Digital display method and system of intelligent building blocks based on artificial intelligence technology

Abstract

The present application relates to the field of three-dimensional modeling technology, and specifically discloses a method and system for digital display of intelligent building blocks based on artificial intelligence technology, which uses computer vision technology based on deep learning to process multiple local perspective images of physical building blocks to mine local surface features of building blocks at various perspectives, and then performs three-dimensional correlation modeling on the local surface features of building blocks at multiple perspectives to obtain global structural information of the physical building blocks, thereby generating corresponding three-dimensional digital building block models. In this way, rapid digital display of physical building blocks can be achieved, effectively improving the accuracy and efficiency of physical building block modeling, thereby achieving widespread dissemination and sharing of building block ideas.

Description

Technical Field

[0001] The present application relates to the field of three-dimensional modeling technology, and more specifically, to a method and system for digitally displaying smart building blocks based on artificial intelligence technology. Background Art

[0002] As a classic creative toy, building blocks have a wide audience and entertainment value. However, traditional building block play mainly relies on the construction and display of physical building blocks. Due to the inherent physical properties of physical building blocks and space limitations, building block works have certain difficulties in preservation, sharing and dissemination.

[0003] In today's digital age, with the rapid development of artificial intelligence technology, the digital display of physical building blocks has become an effective way to share and spread building block works. Through 3D modeling technology and virtual reality technology, a three-dimensional digital model of physical building blocks is constructed to achieve online exhibitions and interactive displays. In this way, users can browse product displays and perform interactive operations through computers, mobile phones and other devices, thereby achieving unlimited expansion and free interaction of building block creativity, greatly enriching the communication and interactive experience of building block enthusiasts.

[0004] However, the existing digital building block display technology mainly relies on manually creating digital models of building blocks using 3D modeling software. This manual modeling method requires a lot of time and effort, is inefficient, and for complex building block works, its modeling accuracy and efficiency often cannot meet actual needs. Therefore, a smart building block digital display method and system based on artificial intelligence technology is expected. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a method and system for digital display of intelligent building blocks based on artificial intelligence technology, which uses computer vision technology based on deep learning to process multiple local perspective images of physical building blocks to mine the local surface features of the building blocks at various perspectives, and then obtains the global structural information of the physical building blocks by performing three-dimensional correlation modeling on the local surface features of the building blocks at multiple perspectives, thereby generating a corresponding three-dimensional digital building block model. In this way, the rapid digital display of physical building blocks can be achieved, and the accuracy and efficiency of physical building block modeling can be effectively improved, thereby realizing the wide dissemination and sharing of building block ideas.

[0006] Accordingly, according to one aspect of the present application, a method for digitally displaying smart building blocks based on artificial intelligence technology is provided, which includes:

[0007] Acquire multiple local perspective images of the physical building block to be displayed captured by a camera;

[0008] Performing image feature extraction and feature manifestation processing on each of the multiple local perspective images to obtain multiple salient building block local surface feature matrices;

[0009] Performing association modeling on the plurality of significant building block local surface feature matrices to obtain a building block local adjustment association feature map;

[0010] Based on the local adjustment of the associated feature map of the building blocks, a three-dimensional digital building block is generated;

[0011] Demonstrate the generation of three-dimensional digital building blocks.

[0012] In the above-mentioned intelligent building block digital display method based on artificial intelligence technology, image feature extraction and feature manifestation processing are performed on each of the multiple local perspective images to obtain multiple significant building block local surface feature matrices, including: image preprocessing is performed on the multiple local perspective images to obtain multiple preprocessed local perspective images; the building block surface features of each preprocessed local perspective image in the multiple preprocessed local perspective images are extracted to obtain multiple building block local surface feature matrices; image content attention distinction is performed on each building block local surface feature matrix in the multiple building block local surface feature matrices to obtain the multiple significant building block local surface feature matrices.

[0013] In the above-mentioned intelligent building block digital display method based on artificial intelligence technology, the building block surface features of each preprocessed local perspective image in the multiple preprocessed local perspective images are extracted respectively to obtain multiple building block local surface feature matrices, including: each preprocessed local perspective image in the multiple preprocessed local perspective images is passed through a building block surface feature extractor based on a void convolutional neural network model to obtain the multiple building block local surface feature matrices.

[0014] In the above-mentioned intelligent building block digital display method based on artificial intelligence technology, image content attention distinction is performed on each of the multiple building block local surface feature matrices to obtain the multiple significant building block local surface feature matrices, including: each of the multiple building block local surface feature matrices is input into a content distinction module based on a spatial mask to obtain the multiple significant building block local surface feature matrices.

[0015] In the above-mentioned intelligent building block digital display method based on artificial intelligence technology, each of the multiple building block local surface feature matrices is respectively input into a content differentiation module based on a spatial mask to obtain the multiple significant building block local surface feature matrices, including: calculating a feature significance score matrix of the building block local surface feature matrix; masking the feature significance score matrix based on a predetermined threshold to obtain a building block local surface feature attention mask matrix; using the building block local surface feature attention mask matrix as a weight matrix, calculating the position point multiplication of the building block local surface feature matrix and the building block local surface feature attention mask matrix to obtain the significant building block local surface feature matrix.

[0016] In the above-mentioned intelligent building block digital display method based on artificial intelligence technology, the feature significance score matrix of the local surface feature matrix of the building block is calculated, including: calculating the reciprocal of the exponential function value with natural constant as the base of each eigenvalue of the local surface feature matrix of the building block to obtain an indexed local surface feature matrix of the building block; calculating the cumulative sum of each eigenvalue of the local surface feature matrix of the indexed building block to obtain the global feature strength of the local surface feature matrix of the building block; calculating the addition result of the global feature strength and one, and dividing the eigenvalues ​​of each position of the local surface feature matrix of the indexed building block by the addition result to obtain the feature significance score matrix.

[0017] In the above-mentioned intelligent building block digital display method based on artificial intelligence technology, the multiple significant building block local surface feature matrices are associated with each other to obtain a building block local adjustment associated feature map, including: arranging the multiple significant building block local surface feature matrices into feature tensors and then passing them through a building block local feature association modeling module based on a three-dimensional convolutional neural network model to obtain the building block local adjustment associated feature map.

[0018] In the above-mentioned intelligent building block digital display method based on artificial intelligence technology, based on the local adjustment of the building block associated feature map, a three-dimensional digital building block is generated, including: inputting the local adjustment of the building block associated feature map into a digital building block generator based on a diffusion model to obtain the generated three-dimensional digital building block.

[0019] According to another aspect of the present application, there is provided a smart building block digital display system based on artificial intelligence technology, which includes:

[0020] A multi-view image acquisition module for a physical building block, used to acquire multiple local view images of the physical building block to be displayed collected by a camera;

[0021] A building block local surface feature extraction module, used for performing image feature extraction and feature manifestation processing on each of the multiple local view images to obtain multiple significant building block local surface feature matrices;

[0022] A building block local surface feature association modeling module, used for performing association modeling on the plurality of significant building block local surface feature matrices to obtain a building block local adjustment association feature map;

[0023] A three-dimensional digital building block generation module, used to locally adjust the associated feature map based on the building block to generate a three-dimensional digital building block;

[0024] The digital building block display module is used to display the generated three-dimensional digital building blocks.

[0025] Compared with the prior art, the method and system for digital display of intelligent building blocks based on artificial intelligence technology provided by the present application uses computer vision technology based on deep learning to process multiple local perspective images of physical building blocks to mine the local surface features of the building blocks at various perspectives, and then performs three-dimensional correlation modeling on the local surface features of the building blocks at multiple perspectives to obtain the global structural information of the physical building blocks, thereby generating a corresponding three-dimensional digital building block model. In this way, the rapid digital display of physical building blocks can be achieved, and the accuracy and efficiency of physical building block modeling can be effectively improved, thereby realizing the wide dissemination and sharing of building block ideas. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0027] Figure 1 The present invention is a flowchart of a method for digitally displaying smart building blocks based on artificial intelligence technology according to an embodiment of the present application.

[0028] Figure 2 Schematic diagram of the architecture of the digital display method of smart building blocks based on artificial intelligence technology according to an embodiment of the present application.

[0029] Figure 3 The present invention is a flowchart of performing image feature extraction and feature manifestation processing on each of the multiple local perspective images in the smart building block digital display method based on artificial intelligence technology according to an embodiment of the present application to obtain multiple significant building block local surface feature matrices.

[0030] Figure 4A flowchart of inputting each of the multiple building block local surface feature matrices into a content differentiation module based on a spatial mask to obtain the multiple significant building block local surface feature matrices in the smart building block digital display method based on artificial intelligence technology according to an embodiment of the present application.

[0031] Figure 5 4 is a block diagram of a smart building block digital display system based on artificial intelligence technology according to an embodiment of the present application. DETAILED DESCRIPTION

[0032] Below, the embodiments of the present application will be described in more detail in conjunction with the accompanying drawings, and the above and other purposes, features and advantages of the present application will become more apparent. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the example embodiments described herein.

[0033] As mentioned in the above background technology, the existing digital building block display technology mainly relies on three-dimensional modeling software to build digital models of building blocks through manual operation. However, this manual modeling method not only consumes a lot of time and human resources, but also has low efficiency. Moreover, when faced with complex building block works, its modeling accuracy and efficiency often cannot reach the expected standards, and it is difficult to meet the needs of practical applications. In view of the above technical problems, the technical concept of the present application is to use computer vision technology based on deep learning to perform image processing on multiple local perspective images of physical building blocks to dig out the local surface features of the building blocks under various perspectives, and then perform three-dimensional correlation modeling on the local surface features of the building blocks from multiple perspectives to obtain the global structural information of the physical building blocks, thereby generating corresponding three-dimensional digital building block models. In this way, the rapid digital display of physical building blocks can be realized, and the accuracy and efficiency of physical building block modeling can be effectively improved, thereby realizing the wide dissemination and sharing of building block creativity.

[0034] Figure 1 The present invention is a flowchart of a method for digitally displaying smart building blocks based on artificial intelligence technology according to an embodiment of the present application. Figure 2 FIG. 1 is a schematic diagram of the architecture of a method for digitally displaying smart building blocks based on artificial intelligence technology according to an embodiment of the present application. Figure 1 and Figure 2As shown, according to the embodiment of the present application, the method for digital display of smart building blocks based on artificial intelligence technology includes the following steps: S110, acquiring multiple local perspective images of the physical building blocks to be displayed captured by a camera; S120, performing image feature extraction and feature manifestation processing on each of the multiple local perspective images to obtain multiple significant building block local surface feature matrices; S130, performing association modeling on the multiple significant building block local surface feature matrices to obtain a building block local adjustment association feature map; S140, generating a three-dimensional digital building block based on the building block local adjustment association feature map; S150, displaying the generated three-dimensional digital building block.

[0035] In the above-mentioned intelligent building block digital display method based on artificial intelligence technology, the step S110 obtains multiple local perspective images of the real building block to be displayed collected by the camera. It should be understood that the real building block is a three-dimensional solid object, and its shape, color, texture and other features will be different under different perspectives. Under a single perspective, some parts of the building block will be blocked by other parts, resulting in the inability to fully capture all the features of the building block, affecting the accuracy of three-dimensional modeling. Therefore, in the technical solution of the present application, the surface features of the building block in all directions are fully obtained by shooting local images of the real building block from multiple perspectives. In this way, the building block image information of multiple perspectives can be used to complement and verify each other, understand the various parts of the building block and their mutual relationship, so as to infer the three-dimensional structure of the object, more accurately estimate the spatial position, shape and size of the building block, avoid information loss or misunderstanding caused by a single perspective, thereby improving the accuracy and authenticity of the digital three-dimensional modeling of the real building block.

[0036] In the above-mentioned method for digital display of intelligent building blocks based on artificial intelligence technology, the step S120 is to extract image features and perform feature manifestation processing on each of the multiple local perspective images to obtain multiple significant building block local surface feature matrices. Figure 3 The flowchart is a method for digitally displaying smart building blocks based on artificial intelligence technology according to an embodiment of the present application, in which each of the multiple local perspective images is subjected to image feature extraction and feature manifestation processing to obtain multiple significant building block local surface feature matrices. Figure 3 As shown, the step S120 includes: S121, performing image preprocessing on the multiple local perspective images to obtain multiple preprocessed local perspective images; S122, respectively extracting the block surface features of each of the multiple preprocessed local perspective images to obtain multiple block local surface feature matrices; S123, respectively performing image content attention differentiation on each of the multiple block local surface feature matrices to obtain the multiple salient block local surface feature matrices.

[0037] Specifically, the step S121 performs image preprocessing on the multiple local perspective images to obtain multiple preprocessed local perspective images. It should be understood that, considering the influence of factors such as lighting, camera angle, focal length, etc. in the actual shooting process, the captured local perspective images may have problems such as noise, blur, distortion, etc., which in turn affect the accuracy of subsequent feature extraction and modeling. Therefore, in the technical solution of the present application, image preprocessing technology is further used to perform denoising, detail enhancement and distortion correction on multiple local perspective images to improve image quality and lay a good foundation for subsequent feature extraction and modeling. In an embodiment of the present application, image filtering, histogram equalization and other technologies can be used for image preprocessing to eliminate noise and distortion in the image, enhance the detail information of the image, and make the image clearer and more accurate.

[0038] Specifically, the step S122 extracts the block surface features of each of the multiple preprocessed local perspective images to obtain multiple block local surface feature matrices. In a specific example of the present application, the processing method for extracting the block surface features of each of the multiple preprocessed local perspective images is to pass each of the multiple preprocessed local perspective images through a block surface feature extractor based on a dilated convolutional neural network model to obtain the multiple block local surface feature matrices. It should be understood that the dilated convolutional neural network (DCNN) increases the receptive field while maintaining the resolution of the feature map by introducing the concept of dilation rate, allowing the network to capture a wider range of contextual information without increasing the number of parameters and the amount of calculation, thereby reducing the information loss problem that may occur during the feature extraction process. Therefore, based on the characteristics of the dilated convolution, the block surface feature extractor can effectively extract the basic local surface detail features of the blocks such as color, texture, shape, etc. at various viewing angles, as well as contextual information such as the spatial structural relationship of each part of the blocks, providing a rich and accurate information basis for subsequent three-dimensional digital block modeling.

[0039] Specifically, the step S123 performs image content attention distinction on each of the local surface feature matrices of the multiple building blocks to obtain the multiple significant local surface feature matrices of the building blocks. In a specific example of the present application, the processing method of performing image content attention distinction on each of the local surface feature matrices of the multiple building blocks is to input each of the local surface feature matrices of the multiple building blocks into a content distinction module based on a spatial mask to obtain the multiple significant local surface feature matrices of the building blocks. It should be understood that in the multiple local perspective images of the building block entity, in addition to the features of the building blocks themselves, some background noise or other interference factors may also be included, which may interfere with the subsequent three-dimensional modeling. Therefore, in the technical solution of the present application, a content distinction module based on a spatial mask is further introduced to process each of the local surface feature matrices of the multiple building blocks in the local surface feature matrices of the building blocks, accurately distinguish and remove irrelevant features therein, so as to ensure the accuracy and effectiveness of data processing. Specifically, the content differentiation module based on the spatial mask calculates the attention weight of each position in the local surface feature matrix of the building block, generates a corresponding spatial mask based on the feature content, and then uses the generated spatial mask to perform feature screening on the original feature matrix, shields or ignores the non-building block area features in the feature matrix, and only retains important features related to the building blocks, thereby significantly improving the accuracy and efficiency of subsequent three-dimensional modeling, reducing the interference of irrelevant features on the modeling results, and helping to reduce the amount of calculation and improve processing efficiency.

[0040] Figure 4 The present invention provides a flowchart of inputting each of the plurality of local surface feature matrices of building blocks into a content differentiation module based on a spatial mask to obtain the plurality of significant local surface feature matrices of building blocks in the method for digital display of intelligent building blocks based on artificial intelligence technology according to an embodiment of the present invention. Figure 4 As shown, the step S123 includes: S1231, calculating the feature significance score matrix of the building block local surface feature matrix; S1232, masking the feature significance score matrix based on a predetermined threshold to obtain a building block local surface feature attention mask matrix; S1233, using the building block local surface feature attention mask matrix as a weight matrix, calculating the position point multiplication of the building block local surface feature matrix and the building block local surface feature attention mask matrix to obtain the significant building block local surface feature matrix.

[0041] Specifically, the step S1231 includes: calculating the reciprocal of the exponential function value with natural constant as the base of each eigenvalue of the local surface feature matrix of the building block to obtain an indexed local surface feature matrix of the building block; calculating the cumulative sum of each eigenvalue of the local surface feature matrix of the building block to obtain the global feature strength of the local surface feature matrix of the building block; calculating the addition result of the global feature strength and one, and dividing the eigenvalue of each position of the local surface feature matrix of the indexed building block by the addition result to obtain the feature significance score matrix.

[0042] That is, the step S123 includes: processing the local surface feature matrix of the building block with the following spatial mask formula to obtain the local surface feature matrix of the significant building block; wherein the spatial mask formula is:

[0043]

[0044]

[0045] M' i,j =M(i,h)⊙A(i,j)

[0046] Where A(i,h) is the eigenvalue of the position (i,j) in the local surface feature matrix of the building block, mask(·) represents masking, ε represents a predetermined threshold, M(i,j) is the eigenvalue of the position (i,j) in the local surface feature attention mask matrix of the building block, ⊙ represents the point multiplication by position, and M' i,j is the eigenvalue of the (i, j) position of the local surface feature matrix of the significant building block.

[0047] In the above-mentioned intelligent building block digital display method based on artificial intelligence technology, the step S130 performs association modeling on the multiple significant building block local surface feature matrices to obtain the building block local adjustment association feature map. In a specific example of the present application, the processing method of the association modeling of the multiple significant building block local surface feature matrices is to arrange the multiple significant building block local surface feature matrices into feature tensors and then pass the building block local feature association modeling module based on the three-dimensional convolutional neural network model to obtain the building block local adjustment association feature map. It should be understood that since the digital display of the building block needs to construct its global structural information, it is not possible to only consider the building block local features of a single perspective. Therefore, in the technical solution of the present application, the multiple significant building block local surface feature matrices are further arranged as feature tensors to integrate the building block surface feature information under different perspectives. Furthermore, the integrated feature tensor is processed using the building block local feature association modeling module based on the three-dimensional convolutional neural network model to capture the spatial correlation and mutual dependence between the local features of the building blocks under multiple perspectives. Specifically, the 3D Convolutional Neural Network (3DCNN) can effectively capture the spatial correlation and interdependence between the local features of the building blocks from multiple perspectives by using a 3D convolution kernel in the convolution layer to perform convolution operations in the three directions (width, height and depth) of the feature tensor, fully understand the overall shape and structure of the building blocks, and thus construct the complete structural information of the building blocks in three-dimensional space, so as to obtain the local adjustment correlation feature map of the building blocks, providing strong support for the subsequent generation of digital building blocks.

[0048] In the above-mentioned intelligent building block digital display method based on artificial intelligence technology, the step S140, based on the local adjustment of the associated feature map of the building block, obtains the generated three-dimensional digital building block. In a specific example of the present application, the implementation method of obtaining the generated three-dimensional digital building block based on the local adjustment of the associated feature map of the building block is to input the local adjustment of the associated feature map of the building block into a digital building block generator based on a diffusion model to obtain the generated three-dimensional digital building block. That is, in order to ensure that the generated three-dimensional model of the digital building block not only conforms to the shape and structure of the real building block, but also can maintain sufficient smoothness and continuity, in the technical solution of the present application, a digital building block generator based on a diffusion model is introduced to simulate the generation process of the digital building block, and high-quality three-dimensional digital building blocks are generated by gradually removing noise and optimizing details. Specifically, the diffusion model is a generative model that gradually recovers real data from noise by simulating the diffusion process of data distribution, and has strong generation ability and flexibility. Here, the digital building block generator based on the diffusion model maps the local adjustment associated feature map of the building block to a high-dimensional latent space, simulates the diffusion process of the building block data in the latent space through a step-by-step iterative approach, and gradually recovers the real building block data from the noise until high-quality three-dimensional digital building blocks are generated.

[0049] In the technical solution of the present application, the multiple local surface feature matrices of the building blocks represent the image semantic features in the local image semantic space domain under the global image semantic space domain for each local perspective image. When distinguishing the content of the spatial mask, the difference in the distribution of image semantic features in the local image semantic space domain will cause the difference in the weight of the spatial semantic association mask, resulting in the problem of insufficient distribution aggregation of the image semantic feature association modeling in the local adjustment association feature map of the building block, thereby affecting the generation efficiency of the digital building block generator based on the diffusion model and affecting the generation quality of the generated three-dimensional digital building blocks.

[0050] Based on this, in a preferred embodiment of the present application, the locally adjusted associated feature map of the building blocks is input into a digital building block generator based on a diffusion model to obtain a three-dimensional digital building block, including: multiplying the locally adjusted associated feature map of the building blocks by the scale of the locally adjusted associated feature map of the building blocks and the square root of the scale of the locally adjusted associated feature map of the building blocks to obtain a locally adjusted associated full-width representation map of the building blocks and a locally adjusted associated half-width representation map of the building blocks, wherein the scale of the locally adjusted associated feature map of the building blocks is the width of the feature matrix of the locally adjusted associated feature map of the building blocks multiplied by the height and then multiplied by the number of channels of the locally adjusted associated feature map of the building blocks; performing point subtraction on the sum of the absolute values ​​of each feature value of the locally adjusted associated full-width representation map of the building blocks and the locally adjusted associated feature map of the building blocks, and calculating the square root of the absolute value of each position of the point subtraction result to obtain the semantic change of the locally adjusted associated full-width of the building blocks. Figure; perform point subtraction on the square root of the sum of the squares of the eigenvalues ​​of the local adjustment associated half-width representation map of the building block and the local adjustment associated feature map of the building block, and calculate the square root of the absolute value of each position of the point subtraction result to obtain the local adjustment associated half-width semantic change map of the building block; respectively calculate the logarithm with base 2 of each eigenvalue of the local adjustment associated full-width semantic change map of the building block and the local adjustment associated half-width semantic change map of the building block to obtain the local adjustment associated full-width semantic change information map of the building block and the local adjustment associated half-width semantic change information map of the building block; calculate the weighted sum of the local adjustment associated full-width semantic change information map of the building block and the local adjustment associated half-width semantic change information map of the building block with a balanced hyperparameter as the weight to obtain the optimized local adjustment associated feature map of the building block; input the optimized local adjustment associated feature map of the building block into a digital building block generator based on a diffusion model to generate a three-dimensional digital building block.

[0051] That is, in this preferred embodiment, the local adjustment associated feature map of the building blocks is processed by the following optimization formula to obtain the optimized local adjustment associated feature map of the building blocks, wherein the optimization formula is:

[0052]

[0053] Wherein, F represents the local adjustment associated feature map of the building block, L represents the scale of the local adjustment associated feature map of the building block, ||·||1 represents the first norm of the feature map, ||·||2 represents the second norm of the feature map, |·|⊙ 1 / 2 Represents the square root of the absolute value of each eigenvalue of the feature map, log represents the logarithmic function operation with base 2, ω is the balance hyperparameter, ⊙ represents the dot product, Indicates point reduction, represents point addition, and F' represents the local adjustment associated feature map of the optimized building blocks.

[0054] Here, the local adjustment associated feature map of the building block is taken as a feature set, wherein the change semantics in units of the feature value of the local adjustment associated feature map of the building block is expressed, in order to dynamically aggregate the semantic set composed of different change semantics of the local adjustment associated feature map of the building block without ignoring the individual semantic change information, the individual features of the local adjustment associated feature map of the building block and the collective expression of the aggregation scale of the local adjustment associated feature map of the building block are taken as the feature full amplitude and half amplitude, and the low-rank negative correlation of different dimensions of the overall semantics of the feature set of the local adjustment associated feature map of the building block is taken as the phase and scaled, so as to dynamically adjust the change relationship of the semantic content of the local adjustment associated feature map of the building block, thereby improving the aggregation of the semantic information expression of the overall feature set of the local adjustment associated feature map of the building block, improving the generation efficiency of the local adjustment associated feature map of the building block through the digital building block generator based on the diffusion model, and improving the generation quality of the generated three-dimensional digital building blocks.

[0055] In the above-mentioned intelligent building block digital display method based on artificial intelligence technology, the step S150 displays the generated three-dimensional digital building blocks. That is, the generated three-dimensional digital building blocks are visualized so that users can intuitively understand the real shape and structure of the physical building blocks in three-dimensional space, feel their three-dimensional sense and spatial sense, and facilitate users to observe and analyze from multiple angles and in all directions. In this way, the communication and interactive experience of building block enthusiasts can be greatly enriched, and the widespread dissemination and development of building block culture can be promoted.

[0056] In summary, the method for digital display of intelligent building blocks based on artificial intelligence technology according to the embodiment of the present application is explained, which uses computer vision technology based on deep learning to process multiple local perspective images of physical building blocks to mine the local surface features of the building blocks at various perspectives, and then performs three-dimensional correlation modeling on the local surface features of the building blocks at multiple perspectives to obtain the global structural information of the physical building blocks, thereby generating a corresponding three-dimensional digital building block model. In this way, the rapid digital display of physical building blocks can be achieved, and the accuracy and efficiency of physical building block modeling can be effectively improved, thereby realizing the wide dissemination and sharing of building block creativity.

[0057] Figure 5 FIG. 1 is a block diagram of a digital display system for intelligent building blocks based on artificial intelligence technology according to an embodiment of the present application. Figure 5As shown, according to the embodiment of the present application, the intelligent building block digital display system 100 based on artificial intelligence technology includes: a real-world building block multi-view image acquisition module 110, which is used to acquire multiple local view images of the real-world building block to be displayed collected by a camera; a building block local surface feature extraction module 120, which is used to perform image feature extraction and feature manifestation processing on each local view image of the multiple local view images to obtain multiple significant building block local surface feature matrices; a building block local surface feature association modeling module 130, which is used to perform association modeling on the multiple significant building block local surface feature matrices to obtain a building block local adjustment association feature map; a three-dimensional digital building block generation module 140, which is used to generate a three-dimensional digital building block based on the building block local adjustment association feature map; a digital building block display module 150, which is used to display the generated three-dimensional digital building block.

[0058] Here, those skilled in the art can understand that the specific operations of each module in the above-mentioned intelligent building block digital display system based on artificial intelligence technology have been referred to above. Figures 1 to 4 The method for digital display of intelligent building blocks based on artificial intelligence technology has been introduced in detail, and therefore, its repeated description will be omitted.

[0059] The basic principle of the present invention is described above in conjunction with specific embodiments. However, it should be pointed out that the advantages, strengths, effects, etc. mentioned in the present invention are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. must be possessed by each embodiment of the present invention. In addition, the specific details of the above embodiments are only for the purpose of illustration and facilitation of understanding, rather than limitation, and the above details do not limit the present invention to being implemented by adopting the above specific details.

[0060] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided by the present invention, it should be understood that the disclosed system and method can be implemented in other ways. For example, the system embodiment described above is only schematic. For example, the module division is only a logical function division, and there may be other division methods in actual implementation. The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0061] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference to a figure in a claim should not be considered as limiting the claim to which it relates.

[0062] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units stated in the system claims can also be implemented by one unit through software or hardware.

[0063] Finally, it should be noted that the above description has been given for the purpose of illustration and description. In addition, the above embodiments are only used to illustrate the technical solution of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A digital display method of intelligent building blocks based on artificial intelligence technology, characterized in that: include: Acquire multiple local perspective images of the physical building block to be displayed captured by a camera; Performing image feature extraction and feature manifestation processing on each of the multiple local perspective images to obtain multiple salient building block local surface feature matrices; Performing association modeling on the plurality of significant building block local surface feature matrices to obtain a building block local adjustment association feature map; Based on the local adjustment of the associated feature map of the building blocks, a three-dimensional digital building block is generated; Displaying the generated three-dimensional digital building blocks; The image feature extraction and feature manifestation processing are performed on each of the multiple local perspective images to obtain multiple significant building block local surface feature matrices, including: Performing image preprocessing on the multiple local perspective images to obtain multiple preprocessed local perspective images; Extracting the building block surface features of each preprocessed local perspective image in the plurality of preprocessed local perspective images respectively to obtain a plurality of building block local surface feature matrices; Performing image content attention differentiation on each of the plurality of building block local surface feature matrices to obtain the plurality of salient building block local surface feature matrices; The method of performing image content attention differentiation on each of the plurality of building block local surface feature matrices to obtain the plurality of salient building block local surface feature matrices comprises: Inputting each of the plurality of building block local surface feature matrices into a content differentiation module based on a spatial mask to obtain the plurality of salient building block local surface feature matrices; The method of inputting each of the plurality of building block local surface feature matrices into a content differentiation module based on a spatial mask to obtain the plurality of salient building block local surface feature matrices comprises: Calculating a feature significance score matrix of the local surface feature matrix of the building block; Performing masking processing on the feature significance score matrix based on a predetermined threshold to obtain a building block local surface feature attention mask matrix; The building block local surface feature attention mask matrix is ​​used as a weight matrix, and the position point multiplication of the building block local surface feature matrix and the building block local surface feature attention mask matrix is ​​calculated to obtain the significant building block local surface feature matrix.

2. The method for digital display of intelligent building blocks based on artificial intelligence technology according to claim 1, characterized in that: Extracting the building block surface features of each preprocessed local perspective image in the plurality of preprocessed local perspective images respectively to obtain a plurality of building block local surface feature matrices, including: Each of the multiple preprocessed local perspective images is passed through a building block surface feature extractor based on a hole convolutional neural network model to obtain the multiple building block local surface feature matrices.

3. The method for digital display of intelligent building blocks based on artificial intelligence technology according to claim 2, characterized in that: Calculating a feature significance score matrix of the local surface feature matrix of the building block, comprising: Calculating the reciprocal of the exponential function value with the natural constant as the base of each eigenvalue of the local surface feature matrix of the building block to obtain an exponential building block local surface feature matrix; Calculating the cumulative sum of each eigenvalue of the indexed building block local surface feature matrix to obtain the global feature strength of the building block local surface feature matrix; The sum of the global feature strength and one is calculated, and the feature values ​​of each position of the indexed building block local surface feature matrix are divided by the sum to obtain the feature significance score matrix.

4. The method for digital display of intelligent building blocks based on artificial intelligence technology according to claim 3 is characterized in that: Performing association modeling on the plurality of significant building block local surface feature matrices to obtain a building block local adjustment association feature map, including: After arranging the multiple significant building block local surface feature matrices into feature tensors, a building block local feature association modeling module based on a three-dimensional convolutional neural network model is used to obtain the building block local adjustment association feature map.

5. The method for digital display of intelligent building blocks based on artificial intelligence technology according to claim 4, characterized in that: Based on the local adjustment of the associated feature map of the building blocks, a three-dimensional digital building block is generated, including: The local adjustment associated feature map of the building block is input into a digital building block generator based on a diffusion model to obtain the generated three-dimensional digital building block.

6. A digital display system of intelligent building blocks based on artificial intelligence technology, using the digital display method of intelligent building blocks based on artificial intelligence technology according to claim 1, characterized in that: include: A multi-view image acquisition module for a physical building block, used to acquire multiple local view images of the physical building block to be displayed collected by a camera; A building block local surface feature extraction module, used for performing image feature extraction and feature manifestation processing on each of the multiple local view images to obtain multiple significant building block local surface feature matrices; A building block local surface feature association modeling module, used for performing association modeling on the plurality of significant building block local surface feature matrices to obtain a building block local adjustment association feature map; A three-dimensional digital building block generation module, used to locally adjust the associated feature map based on the building block to generate a three-dimensional digital building block; The digital building block display module is used to display the generated three-dimensional digital building blocks.

Images