coordinates in the two-dimensional layout 302. The multiple planes are then set according to the spatial relationship indicated by the three-dimensional mesh 301 to obtain a three-dimensional representation of the object.

[0076] In this way, the spatial constraint information in Example 300 pre-associates the three-dimensional structural information of multiple planes with the two-dimensional layout in the image, which can alleviate the need for repeated manual alignment between image content and planar structure. The three-dimensional structural information represented by the three-dimensional mesh 301 and the two-dimensional layout 302 can be reused in multiple generations. The electronic device 110 only needs to update the image data to form another appearance, thereby reducing the amount of processing required for geometric calculations and texture coordinate adjustments, and improving structural consistency.

[0077] Figure 4 Example images 400 are shown, corresponding to different forms of the object. For example... Figure 4 As shown, example image 400 includes tall-flowered image 401, grass-like image 402, and medium-flowered image 403.

[0078] Image 401 of tall-flowered plants can form content of stems, leaves, and flowers located at higher positions within a section corresponding to the vertical plane. Image 402 of grasses can form content of slender and relatively dense leaves within multiple sections. Image 403 of medium-flowered plants can form content of flowers and leaves with a height and spread between that of tall-flowered and dwarf-flowered plants within each section.

[0079] It should be understood that tall flowers, grasses, and medium-sized flowers are merely examples of object categories and are not restrictive. In some cases, dwarf flowers, shrubs, or other object categories may also be included. Each of the aforementioned object categories can be associated with candidate spatial constraint information that matches its planar arrangement. Candidate spatial constraint information may include candidate 3D structural information and candidate 2D layout.

[0080] Electronic device 110 can obtain a first category based on a first indication. For example, when the first indication points to an object with a tall stem and concentrated flowers, such as a sunflower, electronic device 110 can determine the first category as a tall-flowered species. As another example, when the first indication points to an object with slender leaves and a low-lying, dense distribution, electronic device 110 can determine the first category as a grass. The first category is used to select spatial constraint information that matches the morphology, and is not used to limit the specific object category in the image. In some cases, the spatial constraint information may include a three-dimensional grid and a two-dimensional layout that match the first category.

[0081] For the same first category, electronic device 110 can obtain multiple candidate spatial constraint information and receive a selection operation for selecting spatial constraint information. Different candidate spatial constraint information can have different numbers of partitions, partition ratios, and angles, scaling ratios, or spatial positions of multiple planes, thereby forming different insert appearances.

[0082] The electronic device 110 can also obtain visual information based on the first indication and generate an image based on the visual style indicated by the visual information. In some cases, the electronic device 110 can also generate an image under two-dimensional layout constraints corresponding to the first category. The visual information can make the high-flowering image 401, the grass image 402, or the medium-flowering image 403 present different colors, seasonal characteristics, visual styles, and densities, while keeping the graphic elements in the corresponding partitions.

[0083] In this way, example image 400 selects reusable spatial constraint information in the form of an object, and changes the specific image content within the structural template included in the spatial constraint information. This alleviates the need to redesign the three-dimensional structure and two-dimensional layout for each plant category. Electronic device 110 can reuse limited spatial constraint information to generate multiple appearances, reducing the repeated storage and selection range of structural resources, and improving the throughput of batch generation of different plant resources.

[0084] Figure 5 An example of image generation 500 is shown. For example... Figure 5 As shown, image generation example 500 includes a style reference image 501, a layout reference image 502, and an image 503. In some cases, the style reference image 501 may be included in a first indication, the layout reference image 502 may correspond to a two-dimensional layout obtained through mapping relationships, and the image 503 may be mapped to multiple planes arranged according to spatial relationships.

[0085] The style reference image 501 can be provided by the user 140 via the interface 150 to indicate the appearance of the object to be generated. The first instruction may also include only text, or both text and the style reference image 501. The electronic device 110 can extract at least one of category information, color information, shape information, and generation parameters from the first instruction to obtain structured instruction information.

[0086] The layout reference image 502 can convert multiple partitions, partition boundaries, and spacing regions in a two-dimensional layout into layout constraint information that can be processed by an image generation tool. The layout reference image 502 can be generated by the electronic device 110 based on the two-dimensional layout, or it can be stored together with the two-dimensional layout as part of the mapping relationship.

[0087] The processing entity can invoke the image generation tool and the assembly tool respectively. The image generation tool can receive at least one of structured instruction information, style reference image 501, and layout reference image 502. In some cases, the processing entity can determine the visual style of the image content based on the style reference image 501 and generate image 503 within the partition defined by the layout reference image 502.

[0088] Image 503 may include multiple graphic elements, each representing an object. When the object is a plant, the graphic elements in different sections can form flower, stem, or leaf content whose orientation, length, and shape match the corresponding plane. The correspondence between multiple graphic elements and multiple sections can be indicated by two-dimensional layout or layout constraint information.

[0089] Image 503 can include color channels and alpha channels. Color channels can carry red, green, and blue (RGB) color information, while alpha channels can carry transparency (alpha) information. Transparency information ensures that image areas outside of graphic elements do not obscure scene content during rendering.

[0090] The image generation tool can generate multiple candidate images 503 through multiple samplings. The electronic device 110 can obtain the quality information of each candidate image, determine whether the candidate graphic elements are located in the corresponding partition, whether the transparent edges are complete, and whether the visual styles of the multiple candidate graphic elements are consistent, and select the image 503 for assembly from the candidate images that meet the quality requirements.

[0091] In some cases, if the first candidate image does not meet the quality requirements, the electronic device 110 can adjust the first parameter used to generate the first candidate image and regenerate the first candidate image based on the adjusted parameter. The adjustment can involve the layout constraint strength, sampling method, visual style, or other appropriate parameters without changing the correspondence between the two-dimensional layout and the three-dimensional structural information.

[0092] The assembly tool can receive image 503 and three-dimensional structural information corresponding to layout reference image 502. This three-dimensional structural information can be represented by a three-dimensional mesh. Based on image 503 and the three-dimensional structural information, the assembly tool can map the entire image 503 onto multiple planes. Each plane reads the corresponding partition of image 503 according to its texture coordinates and is set according to the spatial relationships indicated by the three-dimensional structural information to jointly present the appearance of the object.

[0093] In this way, image generation example 500 provides appearance instructions and layout constraints to the image generation tool, enabling the generated content to adapt to the predetermined partitions while maintaining the desired visual style. This mitigates misalignment and edge anomalies caused by inconsistencies between the generated image and the texture coordinate layout. Quality determination and regeneration improve the stability of usable images, reduce image rework and repeated tool calls in subsequent assembly, thereby shortening the overall generation time of the 3D representation.

[0094] Figure 6 An application example 600 of a three-dimensional representation of an object is shown. For example... Figure 6As shown, application example 600 includes a 3D representation 601 of an object within a rendering engine, a 3D representation 602 of a single object presented to user 140, and an example scene 603 obtained by reusing the 3D representation of the object.

[0095] When generating a 3D representation 601 of an object, the assembly tool can map the image onto multiple planes based on the image and 3D structural information, and set up multiple planes according to the spatial relationships indicated by the 3D structural information, thereby obtaining the 3D representation 601 of the object. In the case where the 3D structural information is represented by a 3D mesh, each plane can read the corresponding partition of the image through texture coordinates in the 2D layout.

[0096] When an image includes a first channel and a second channel, the assembly tool can generate a 3D representation 601 of the object based on the color information indicated by the first channel and the transparency information indicated by the second channel. In some cases, the assembly tool can also combine 3D structural information represented by a 3D mesh to generate this 3D representation. The assembly tool can also bind at least one of the following parameters: material, transparency clipping, two-sided rendering, normal processing, or wobbling parameters, so that multiple planes are rendered in the rendering engine in the desired manner.

[0097] A 3D representation 602 of a single object can be loaded, edited, or rendered as a reusable resource by application 120. When the object is a plant, the 3D representation 602 can present the spatial appearance of intercalary plants such as flowers, grass, or shrubs, and can generate dynamic effects that change with the scene based on sway parameters. This document does not restrict the subsequent use of the 3D representation.

[0098] Electronic device 110 can place the 3D representation of the same object multiple times in example scene 603, and can also reuse the same spatial constraint information and replace different images to form multiple objects with different appearances. In some cases, the spatial constraint information may include 3D meshes and 2D layouts. Since duplicate objects can share at least part of the spatial constraint information or material configuration, application 120 can reduce the storage and loading of duplicate resources.

[0099] In some scenarios, electronic device 110 can receive a second instruction describing another object, generate a second image based on the second instruction, and generate a three-dimensional representation of the other object based on the second image. In some scenarios, electronic device 110 can reuse the same spatial constraint information to generate a three-dimensional representation of another object. Electronic device 110 can save the three-dimensional representations of multiple objects to a resource collection and load or replace them in batches as needed for the scenario.

[0100] The 3D representation of objects can be used in scene editing, real-time rendering, virtual environment construction, digital content creation, or other applications that require presenting multiple objects with a limited number of geometric faces. In some cases, the above approach can also be applied to planar composite objects other than plants. It should be understood that the above application scenarios are merely illustrative and not restrictive.

[0101] Using the methods described above, application example 600 maps images onto multiple planes set according to spatial relationships, and can assemble color information, transparency information, and rendering parameters together, allowing objects to maintain their structural correspondence with the image in both single-object rendering and repetitive scenes. When using spatial constraint information, images generated under 2D layout constraints and reusable 3D structural information can mitigate stretching, misalignment, and transparent edge anomalies. Sharing spatial constraint information and material configurations reduces the storage and loading of repetitive geometric data and lowers the geometric processing and rendering overhead when a large number of objects are rendered simultaneously, thereby supporting more efficient scene construction and real-time rendering.

[0102] Figure 7 A flowchart of an example process 700 for generating a three-dimensional representation of an object is shown, depending on certain circumstances. Process 700 can be implemented at electronic device 110. (See below for reference.) Figure 7 To describe process 700.

[0103] like Figure 7 As shown, in block 710, electronic device 110 receives a first instruction, which describes an object.

[0104] In frame 720, electronic device 110 generates an image based on a first instruction. The image includes multiple graphic elements, each of which represents an object.

[0105] In box 730, electronic device 110 generates a three-dimensional representation of an object based on an image. The three-dimensional representation includes multiple planes arranged according to spatial relationships, and the image is mapped to each of the multiple planes.

[0106] In some cases, the electronic device 110 may obtain a first category based on a first indication, the first category corresponding to the shape of an object; and obtain spatial constraint information based on the first category and a mapping relationship, the spatial constraint information indicating structural constraints on the three-dimensional representation of the object, the mapping relationship including the mapping between the first category and the spatial constraint information, wherein the three-dimensional representation of the object is generated based on the spatial constraint information.

[0107] In some cases, spatial constraint information may include two-dimensional layout and three-dimensional structural information. The two-dimensional layout indicates the arrangement of multiple graphic elements in the image, while the three-dimensional structural information indicates the spatial relationship between multiple planes.

[0108] In some cases, electronic device 110 can generate an image based on a first instruction and a two-dimensional layout, the image including multiple partitions, with multiple graphic elements distributed in the multiple partitions according to the two-dimensional layout.

[0109] In some cases, electronic device 110 can generate a three-dimensional representation of an object based on images and three-dimensional structural information.

[0110] In some cases, the mapping relationship may include a mapping between a first category and multiple candidate spatial constraint information. Electronic device 110 may obtain multiple candidate spatial constraint information based on the first category and the mapping relationship; and receive a selection operation for selecting spatial constraint information from the multiple candidate spatial constraint information.

[0111] In some cases, electronic device 110 may obtain visual information based on a first instruction, the visual information indicating the visual style of an image; and generate an image based on the visual style.

[0112] In some cases, the electronic device 110 may generate multiple candidate images based on a first instruction, each candidate image including multiple candidate graphic elements, each candidate graphic element representing an object; obtain quality information indicating the image quality of each of the multiple candidate images; and select an image from the multiple candidate images based on the quality information.

[0113] In some cases, the electronic device 110 may, in response to the image quality of the first candidate image not meeting the quality requirements, adjust a first value of the first parameter to obtain a second value of the first parameter, the first value being used to generate a first candidate image, a plurality of candidate images including the first candidate image; and regenerate the first candidate image based on the second value of the first parameter.

[0114] In some cases, the image quality of multiple candidate images can indicate at least one of the following: the alignment of multiple candidate graphic elements with multiple partitions in the image; the edge integrity of multiple candidate graphic elements; or the style consistency among multiple candidate graphic elements.

[0115] In some cases, the electronic device 110 can map an image onto multiple planes based on the image and three-dimensional structural information; and set multiple planes according to spatial relationships to obtain a three-dimensional representation of the object.

[0116] In some cases, the first instruction may include at least one of the text or reference image.

[0117] In some cases, an image may include a first channel and a second channel, where the first channel indicates the color information of the image and the second channel indicates the transparency information of the image.

[0118] In some cases, electronic device 110 can generate a three-dimensional representation of an object based on color information and transparency information.

[0119] In some cases, the spatial relationship of multiple planes can indicate at least one of the following: the angles of the multiple planes; the scaling ratios of the multiple planes; or the spatial positions of the multiple planes.

[0120] In some cases, the object can be a plant.

[0121] In some cases, the electronic device 110 can extract at least one of category information, color information, shape information and generation parameters from the first indication to obtain structured indication information; and obtain spatial constraint information based on the structured indication information.

[0122] In some cases, the three-dimensional representation of an image and an object can be generated by processing the entity by separately invoking an image generation tool and an assembly tool.

[0123] In some cases, the electronic device 110 may receive a second instruction describing another object; generate a second image based on the second instruction; and generate a three-dimensional representation of the other object based on the second image. In some cases, the electronic device 110 may reuse previous spatial constraint information to generate a three-dimensional representation of the other object.

[0124] Through the above scheme, electronic device 110 can generate an image comprising multiple graphic elements representing the object based on a first indication used to describe the object, and map the image onto multiple planes arranged according to spatial relationships, thereby generating a three-dimensional representation of the object. In cases where spatial constraint information is used, electronic device 110 can also utilize the mapping relationship between the first category and the spatial constraint information to select two-dimensional layout and three-dimensional structural information, and regenerate the image if the candidate image does not meet quality requirements. Spatial constraint information can be reused; candidate selection, color and transparency information binding, and image mapping to multiple planes can be completed in the same process, thereby reducing the number of repetitive geometry constructions, manual alignments, and invalid image assemblies, and lowering the overhead of geometry processing, storage, and rendering resources.

[0125] A corresponding apparatus for implementing the above methods or processes is also provided.

[0126] Figure 8 Block diagrams of apparatus 800 for generating three-dimensional representations of objects are shown in some scenarios. Apparatus 800 can be implemented as or included in electronic device 110. The various modules / components in apparatus 800 can be implemented by hardware, software, firmware, or any combination thereof.

[0127] like Figure 8As shown, the device 800 includes: an indication receiving module 810 configured to receive a first indication for describing an object; an image generation module 820 configured to generate an image based on the first indication, the image including multiple graphic elements, each graphic element representing an object; and a representation generation module 830 configured to generate a three-dimensional representation of the object based on the image, the three-dimensional representation including multiple planes arranged according to spatial relationships, the image being mapped to the multiple planes respectively.

[0128] In some cases, the device 800 may also be configured to obtain a first category based on a first indication, the first category corresponding to the shape of an object; and to obtain spatial constraint information based on the first category and a mapping relationship, the spatial constraint information indicating structural constraints on the three-dimensional representation of the object, the mapping relationship including the mapping between the first category and the spatial constraint information, wherein the three-dimensional representation of the object is generated based on the spatial constraint information.

[0129] In some cases, spatial constraint information may include two-dimensional layout and three-dimensional structural information. The two-dimensional layout indicates the arrangement of multiple graphic elements in the image, while the three-dimensional structural information indicates the spatial relationship between multiple planes.

[0130] In some cases, the image generation module 820 can also be configured to generate an image based on a first instruction and a two-dimensional layout, the image including multiple partitions, and multiple graphic elements distributed in the multiple partitions according to the two-dimensional layout.

[0131] In some cases, the representation generation module 830 can also be configured to generate a three-dimensional representation of an object based on the image and three-dimensional structural information.

[0132] In some cases, the mapping relationship may include a mapping between a first category and multiple candidate spatial constraint information. The apparatus 800 may also be configured to obtain multiple candidate spatial constraint information based on the first category and the mapping relationship; and to receive a selection operation for selecting spatial constraint information from the multiple candidate spatial constraint information.

[0133] In some cases, the image generation module 820 may also be configured to obtain visual information based on a first indication, the visual information indicating the visual style of the image; and generate an image based on the visual style.

[0134] In some cases, the image generation module 820 may also be configured to generate multiple candidate images based on a first instruction, each candidate image including multiple candidate graphic elements, each candidate graphic element representing an object; obtain quality information indicating the image quality of each of the multiple candidate images; and select an image from the multiple candidate images based on the quality information.

[0135] In some cases, the image generation module 820 may also be configured to, in response to the image quality of the first candidate image not meeting the quality requirements, adjust a first value of the first parameter to obtain a second value of the first parameter, the first value being used to generate a first candidate image, the first candidate image being included among multiple candidate images; and regenerate the first candidate image based on the second value of the first parameter.

[0136] In some cases, the image quality of multiple candidate images can indicate at least one of the following: the alignment of multiple candidate graphic elements with multiple partitions in the image; the edge integrity of multiple candidate graphic elements; or the style consistency among multiple candidate graphic elements.

[0137] In some cases, the representation generation module 830 can also be configured to map an image to multiple planes based on the image and three-dimensional structural information; and to set multiple planes according to spatial relationships to obtain a three-dimensional representation of the object.

[0138] In some cases, the first instruction may include at least one of the text or reference image.

[0139] In some cases, an image may include a first channel and a second channel, where the first channel indicates the color information of the image and the second channel indicates the transparency information of the image.

[0140] In some cases, the representation generation module 830 can also be configured to generate a three-dimensional representation of an object based on color and transparency information.

[0141] In some cases, the spatial relationship of multiple planes can indicate at least one of the following: the angles of the multiple planes; the scaling ratios of the multiple planes; or the spatial positions of the multiple planes.

[0142] In some cases, the object can be a plant.

[0143] In some cases, the device 800 may also be configured to extract at least one of category information, color information, shape information and generation parameters from the first indication to obtain structured indication information; and to obtain spatial constraint information based on the structured indication information.

[0144] In some cases, the three-dimensional representation of an image and an object can be generated by processing the entity by separately invoking an image generation tool and an assembly tool.

[0145] In some cases, the instruction receiving module 810, the image generation module 820, and the representation generation module 830 may also be configured to receive a second instruction describing another object; generate a second image based on the second instruction; and generate a three-dimensional representation of the other object based on the second image. In some cases, the device 800 may also reuse previous spatial constraint information to generate a three-dimensional representation of the other object.

[0146] The modules included in device 800 can be implemented in various ways, including software, hardware, firmware, or any combination thereof. In some cases, one or more modules can be implemented using software and / or firmware, such as machine-executable instructions stored on a storage medium. In addition to or as an alternative to machine-executable instructions, some or all of the units in device 800 can be implemented at least partially by one or more hardware logic components. By way of example, and not limitation, exemplary types of hardware logic components that can be used include field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard parts (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), and so on.

[0147] Figure 9 A block diagram of an electronic device 900 in which one or more examples may be implemented is shown. It should be understood that... Figure 9 The electronic device 900 shown is merely exemplary and should not be construed as limiting the functionality and scope of the examples described herein. Figure 9 The illustrated electronic device 900 can be used to implement the electronic device 110 discussed above.

[0148] like Figure 9 As shown, electronic device 900 is in the form of a general-purpose electronic device. Components of electronic device 900 may include, but are not limited to, one or more processing units or processors 910, memory 920, storage devices 930, one or more communication units 940, one or more input devices 950, and one or more output devices 960. Processor 910 may be a physical or virtual processor and is capable of performing various processes according to programs stored in memory 920. In a multiprocessor system, multiple processors execute computer-executable instructions in parallel to improve the parallel processing capability of electronic device 900.

[0149] Electronic device 900 typically includes multiple computer storage media. Such media can be any accessible media that is accessible to electronic device 900, including but not limited to volatile and non-volatile media, removable and non-removable media. Memory 920 can be volatile memory (e.g., registers, cache, random access memory (RAM)), non-volatile memory (e.g., read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof). Storage device 930 can be removable or non-removable media and can include machine-readable media, such as flash drives, disks, or any other media that can be used to store information and / or data and can be accessed within electronic device 900.

[0150] Electronic device 900 may further include additional removable / non-removable, volatile / non-volatile storage media. Although not explicitly stated... Figure 9 As shown, disk drives for reading from or writing to removable, non-volatile disks (e.g., "floppy disks") and optical disk drives for reading from or writing to removable, non-volatile optical disks can be provided. In these cases, each drive can be connected to a bus (not shown) via one or more data media interfaces. Memory 920 may include computer program product 925 having one or more program modules configured to perform various methods or actions of various examples.

[0151] The communication unit 940 enables communication with other electronic devices via a communication medium. Additionally, the functionality of the components of the electronic device 900 can be implemented using a single computing cluster or multiple computing machines capable of communicating via communication connections. Therefore, the electronic device 900 can operate in a networked environment using logical connections to one or more other servers, networked personal computers, or another network node.

[0152] Input device 950 can be one or more input devices, such as a mouse, keyboard, trackball, etc. Output device 960 can be one or more output devices, such as a monitor, speaker, printer, etc. Electronic device 900 can also communicate with one or more external devices (not shown) via communication unit 940 as needed. External devices include storage devices, display devices, etc., and can communicate with one or more devices that enable user interaction with electronic device 900, or with any device that enables electronic device 900 to communicate with one or more other electronic devices (e.g., network card, modem, etc.). Such communication can be performed via input / output (I / O) interface (not shown).

[0153] A computer-readable storage medium is provided having computer-executable instructions stored thereon, wherein the computer-executable instructions are executed by a processor 910 to implement the method described above. A computer program product 925 is also provided, which is tangibly stored on a non-transient computer-readable medium and includes computer-executable instructions, which are executed by a processor 910 to implement the method described above.

[0154] The flowcharts and / or block diagrams of the methods, apparatus, devices, and computer program products referred to herein describe various aspects. It should be understood that each block of the flowcharts and / or block diagrams, as well as combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer-readable program instructions.

[0155] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0156] Computer-readable program instructions can be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions that execute on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0157] The flowcharts and block diagrams in the accompanying figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products under various scenarios. In this respect, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing the specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the figures. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0158] Various examples have been described above. The foregoing descriptions are exemplary and not exhaustive, nor are they limited to the disclosed implementations. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described implementations. The terminology used herein is chosen to best explain the principles, practical applications, or improvements to technology in the market, or to enable others skilled in the art to understand the various implementations described herein.

Claims

1. A method for generating a three-dimensional representation of an object, comprising: Receive a first instruction, which describes the object; Based on the first instruction, an image is generated, the image comprising a plurality of graphic elements, each of which represents the object; and Based on the image, a three-dimensional representation of the object is generated, the three-dimensional representation including multiple planes arranged according to spatial relationships, and the image is mapped to the multiple planes respectively.

2. The method according to claim 1, further comprising: Based on the first instruction, a first category is obtained, the first category corresponding to the form of the object; as well as Based on the first category and the mapping relationship, spatial constraint information is obtained. The spatial constraint information indicates structural constraints on the three-dimensional representation. The mapping relationship includes the mapping between the first category and the spatial constraint information. The three-dimensional representation of the object is generated based on the spatial constraint information.

3. The method according to claim 2, wherein the spatial constraint information includes: Two-dimensional layout and three-dimensional structural information, wherein the two-dimensional layout indicates the arrangement of the plurality of graphic elements in the image, and the three-dimensional structural information indicates the spatial relationship between the plurality of planes.

4. The method of claim 3, wherein generating the image based on the first instruction comprises: Based on the first instruction and the two-dimensional layout, the image is generated, the image including multiple partitions, and the multiple graphic elements are distributed in the multiple partitions according to the two-dimensional layout.

5. The method of claim 3, wherein generating a three-dimensional representation of the object based on the image comprises: Based on the image and the three-dimensional structural information, the three-dimensional representation of the object is generated.

6. The method of claim 2, wherein the mapping relationship includes a mapping between the first category and a plurality of candidate spatial constraint information, and the step of obtaining the spatial constraint information based on the first category and the mapping relationship includes: Based on the first category and the mapping relationship, the multiple candidate space constraint information is obtained; as well as A selection operation is received, which is used to select the spatial constraint information from the plurality of candidate spatial constraint information.

7. The method of claim 1, wherein generating the image based on the first instruction comprises: Based on the first indication, visual information is obtained, the visual information indicating the visual style of the image; as well as The image is generated based on the visual style.

8. The method of claim 1, wherein generating the image based on the first instruction comprises: Based on the first instruction, a plurality of candidate images are generated, each of the plurality of candidate images including a plurality of candidate graphic elements, the plurality of candidate graphic elements respectively representing the object; Obtain quality information, which indicates the image quality of each of the plurality of candidate images; as well as Based on the quality information, the image is selected from the plurality of candidate images.

9. The method of claim 8, further comprising: In response to the first candidate image's image quality not meeting quality requirements, a first value of a first parameter is adjusted to obtain a second value of the first parameter. This first value is used to generate the first candidate image, and the plurality of candidate images includes the first candidate image. The first candidate image is regenerated based on the second value of the first parameter.

10. The method of claim 8, wherein the image quality of each of the plurality of candidate images indicates at least one of the following: The degree of alignment between the multiple candidate graphic elements and the multiple partitions in the image. The edge integrity of the multiple candidate graphic elements, or Style consistency among the multiple candidate graphic elements.

11. The method of claim 5, wherein generating a three-dimensional representation of the object based on the image and the three-dimensional structural information comprises: Map the image onto the plurality of planes; as well as By setting up the multiple planes according to the spatial relationship, a three-dimensional representation of the object is obtained.

12. The method of claim 1, wherein the first indication comprises at least one of text or a reference image.

13. The method of claim 1, wherein the image includes a first channel and a second channel, the first channel indicating color information of the image and the second channel indicating transparency information of the image.

14. The method of claim 13, wherein generating a three-dimensional representation of the object based on the image comprises: The three-dimensional representation of the object is generated based on the color information and the transparency information.

15. The method of claim 1, wherein the spatial relationship indicates at least one of the following: The angles of each of the multiple planes The scaling ratio of each of the multiple planes, or The spatial positions of the multiple planes.

16. The method of claim 1, wherein the object is a plant.

17. An apparatus for generating a three-dimensional representation of an object, comprising: An instruction receiving module is configured to receive a first instruction, the first instruction being used to describe the object; An image generation module is configured to generate an image based on the first instruction, the image including a plurality of graphic elements, each of which represents the object; and The representation generation module is configured to generate a three-dimensional representation of the object based on the image. The three-dimensional representation includes multiple planes arranged according to spatial relationships, and the image is mapped to the multiple planes respectively.

18. An electronic device comprising: At least one processor; as well as At least one memory coupled to the at least one processor and storing instructions for execution by the at least one processor, the instructions causing the electronic device to perform the method according to any one of claims 1 to 16 when executed by the at least one processor.

19. A computer-readable storage medium having a computer program stored thereon, the computer program being executable by a processor to implement the method according to any one of claims 1 to 16.

20. A computer program product tangibly stored in a computer storage medium and comprising computer-executable instructions that, when executed by a device, cause the device to perform the method according to any one of claims 1 to 16.

Images