Texture synthesis method and system based on Maya 3D model data

Abstract

This invention discloses a texture synthesis method based on Maya 3D model data, belonging to the field of texture synthesis technology. The method includes the following steps: model creation, 3D model parameterization, texture matching, using a texture detection algorithm in the 3D model data texture synthesis system to extract texture structure features from a texture sample image, obtaining a texture structure feature image using the texture detection algorithm, texture synthesis, selecting a rectangular texture block from the texture sample image as the first texture block of the texture image, globally scanning the texture sample image to minimize the error in the overlapping portion between the selected texture block and the first texture block, comparing the structural feature images of the texture blocks to ensure the overlapping area structures match, and then cutting and stitching to obtain a complete texture image. This invention avoids the situation where the texture structure is completely mismatched due to pixel error values ​​in the overlapping area during texture block matching, and can synthesize more detailed texture images.

Description

Technical Field

[0001] This invention relates to the field of texture synthesis technology, specifically to a texture synthesis method and system based on Maya 3D model data. Background Technology

[0002] In the real world, the surfaces of objects have complex, rich and different texture details. With the development of the Internet, whether it is the game industry, the construction industry or the technology industry, there is a need to reproduce the details of realistic object models on the computer. The simulation of the texture details of the object surface plays a very important role in realistic image rendering technology. There are two main methods to simulate texture details on objects generated on the computer. The most common method is texture mapping. Texture synthesis was proposed to solve the problems of seam distortion in texture mapping.

[0003] However, current texture synthesis methods cannot meet the requirements of realistic rendering for objects with complex features using texture samples with single features. Furthermore, if only simple repeated mapping is performed, obvious texture seams and aliasing problems will occur at the texture boundaries. Pixel-matching-based multi-image texture synthesis has proposed a solution to this problem. However, this method is based on pixel matching, and each global scan can only synthesize one pixel. If the 3D model is a terrain model, the texture synthesis speed of this method will be very slow when the terrain scale is large. Moreover, the synthesis effect of textures with complex texture features is not good, and it cannot obtain detailed texture images. This results in the 3D model's texture details not being realistic enough after texture mapping, making the 3D model lack realism. In addition, when mapping the synthesized texture image onto the 3D model, the problem of stretching and deformation of the texture image when the 2D texture image is directly mapped onto the surface of the 3D model is not considered, causing texture deformation on the surface of the 3D model and losing realism, thus failing to render the 3D model well. Summary of the Invention

[0004] The purpose of this invention is to provide a texture synthesis method and system based on Maya 3D model data to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a texture synthesis method based on Maya 3D model data, comprising the following steps:

[0006] S1. Creating a Model: Use Maya to create a 3D model;

[0007] S2. 3D model parameterization: Obtain the data of the 3D solid model, parameterize the 3D solid model onto a 2D plane, so that the points on the 2D plane are mapped one by one to the points on the 3D model mesh, and maintain the consistency of the topological information on the 3D mesh with that on the 2D plane.

[0008] S3. Texture matching: In the texture synthesis system of 3D model data, the texture detection algorithm is used to extract the texture structure features of the texture sample image from the texture sample image. The texture detection algorithm retains the pixels with significant changes in gray value in the texture sample image and removes the rest of the pixels to obtain the texture structure feature image.

[0009] S4. Texture Synthesis: Randomly select a rectangular texture block from the texture sample image as the first texture block of the texture image. Then, scan the texture sample image globally to minimize the error of the overlapping part between the selected texture block and the first texture block. Compare the structural feature images of the two texture blocks to match the structure of the overlapping area. Then, randomly select a point on the overlapping area of ​​the two texture blocks as the best matching point. After that, find the best cutting line through the cutting algorithm, remove the overlapping area according to the cutting line, and stitch the texture blocks together. Repeat this process until a complete texture image is obtained.

[0010] S5. Texture Mapping: Maps the texture image obtained through texture synthesis in S4 onto the three-dimensional model.

[0011] Preferably, a texture synthesis system for 3D model data utilizes the aforementioned texture synthesis method, including a texture sample acquisition unit, a texture sample integration unit, a texture sample structure extraction unit, a texture synthesis unit, and a texture storage unit.

[0012] The texture sample extraction unit extracts the texture structure feature image of the texture sample image from the texture sample image using a texture detection algorithm. The transmitter of the texture sample extraction unit is electrically connected to the receiver of the texture synthesis unit. The texture sample extraction unit sends the extracted texture structure feature image and the texture sample image to the texture synthesis unit.

[0013] The texture synthesis unit is used to receive data from the texture sample structure extraction unit. The texture synthesis unit randomly selects a rectangular texture block from the texture sample image as the first texture block of the texture image. Then, it globally scans the texture sample image to select a second texture block, minimizing the error of the overlapping part between the selected texture block and the first texture block. The structural feature images of the two texture blocks are compared to match the structure of the overlapping area. Then, a point is randomly selected on the overlapping area of ​​the two texture blocks as the best matching point. After that, the best cutting line is found through a cutting algorithm. The overlapping area is removed according to the cutting line and the texture blocks are stitched together. The transmitting end of the texture synthesis unit is electrically connected to the receiving end of the texture storage unit.

[0014] Preferably, the texture sample acquisition unit is used to acquire texture sample images from captured images or hand-drawn images. The transmitting end of the texture sample acquisition unit is electrically connected to the receiving end of the texture sample integration unit, and the acquired texture sample images are sent to the texture sample integration unit.

[0015] Preferably, the texture sample integration unit is used to receive the texture sample image transmitted by the texture sample acquisition unit, and to remove duplicate and unclear texture sample images from the acquired texture sample image. The transmitting end of the texture sample integration unit is electrically connected to the receiving end of the texture sample extraction structure unit, and the texture sample integration unit sends the texture sample image after removing duplicates and unclear images to the texture sample extraction structure unit.

[0016] Preferably, the texture storage unit (5) is used to receive the texture-synthesized image from the texture synthesis unit (4) and store it.

[0017] Preferably, the texture synthesis unit further includes a texture image cutting module, which is used to find the optimal cutting line to cut and remove the overlapping area between two overlapping texture blocks using a cutting algorithm.

[0018] Preferably, the texture synthesis unit further includes a texture image stitching module, which is used to stitch together texture blocks after removing overlapping areas to obtain a complete texture image.

[0019] Preferably, the texture sample structure extraction unit further includes a texture detection module. The texture detection module extracts the texture structure feature image of the texture sample image from the texture sample image after removing duplicates and unclear elements using a texture detection algorithm. The texture detection algorithm is as follows:

[0020]

[0021] G1 = I * K1, G2 = I * K2

[0022] in, I represents the image being detected, i.e., the texture sample image.

[0023] Compared with the prior art, the beneficial effects of the present invention are:

[0024] This invention adds a texture detection algorithm, which can extract the feature structure of texture samples and compare the feature structure maps of two texture blocks during texture synthesis. This ensures that the structures of the overlapping areas of the two texture blocks match, avoiding the situation where the texture structures are completely mismatched due to the error values ​​of pixels in the overlapping areas during the texture block matching process. This allows for the synthesis of more detailed texture images. Furthermore, it speeds up the texture synthesis process when the model is a large terrain model. Additionally, the 3D model is meshed and parameterized, avoiding the situation where the texture image is stretched and deformed during texture mapping, which would cause the surface texture of the 3D model to be deformed. Attached Figure Description

[0025] Figure 1 This is a schematic diagram illustrating the steps provided in an embodiment of the present invention;

[0026] Figure 2 This is an overall structural diagram provided for an embodiment of the present invention;

[0027] Figure 3 This is a unit structure diagram of the texture sample structure extraction provided in an embodiment of the present invention;

[0028] Figure 4 This is a unit structure diagram of the texture synthesis unit provided in an embodiment of the present invention;

[0029] Figure 5 The diagram shows the structure of the texture sample extraction unit and the texture synthesis unit provided in the embodiments of the present invention.

[0030] In the diagram: 1. Texture sample acquisition unit; 2. Texture sample integration unit; 3. Texture sample structure extraction unit; 4. Texture synthesis unit; 5. Texture storage unit; 31. Texture detection module; 41. Texture image cropping module; 42. Texture image stitching module. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] Please see Figure 1-5This invention provides a technical solution: a texture synthesis method based on Maya 3D model data, comprising the following steps:

[0033] S1. Creating a Model: Use Maya to create a 3D model;

[0034] S2, 3D model parameterization: Obtain the data of the 3D solid model, parameterize the 3D solid model onto a 2D plane, so that the points on the 2D plane are mapped one by one to the points on the 3D model mesh, and keep the topological information on the 3D mesh consistent with that on the 2D plane, so as to avoid the problem of stretching and deformation of the texture image during subsequent mapping.

[0035] S3. Texture Matching: In the texture synthesis system of 3D model data, the texture detection algorithm is used to extract the texture structure features of the texture sample image from the texture sample image. The texture detection algorithm retains the pixels with significant changes in gray value in the texture sample image and removes the rest of the pixels to obtain the texture structure feature image. This avoids the situation where the texture structure is completely mismatched due to the error value of the pixels in the overlapping area during the texture block matching process.

[0036] S4. Texture Synthesis: Randomly select a rectangular texture block from the texture sample image as the first texture block of the texture image. Then, scan the texture sample image globally to minimize the error of the overlapping part between the selected texture block and the first texture block. Compare the structural feature images of the two texture blocks to match the structure of the overlapping area. Then, randomly select a point on the overlapping area of ​​the two texture blocks as the best matching point. After that, find the best cutting line through the cutting algorithm, remove the overlapping area according to the cutting line, and stitch the texture blocks together. Repeat this process until a complete texture image is obtained to avoid the problem of seam distortion in the texture image.

[0037] S5. Texture Mapping: Maps the texture image obtained through texture synthesis in S4 onto the three-dimensional model.

[0038] A texture synthesis system for 3D model data, which utilizes the aforementioned texture synthesis method, includes a texture sample acquisition unit 1, a texture sample integration unit 2, a texture sample structure extraction unit 3, a texture synthesis unit 4, and a texture storage unit 5.

[0039] The texture sample extraction unit 3 extracts the texture structure feature image of the texture sample image from the texture sample image using a texture detection algorithm. The transmitter of the texture sample extraction unit 3 is electrically connected to the receiver of the texture synthesis unit 4. The texture sample extraction unit 3 sends the extracted texture structure feature image and the texture sample image to the texture synthesis unit 4. This avoids the situation where the texture structure is completely mismatched due to the error value of the pixel in the overlapping area during the texture block matching process. It can synthesize a more detailed texture image, and it also speeds up the texture synthesis speed if the model is a very large terrain model.

[0040] The texture synthesis unit 4 receives data from the texture sample structure extraction unit 3. The texture synthesis unit 4 randomly selects a rectangular texture block from the texture sample image as the first texture block of the texture image. Then, it globally scans the texture sample image to select a second texture block, minimizing the error of the overlapping part between the selected texture block and the first texture block. The structural feature images of the two texture blocks are compared to match the structure of the overlapping area. Then, a point is randomly selected on the overlapping area of ​​the two texture blocks as the best matching point. After that, the best cutting line is found through the cutting algorithm. The overlapping area is removed according to the cutting line and the texture blocks are stitched together. The transmitting end of the texture synthesis unit 4 is electrically connected to the receiving end of the texture storage unit 5 to avoid the problem of seam distortion in the texture image.

[0041] The texture sample acquisition unit is used to acquire texture sample images from captured images or hand-drawn images. The transmitter of the texture sample acquisition unit 1 is electrically connected to the receiver of the texture sample integration unit 2, and the acquired texture sample image is sent to the texture sample integration unit 2, thus providing texture sample image data for subsequent texture synthesis.

[0042] The texture sample integration unit 2 is used to receive the texture sample image transmitted from the texture sample acquisition unit 1, and to remove duplicate and unclear texture sample images in the acquired texture sample image. The transmitting end of the texture sample integration unit 2 is electrically connected to the receiving end of the texture sample extraction structure unit 3. The texture sample integration unit 2 sends the texture sample image after removing duplicates and unclear images to the texture sample extraction structure unit 3, thereby removing duplicate and unclear texture sample images and avoiding the generation of low-resolution textures through texture synthesis.

[0043] The texture storage unit 5 is used to receive the texture-synthesized image from the texture synthesis unit 4 and store it.

[0044] The texture synthesis unit 4 also includes a texture image cutting module 41. The texture image cutting module 41 is used to find the best cutting line to cut two overlapping texture blocks and remove the overlapping area, so as to avoid the problem of seam distortion in the texture image.

[0045] The texture synthesis unit also includes a texture image stitching module 42, which is used to stitch together texture blocks after removing overlapping areas to obtain a complete texture image.

[0046] The sample texture structure extraction unit 3 also includes a texture detection module 31. The texture detection module 31 extracts the texture structure feature image of the texture sample image from the texture sample image after removing duplicates and unclear parts using a texture detection algorithm. The texture detection algorithm is as follows:

[0047]

[0048] G1 = I * K1, G2 = I * K2

[0049] in, I represents the image being detected, i.e., the texture sample image.

[0050] Working Principle: This invention creates a 3D model using Maya. After acquiring the 3D model data, it parameterizes the 3D model onto a 2D plane, mapping each point on the 2D plane to a point on the 3D model mesh. Then, in the texture synthesis system of the 3D model data, texture sample acquisition unit 1 acquires texture sample images from captured or hand-drawn images and sends these images to texture sample integration unit 2. Texture sample integration unit 2 removes duplicate and unclear texture sample images and then sends the integrated texture sample image to texture sample structure extraction unit 3. Texture sample structure extraction unit 3 uses a texture detection algorithm to obtain the texture structure feature map and then integrates the texture structure. The structural feature map is sent to the texture synthesis unit 4. The texture synthesis unit 4 randomly selects a rectangular texture block from the texture sample image as the first texture block of the texture image. Then, it globally scans the texture sample image to select a second texture block, minimizing the error of the overlapping part between the selected texture block and the first texture block. The structural feature images of the two texture blocks are compared to match the structure of the overlapping area. Then, a point is randomly selected on the overlapping area of ​​the two texture blocks as the best matching point. After that, the best cutting line is found through the cutting algorithm. The overlapping area is removed according to the cutting line and the texture blocks are stitched together. The resulting texture image is sent to the texture storage unit 5 for storage. Finally, the texture image obtained by texture synthesis is mapped onto the three-dimensional model.

[0051] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0052] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for texture synthesis based on Maya-based 3D model data, characterized by, The method comprises the following steps: S1, model creation: creating a three-dimensional model using Maya; S2, three-dimensional model parameterization: obtaining data of the three-dimensional model, parameterizing the three-dimensional model to a two-dimensional plane, mapping points on the two-dimensional plane to points on the three-dimensional model grid one by one, and keeping the topological information on the three-dimensional grid consistent with that on the two-dimensional plane; S3, texture matching: using a texture detection algorithm in a texture synthesis system of 3D model data to extract texture structure features of the texture sample image from the texture sample image, using the texture detection algorithm to retain points with significant changes in gray value in the pixel points in the texture sample image and to remove the remaining pixel points, to obtain a texture structure feature image; S4, texture synthesis: randomly selecting a rectangular texture block from the texture sample image as the first texture block of the texture image, then globally scanning the texture sample image to minimize the error of the overlapping part between the selected texture block and the first texture block, comparing the structure feature images of the two texture blocks to match the structure of the overlapping area, then randomly selecting a point on the overlapping area of the two texture blocks as the best matching point, then finding the best cutting line through a cutting algorithm, removing the overlapping area according to the cutting line and stitching the texture blocks, and repeating the process until a complete texture image is obtained; S5, texture mapping: mapping the texture image obtained by texture synthesis in S4 to the three-dimensional model.

2. A texture synthesis system of 3D model data, which applies the texture synthesis method of claim 1, characterized in that: It comprises a texture sample acquisition unit (1), a texture sample integration unit (2), a texture sample structure extraction unit (3), a texture synthesis unit (4), and a texture storage unit (5): The texture sample structure extraction unit (3) extracts a texture structure feature image of the texture sample image from the texture sample image through a texture detection algorithm, the transmitting end of the texture sample structure extraction unit (3) is electrically connected with the receiving end of the texture synthesis unit (4), and the texture sample structure extraction unit (3) sends the extracted texture structure feature image and the texture sample image to the texture synthesis unit (4); The texture synthesis unit (4) is used for receiving data transmitted by the texture sample structure extraction unit (3), randomly selects a rectangular texture block from the texture sample image as the first texture block of the texture image, then globally scans the texture sample image to select a second texture block, minimizes the error of the overlapping part between the selected texture block and the first texture block, compares the structure feature images of the two texture blocks to match the structure of the overlapping area, then randomly selects a point on the overlapping area of the two texture blocks as the best matching point, then finds the best cutting line through a cutting algorithm, removes the overlapping area according to the cutting line and stitches the texture blocks, and the transmitting end of the texture synthesis unit (4) is electrically connected with the receiving end of the texture storage unit (5).

3. The texture synthesis system of 3D model data according to claim 2, characterized in that: The acquisition texture sample unit (1) is used for acquiring a texture sample image from a photographed image or a hand-drawn image, the transmitting end of the acquisition texture sample unit (1) is electrically connected with the receiving end of the texture sample integration unit (2), and the acquired texture sample image is sent to the texture sample integration unit (2).

4. The texture synthesis system of 3D model data according to claim 2, characterized in that: The texture sample integration unit (2) is used for receiving the texture sample image transmitted by the acquisition texture sample unit (1), and removing repeated and unclear texture sample images in the acquired texture sample image, the transmitting end of the texture sample integration unit (2) is electrically connected with the receiving end of the extraction texture sample structure unit (3), and the texture sample integration unit (2) sends the texture sample image after removing the repeated and unclear texture sample images to the extraction texture sample structure unit (3).

5. The texture synthesis system of 3D model data according to claim 2, characterized in that: The texture storage unit (5) is used for receiving the image after texture synthesis transmitted by the texture synthesis unit (4) and storing the image.

6. The texture synthesis system of 3D model data according to claim 2, characterized in that: The texture synthesis unit (4) further comprises a texture image cutting module (41), the texture image cutting module (41) is used for cutting two overlapping texture blocks to remove the overlapping area by finding the best cutting line through a cutting algorithm.

7. The texture synthesis system of 3D model data according to claim 2, characterized in that: The texture synthesis unit (4) further comprises a texture image splicing module (42), the texture image splicing module (42) is used for splicing the texture blocks after removing the overlapping area to obtain a complete texture image.

8. The texture synthesis system of 3D model data according to claim 2, characterized in that: The extraction texture sample structure unit (3) further comprises a texture detection module (31), the texture detection module (31) extracts a texture structure feature image of the texture sample image from the texture sample image after removing the repeated and unclear texture sample images through a texture detection algorithm.

Images