Method and device for constructing rare bird image database based on three-dimensional model

Abstract

The application provides a method and device for constructing a rare bird image database based on a three-dimensional model, and belongs to the technical field of rare bird image augmentation. After an initial image set including multiple actually photographed rare bird images is obtained, a three-dimensional model of the rare bird is generated from the initial image set by using an artificial intelligence modeling model, and multiple generated images of the rare bird are obtained by adjusting the posture of the three-dimensional model, so that a rare bird image database with more images than the initial image set is constructed. The above process realizes the augmentation of the rare bird images, and is conducive to the research of the rare bird.

Description

Technical Field

[0001] This invention relates to the field of rare bird image augmentation technology, and in particular to a method and apparatus for constructing a rare bird image database based on a three-dimensional model. Background Technology

[0002] Rare birds are generally defined as bird species that are few in number in nature, have a limited distribution range, or face a high risk of extinction (such as critically endangered species with <250 individuals and endangered species with <2500 individuals according to IUCN standards).

[0003] Because rare birds are few in number and depend on special habitats, they are difficult to observe, resulting in a limited number of images and making research on them challenging. Summary of the Invention

[0004] This invention proposes a method and apparatus for constructing a rare bird image database based on a three-dimensional model. It uses an artificial intelligence modeling model to construct a three-dimensional model of rare birds based on sample images, and then uses the three-dimensional model to construct generated images of rare birds. This can realize the amplification of rare bird images, which is beneficial to the research of rare birds.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for constructing a rare bird image database based on a three-dimensional model, comprising: Obtain an initial image set of rare birds; the initial image set includes multiple sample images of each rare bird species among a variety of rare birds; for each rare bird species among a variety of rare birds, the multiple sample images of the rare bird show multiple key parts and multiple morphological features of the rare bird in standing and / or flying postures from multiple shooting angles; the sample images are two-dimensional bird images taken in real-world scenes. Based on the initial image set of rare birds and the artificial intelligence modeling model, a 3D model of each rare bird species was determined; the artificial intelligence modeling model was constructed based on a generative adversarial network or a diffusion model. The poses of the 3D models of each rare bird species in a variety of rare bird species are adjusted to generate a rare bird image database. The rare bird image database includes multiple generated images of each rare bird species in a variety of rare bird species. The multiple generated images of each rare bird species include multiple 2D bird images of each rare bird species in various poses at various shooting angles. The various poses include standing, flying, crouching, preening, alert, intimidating, attacking, and courtship poses.

[0006] In one implementation of the first aspect, based on an initial image set of rare birds and an artificial intelligence modeling model, a 3D model of each rare bird species among a variety of rare birds is determined, including: For each rare bird species among a variety of rare bird species, prompt words are generated based on multiple sample images and 3D models of the rare bird species. An artificial intelligence modeling model is used to analyze the multiple sample images of the rare bird species to generate candidate 3D models. The prompt words generated by the 3D model are used to guide the artificial intelligence modeling model to generate candidate 3D models. Texture mapping and skeletal binding were applied to the candidate 3D models to obtain 3D models of rare birds.

[0007] In one implementation of the first aspect, the 3D model generates prompts describing the species name of the rare bird, multiple key parts, and various morphological features.

[0008] In one implementation of the first aspect, the artificial intelligence modeling model is the Tripo model, the Meshy model, or the ShapeGAN model.

[0009] In one implementation of the first aspect, the pose of the 3D model of each rare bird species among a variety of rare bird species is adjusted to generate a rare bird image database, including: Step 1: For each rare bird species among the various rare bird species, place the 3D model of the rare bird species in the world coordinate system; Step 2: Set up a virtual camera and light source in the world coordinate system to photograph the 3D model of the rare bird. Step 3: Adjust the pose of the 3D model of the rare bird, the shooting angle of the virtual camera, and the lighting effects; Step 4: Render the 3D model of the rare bird based on the virtual camera and light source; Step 5: Blend the rendered rare bird image with the background of the natural environment in which the rare bird lives to obtain a composite image of the rare bird. Step 6: Repeat steps 3 to 5 to obtain multiple generated images of rare birds; Step 7: Integrate multiple generated images of each rare bird species from various rare bird species to obtain a rare bird image database.

[0010] In one implementation of the first aspect, several key parts include the crest, beak, neck, chest, abdomen, legs, claws, and tail feathers; Various morphological characteristics include: feather features of the crest, neck, chest, abdomen, legs, and tail feathers; color and length of the beak and claws; feather features include feather color, pattern, and orientation. Multiple shooting angles are available, including eye-level, upward-looking, downward-looking, frontal, side, and rear angles.

[0011] Secondly, the present invention provides an apparatus for constructing a rare bird image database based on a three-dimensional model, including an image set acquisition module, a three-dimensional model construction module, and a database construction module.

[0012] The image set acquisition module is used to acquire an initial image set of rare birds; the initial image set includes multiple sample images of each rare bird among a variety of rare birds; for each rare bird among a variety of rare birds, the multiple sample images of the rare bird show multiple key parts and multiple morphological features of the rare bird in standing and / or flying postures from multiple shooting angles; the sample images are two-dimensional bird images taken in real-world scenes. The 3D model building module is used to determine the 3D model of each rare bird species among a variety of rare bird species, based on an initial image set of rare birds and an artificial intelligence modeling model; the artificial intelligence modeling model is built based on a generative adversarial network or a diffusion model. The database construction module is used to adjust the pose of the 3D model of each rare bird species among a variety of rare bird species to generate a rare bird image database. The rare bird image database includes multiple generated images of each rare bird species among a variety of rare bird species. The multiple generated images of each rare bird species include multiple 2D bird images of each rare bird species in various poses at various shooting angles. The various poses include standing, flying, crouching, preening, alert, intimidating, attacking, and courtship poses.

[0013] Thirdly, the present invention provides an electronic device including a processor and a memory coupled to the processor; the memory is used to store computer instructions, and when the electronic device is running, the processor executes the computer instructions stored in the memory to cause the electronic device to perform the method described in the first aspect above or any implementation thereof.

[0014] Fourthly, the present invention provides a computer-readable storage medium including computer program instructions that, when executed by a computer, cause the computer to perform the method described in the first aspect above or any implementation thereof.

[0015] Fifthly, the present invention provides a computer program product, including computer program instructions, which, when executed on a computer, cause the computer to perform the method described in the first aspect above or any implementation thereof.

[0016] The technical effects corresponding to the second to fifth aspects and their possible implementations can be referred to the above description of the technical effects of the first aspect and its possible implementations, and will not be repeated here.

[0017] Compared with the prior art, the present invention has the following beneficial effects.

[0018] The method for constructing a rare bird image database based on a 3D model provided by this invention involves obtaining an initial image set including multiple real-shot images of rare birds, then using an artificial intelligence modeling model to generate 3D models of rare birds based on the initial image set, and then adjusting the pose of the 3D model to obtain multiple generated images of rare birds, thereby constructing a rare bird image database with a larger number of images than the initial image set. The above process realizes the expansion of rare bird images, which is beneficial to the research of rare birds. Attached Figure Description

[0019] Figure 1 This is one of the schematic diagrams illustrating the method for constructing a rare bird image database based on a three-dimensional model provided in this application embodiment; Figure 2 This is the second method for constructing a rare bird image database based on a three-dimensional model provided in the embodiments of this application; Figure 3 This is a schematic diagram of the structure of the device for constructing a rare bird image database based on a three-dimensional model provided in the embodiments of this application. Detailed Implementation

[0020] In the specification and claims of this invention, the terms "first" and "second," etc., are used to distinguish different objects, rather than to describe a specific order of objects.

[0021] In the embodiments of this application, "and / or" indicates a relationship between objects. For example, A and / or B can represent the following three situations: A exists alone, B exists alone, and A and B exist simultaneously.

[0022] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0023] In the description of this invention, unless otherwise stated, "multiple images" means two or more images. For example, multiple sample images refer to two or more sample images; multiple generated images refer to two or more generated images.

[0024] The method and apparatus provided in this application relate to the construction of a rare bird image database. They can generate a large number of rare bird images in various specific poses from a small number of rare bird images in specific poses, thereby constructing a rare bird image database.

[0025] Understandably, rare birds, as important indicator species of ecosystems and key components of biodiversity, possess multidimensional research value. This value extends beyond the field of biology, reaching into areas such as ecological policy, technological innovation, and social ethics.

[0026] However, existing images of rare birds are scarce, and there are problems such as: 1. Insufficient sample size due to limited observation conditions in field shooting; 2. Difficulty in directly observing the habitat preferences and breeding behaviors of rare birds; 3. Low image clarity and accuracy of rare birds; 4. Difficulty in directly obtaining images of rare birds with specific movements and angles through field shooting; 5. Lack of images of certain bird species at specific growth stages or in different regions.

[0027] In recent years, with the development of artificial intelligence technology, artificial intelligence can analyze massive images and model data through deep learning models, extract visual features and volume patterns, and then generate three-dimensional models based on reference images based on algorithms.

[0028] To address the problem in the background art that the small number of rare birds and their dependence on specific habitats make observation of rare birds difficult, resulting in a limited number of rare bird images and consequently hindering research on rare birds, this application provides a method for constructing a rare bird image database based on a three-dimensional model. This method employs an artificial intelligence modeling model to construct three-dimensional models of rare birds based on sample images, and then uses these three-dimensional models to generate images of rare birds. This approach can expand the number of rare bird images, thereby facilitating research on rare birds.

[0029] For example, the method for constructing a rare bird image database based on a 3D model provided in this embodiment of the invention can be executed by an electronic device with processing capabilities, such as a computer or server. Taking a computer as an example, the hardware components of the computer may include: a processor, memory, a network interface, a user interface, a communication bus, etc.

[0030] The processor controls the electronic device to perform related processing and computation tasks, such as acquiring an initial image set of rare birds, determining a 3D model of each rare bird species, and generating a rare bird image database. The processor may include a central processing unit (CPU) or other processors, and may be single-core or multi-core; for example, the processor may include multiple CPUs.

[0031] Memory is used to store computer instructions and related data, such as storing initial image sets of rare birds, artificial intelligence modeling models, and generating rare bird image databases. Memory can be random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical storage, disk storage media, or other magnetic storage devices, or any other medium capable of storing program code or data accessible by a computer. Optionally, memory can be integrated into the processor, or it can be independent of the processor.

[0032] A network interface is used for communication between a computer and other devices or communication networks. A network interface can be a transceiver with transmit and receive capabilities. Optionally, a network interface may include standard wired interfaces or wireless interfaces (such as Wi-Fi interfaces, Bluetooth interfaces, and 5G interfaces).

[0033] The communication bus is used to enable communication between different components. For example, the processor, memory, network interface and user interface mentioned above can be interconnected through the communication bus.

[0034] The user interface may include a display screen and an input unit (such as a keyboard). Optionally, the user interface may also include a standard wired interface or a wireless interface.

[0035] Those skilled in the art will understand that the computer described above may include more or fewer components, or combine certain components, or have different component arrangements; the embodiments of this application do not limit this.

[0036] like Figure 1 As shown, the method for constructing a rare bird image database based on a three-dimensional model provided in this application includes steps S101-S103.

[0037] S101. Obtain the initial image set of rare birds.

[0038] In this embodiment of the application, the initial image set includes multiple sample images of each rare bird species. For each rare bird species, the multiple sample images showcase multiple key parts and morphological features of the rare bird in standing and / or flying postures from various shooting angles. The aforementioned sample images refer to two-dimensional bird images taken in real-world conditions.

[0039] Alternatively, the aforementioned rare bird species may include the kakapo, Chinese merganser, heron, Sumatran ground cuckoo, and Philippine eagle. The aforementioned key body parts may include the crest, beak, neck, chest, abdomen, legs, claws, and tail feathers; the aforementioned morphological characteristics may include the plumage features of the crest, neck, chest, abdomen, legs, and tail feathers, as well as the color and length of the beak and claws; plumage characteristics include the color, pattern, and orientation of the feathers.

[0040] Specifically, the selection principles for sample images in the initial image set of the above rare birds are given below.

[0041] (1) From the perspective of image elements: ① Subject: The main subject is a bird, showing the bird in flight or standing posture. The individual bird should occupy more than half of the picture to highlight its body shape, so that the AI ​​can restore the bird's actual body shape and detailed structure to the greatest extent.

[0042] ② Background: Choose bird images with blank, solid color, or transparent backgrounds whenever possible. If no such images are available, select images with relatively simple backgrounds and manually or using AI to extract the bird subjects. If the bird is a raptor, the background should ideally be a sky or open ground scene; if the bird is a forest bird, the foreground should be free of obstructing elements such as branches, and the background branches should minimize the area obscured by the bird. Overall, the background should be free of complex or unnecessary elements that interfere with the presentation of the subject, and the contrast between the bird and the background should be high (e.g., a dark bird against a light sky, or a light bird against a dark forest background). (2) From the perspective of visual features: ① Clarity: Key parts (beak, claws, flight feathers) are clearly detailed, and feather textures (such as barb arrangement and spots) are discernible. This provides AI models with more accurate and detailed features (such as feather layers and eye structure), helping the models learn more realistic bird forms and generate higher-quality new images. Even if the image clarity is low or blurry (such as images of some birds of prey), it can still provide information such as the basic shape, color, and basic features of the bird, with a complete main outline.

[0043] ② Viewpoint selection: mainly from the side, with a few from a slightly overhead angle (showing the full view of the wings when in flight).

[0044] (3) From the perspective of specific morphological characteristics of birds: ① Posture: Images that correspond to the most common behaviors of the bird species (such as standing or flying) should be selected. Images with small movements (such as scratching, flapping wings, etc.) are difficult for AI to identify key subject features and movements when used as modeling reference images, and the generated models are more likely to have distorted or incorrect shapes.

[0045] ②Crown: The crest features are obvious (if the crest is not obvious due to the shooting angle or bird species, it can be ignored).

[0046] ③Beak: The base of a bird's beak has a clear outline where it connects to its head.

[0047] ④ Neck (throat + front of neck), chest, abdomen.

[0048] Neck: From a side or frontal view, the neck feathers are arranged in a straight line, and the throat and foreneck feathers are of a single color or have simple stripes.

[0049] Chest and abdomen: The chest is densely feathered, showing a full muscle outline (especially the chest of birds of prey is broad), while the abdominal feathers are smooth.

[0050] ⑤ Tail feathers Tail feathers: All are in a spread or semi-spread state (such as forked tail feathers during flight, or horizontal tail feathers during gliding), with clear edges on the feather vanes and no breaks or abnormal missing parts.

[0051] ⑥ Legs and claws Legs: In images of a standing posture, the legs are mostly bare (without feathers covering the feet), and the scale patterns on the shanks are visible (such as reticulated or transverse scales). The leg color is mostly gray, yellow, or orange (contrasting with the environment).

[0052] Claws: The claws are all in a clenched or grasping state (such as grasping a branch or prey), with sharp, curved tips. Some pictures show webbed feet (such as waterbirds).

[0053] S102. Based on the initial image set of rare birds and the artificial intelligence modeling model, determine the three-dimensional model of each rare bird among the various rare bird species.

[0054] The aforementioned AI modeling models are constructed based on generative adversarial networks (GANs) or diffusion models. Specifically, when the AI ​​modeling model is constructed based on a diffusion model, it can be a Tripo model or a Meshy model, etc.; specifically, the Tripo model mentioned above is the Tripo (AI version: v3.0) model, and the Meshy model mentioned above is the Meshy (AI version: Meshy 6 preview) model. When the AI ​​modeling model is based on a generative adversarial network (GAN), it can be a ShapeGAN model or a 3D-GAN / VoxGAN model.

[0055] Since the artificial intelligence modeling model used in this application embodiment belongs to the prior art, this application embodiment will not further elaborate on the structure of the above-mentioned artificial intelligence modeling model.

[0056] In one implementation, combined with Figure 1 ,like Figure 2 As shown, S102 includes S1021-S1022.

[0057] S1021. For each rare bird species among a variety of rare bird species, generate prompt words based on multiple sample images and 3D models of the rare bird species, and use an artificial intelligence modeling model to analyze the multiple sample images of the rare bird species to generate candidate 3D models.

[0058] Specifically, for each rare bird species among various rare bird species, multiple sample images and 3D models of the rare bird species are used to generate prompts which are then input into the artificial intelligence modeling model. Guided by the prompts generated by the 3D model, the artificial intelligence modeling model generates and outputs candidate 3D models based on the multiple sample images of the rare bird species.

[0059] Understandably, the aforementioned 3D model generation prompts are used to guide the artificial intelligence modeling model in generating candidate 3D models. These prompts describe the species name of the rare bird, several key body parts, and various morphological features.

[0060] Specifically, the prompts generated by the aforementioned 3D model need to meet the following two requirements.

[0061] (1) The number of keywords should be no more than three.

[0062] When an AI model generates a 3D model, an excessive number of keywords beyond its comprehension range can negatively impact its understanding of the reference image, potentially preventing the generation of a satisfactory 3D model. Therefore, it is advisable to provide no more than three keywords.

[0063] (2) Detailed classification information such as order, family, and genus must be provided.

[0064] When an AI model generates a 3D model based on multiple sample images of rare birds, if it only describes "birds on the ground," the AI ​​model may easily misidentify them as thrushes. Therefore, it is necessary to supplement the model with detailed classification information such as order, family, and genus. Furthermore, the word "eagle" should be avoided in the instructions to prevent the AI ​​model from automatically adding eagle characteristics.

[0065] S1022. Apply textures and rigging to the candidate 3D models to obtain 3D models of rare birds.

[0066] In one application scenario, the specific process of S1022 is as follows.

[0067] (1) Optimize candidate 3D models.

[0068] ① Wiring and Surfaces: The candidate 3D models generated by AI modeling often exhibit significant uncertainty in their wiring, frequently resulting in messy and uneven wiring and a large number of three-sided and polygonal faces, which can cause considerable inconvenience to subsequent model operations. Therefore, it is necessary to first evaluate the wiring of the generated candidate 3D models. If the wiring and surfaces of the candidate 3D models are messy, it is necessary to use modeling software to retopologically reshape the candidate 3D models so that the wiring approaches the formation of a mesh dominated by flat quadrilateral faces.

[0069] ② Model defects: The candidate 3D models generated by the artificial intelligence modeling model may have local structural defects, such as burrs, clipping, distortion and broken surfaces. These defects need to be manually repaired in the modeling software.

[0070] (2) Material texture creation.

[0071] To ensure the final rendered model accurately reflects the original appearance of rare birds, textures must be sourced directly from original bird images, meticulously reproducing all body elements. Key areas (beak, claws, flight feathers, eye structure, etc.) must be clearly detailed, and feather textures (such as barb arrangement, spots, and feather layers) must be discernible, contributing to a more realistic bird appearance. Some detailed textures may be less sharp or blurry due to a lack of high-resolution images in the original image library (e.g., blind spots), but if information such as basic shape, color, and fundamental features is available to ensure complete texture coverage, they can still be used as appropriate. Furthermore, the overall UV mapping image size should be at least 1000*1000 pixels to ensure texture clarity.

[0072] (3) Bind the skeleton.

[0073] Adding skeletons that conform to the structure and movement patterns of rare birds to the candidate 3D model will provide the candidate 3D model with the ability to adjust its posture and ensure that the changes in its movements are regular, so as to restore the specific postures of rare birds.

[0074] S103. Adjust the pose of the 3D model of each rare bird species in a variety of rare bird species to generate a rare bird image database.

[0075] The aforementioned rare bird image database includes multiple generated images of each rare bird species; each generated image of a rare bird species includes multiple two-dimensional bird images of each rare bird species in various poses from multiple shooting angles. It is understood that the aforementioned rare bird image database may also include multiple sample images contained in the initial image set in S101. This application embodiment does not further limit the image content contained in the aforementioned rare bird image database.

[0076] For example, the aforementioned multiple shooting angles may include eye-level angle, upward angle, downward angle, front angle, side angle, and back angle, etc. The aforementioned multiple postures may include standing posture, flying posture, curled-up posture, preening posture, alert posture, intimidating posture, attacking posture, and courtship posture, etc. This application embodiment does not further limit the aforementioned multiple shooting angles and multiple postures.

[0077] Optionally, S103 above includes the following steps.

[0078] Step 1: For each rare bird species among the various rare bird species, place the 3D model of the rare bird species in the world coordinate system; Step 2: Set up a virtual camera and light source in the world coordinate system to photograph the 3D model of the rare bird. Step 3: Adjust the pose of the 3D model of the rare bird, the shooting angle of the virtual camera, and the lighting effects; Step 4: Render the 3D model of the rare bird based on the virtual camera and light source; Step 5: Blend the rendered rare bird image with the background of the natural environment in which the rare bird lives to obtain a composite image of the rare bird.

[0079] It should be noted that, since the world coordinate system, virtual camera, lighting and rendering mentioned above are common image processing techniques, the specific processing procedures of steps 3 to 5 above will not be described in detail in this embodiment of the application.

[0080] Step 6: Repeat steps 3 to 5 to obtain multiple generated images of rare birds.

[0081] Step 7: Integrate multiple generated images of each rare bird species from various rare bird species to obtain a rare bird image database.

[0082] In summary, the method for constructing a rare bird image database based on a 3D model provided in this application involves obtaining an initial image set including multiple real-shot images of rare birds, then using an artificial intelligence modeling model to generate 3D models of rare birds based on the initial image set, and then adjusting the pose of the 3D models to obtain multiple generated images of rare birds, thereby constructing a rare bird image database with a larger number of images than the initial image set. This process expands the number of rare bird images, which is beneficial for the research of rare birds.

[0083] Accordingly, embodiments of this application provide an apparatus for constructing a rare bird image database based on a three-dimensional model, such as... Figure 3 As shown, it includes an image set acquisition module 501, a 3D model construction module 502, and a database construction module 503.

[0084] The image set acquisition module 501 is used to acquire an initial image set of rare birds. The initial image set includes multiple sample images of each rare bird species. For each rare bird species, the multiple sample images show multiple key parts and morphological features of the rare bird in standing and / or flying postures from various shooting angles. The sample images are two-dimensional bird images taken in real-world conditions. For example, the image set acquisition module 501 is used to implement step S101 of the above method.

[0085] The 3D model construction module 502 is used to determine the 3D model of each rare bird species among a variety of rare birds based on an initial image set of rare birds and an artificial intelligence modeling model; the artificial intelligence modeling model is constructed based on a generative adversarial network or a diffusion model. For example, the 3D model construction module 502 is used to implement S102 of the above method.

[0086] The database construction module 503 is used to adjust the pose of the 3D model of each rare bird species among a variety of rare bird species to generate a rare bird image database. The rare bird image database includes multiple generated images of each rare bird species. Each generated image of a rare bird species includes multiple 2D bird images of each rare bird species in various poses from multiple shooting angles. These various poses include standing, flying, crouching, preening, alert, intimidating, attacking, and courtship postures. For example, the database construction module 503 is used to implement step S103 of the above method.

[0087] The modules of the above-mentioned device for constructing a rare bird image database based on a 3D model can also be used to perform other steps in the above method embodiments. All relevant content involved in the above method embodiments can be referred to in the functional description of the corresponding functional module, and will not be repeated here.

[0088] This application also provides an electronic device, including: a processor and a memory coupled to the processor; the memory is used to store computer instructions, and when the electronic device is running, the processor executes the computer instructions stored in the memory to cause the electronic device to perform the methods described in the above embodiments. The processor can implement the image set acquisition module 501, the 3D model construction module 502, and the database construction module 503; the memory can also be used to store an initial image set of rare birds, an artificial intelligence modeling model, and to generate a rare bird image database, etc.

[0089] This application also provides a computer-readable storage medium including a computer program that, when run on a computer, performs the methods described in the above embodiments.

[0090] This application also provides a computer program product, which includes computer program instructions that, when run on a computer, execute the methods described in the above embodiments.

[0091] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for constructing a rare bird image database based on a 3D model, characterized in that, include: Obtain an initial image set of rare birds; The initial image set includes multiple sample images of each rare bird species from a variety of rare bird species; For each of the aforementioned rare bird species, multiple sample images of the rare bird species are presented from various shooting angles, showing multiple key parts and morphological features of the rare bird species in standing and / or flying postures; the sample images are two-dimensional bird images taken in real-world scenes. Based on the initial image set of the rare birds and the artificial intelligence modeling model, a 3D model of each of the various rare birds is determined; the artificial intelligence modeling model is constructed based on a generative adversarial network or a diffusion model. The pose of the 3D model of each rare bird species among the various rare bird species is adjusted to generate a rare bird image database; the rare bird image database includes multiple generated images of each rare bird species among the various rare bird species; the multiple generated images of each rare bird species include multiple 2D bird images of each rare bird species in multiple poses under multiple shooting angles; the multiple poses include standing posture, flight posture, huddled posture, preening posture, alert posture, intimidating posture, attack posture, and courtship posture.

2. The method as described in claim 1, characterized in that, The step of determining a 3D model for each of the various rare bird species based on an initial image set and an artificial intelligence modeling model includes: For each of the various rare bird species, prompt words are generated based on multiple sample images and 3D models of the rare bird species. The artificial intelligence modeling model is then used to analyze the multiple sample images of the rare bird species to generate candidate 3D models. The prompt words generated by the 3D model are used to guide the artificial intelligence modeling model in generating candidate 3D models. Texture mapping and skeletal binding are performed on the candidate 3D model to obtain the 3D model of the rare bird.

3. The method as described in claim 2, characterized in that, The 3D model generates prompts describing the species name, several key parts, and various morphological features of the rare bird.

4. The method as described in claim 1 or 2, characterized in that, The artificial intelligence modeling model is either the Tripo model, the Meshy model, or the ShapeGAN model.

5. The method as described in claim 1, characterized in that, The step of adjusting the pose of the 3D model of each of the various rare bird species to generate a rare bird image database includes: Step 1: For each of the various rare bird species, place the 3D model of the rare bird species in the world coordinate system; Step 2: Set up a virtual camera and light source in the world coordinate system to photograph the 3D model of the rare bird. Step 3: Adjust the posture of the 3D model of the rare bird, the shooting angle of the virtual camera, and the lighting effects; Step 4: Render the 3D model of the rare bird based on the virtual camera and light source; Step 5: Blend the rendered rare bird image with the background of the natural environment in which the rare bird lives to obtain a composite image of the rare bird. Step 6: Repeat steps 3 to 5 to obtain multiple generated images of the rare bird. Step 7: Integrate multiple generated images of each rare bird species to obtain the rare bird image database.

6. The method as described in claim 1 or 3, characterized in that, The key parts include the crest, beak, neck, chest, abdomen, legs, claws, and tail feathers; The various morphological features include: feather characteristics of the crest, neck, chest, abdomen, legs, and tail feathers; and the color and length of the beak and claws; the feather characteristics include the color, pattern, and orientation of the feathers. The various shooting angles include eye-level angle, upward angle, downward angle, front angle, side angle, and back angle.

7. A device for constructing a rare bird image database based on a three-dimensional model, characterized in that, It includes an image set acquisition module, a 3D model construction module, and a database construction module; The image set acquisition module is used to acquire an initial image set of rare birds; the initial image set includes multiple sample images of each rare bird among a variety of rare birds; for each rare bird among the variety of rare birds, the multiple sample images of the rare bird show multiple key parts and multiple morphological features of the rare bird in standing and / or flying postures from multiple shooting angles; the sample images are two-dimensional bird images taken in real-world scenes; The 3D model building module is used to determine the 3D model of each rare bird species among the various rare bird species based on the initial image set of the rare birds and the artificial intelligence modeling model; the artificial intelligence modeling model is constructed based on a generative adversarial network or a diffusion model. The database construction module is used to adjust the posture of the 3D model of each rare bird among the various rare bird species to generate a rare bird image database; the rare bird image database includes multiple generated images of each rare bird among the various rare bird species; the multiple generated images of each rare bird include multiple 2D bird images of each rare bird in multiple postures under multiple shooting angles; the multiple postures include standing posture, flight posture, huddled posture, preening posture, alert posture, intimidating posture, attack posture, and courtship posture.

8. An electronic device, characterized in that, The device includes a processor and a memory coupled to the processor; the memory is used to store computer instructions, which, when the electronic device is running, are executed by the processor to cause the electronic device to perform the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, It includes computer program instructions that, when executed by a computer, cause the computer to perform the method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, It includes computer program instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1 to 6.

Images