Model generation method and device, electronic equipment and storage medium

Abstract

The invention provides a model generation method and device, electronic equipment and a storage medium. The method comprises the steps that a target image is generated through a preset cue word, and the target image comprises a target texture; converting the target image into a first texture map suitable for the target model; fusing a target texture in the first texture map to an original texture map of the target model to obtain a target texture map; and projecting the target texture map to the target model to obtain the target model containing the target texture. According to the method, the target image containing the needed target texture is generated only through the cue word of the large language model, the target image can be at any angle, the target model containing the target texture can be obtained through texture mapping conversion and texture fusion modes, and the efficiency of newly adding, refining and modifying the texture of the target model is higher.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a method, device, electronic device and storage medium for generating a model. Background Art

[0002] In the process of model making, due to the need for refinement or image transformation, the skin texture of the model often needs to be modified. The existing technology usually requires the producer to modify the texture map of the model or directly modify the skin texture of the model, which has the technical problem of low efficiency. Summary of the invention

[0003] In view of this, an object of the present disclosure is to provide a model generation method, device, electronic device and storage medium to improve the efficiency of model generation.

[0004] In a first aspect, an embodiment of the present disclosure provides a method for generating a model, the method comprising: generating a target image by means of preset prompt words, wherein the target image comprises a target texture; converting the target image into a first texture map suitable for a target model; fusing the target texture in the first texture map into the original texture map of the target model to obtain a target texture map; and projecting the target texture map onto the target model to obtain a target model comprising the target texture.

[0005] In a second aspect, an embodiment of the present disclosure provides a device for generating a model, the device comprising: an acquisition module, used to generate a target image through a preset prompt word, the target image comprising a target texture; a conversion module, used to convert the target image into a first texture map suitable for a target model; a fusion module, used to fuse the target texture in the first texture map to the original texture map of the target model to obtain a target texture map; and a projection module, used to project the target texture map to the target model to obtain a target model comprising the target texture.

[0006] In a third aspect, an embodiment of the present disclosure provides an electronic device, including a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the above-mentioned model generation method.

[0007] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement the above-mentioned model generation method.

[0008] The embodiments of the present disclosure bring the following beneficial effects:

[0009] The generation method, device, electronic device and storage medium of the above-mentioned model only need to generate a target image containing the required target texture through the prompt words of the large language model. The target image can be at any angle, and the target model containing the target texture can be obtained through texture map conversion and texture fusion, so that the efficiency of adding, refining and modifying the texture of the target model is higher.

[0010] Other features and advantages of the present disclosure will be described in the following description, and partly become apparent from the description, or understood by practicing the present disclosure. The purpose and other advantages of the present disclosure are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0011] In order to make the above-mentioned objectives, features and advantages of the present disclosure more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the specific embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the specific embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative work.

[0013] Figure 1 is a flow chart of an embodiment of a method for generating a model in an embodiment of the present disclosure;

[0014] Figure 2 is a schematic diagram of a method for generating a model in an embodiment of the present disclosure;

[0015] Figure 3 is a schematic diagram of a method for generating a model in an embodiment of the present disclosure;

[0016] Figure 4 is a schematic diagram of a method for generating a model in an embodiment of the present disclosure;

[0017] Figure 5 is a schematic diagram of a method for generating a model in an embodiment of the present disclosure;

[0018] Figure 6 A schematic diagram of a model generation device provided in an embodiment of the present disclosure;

[0019] Figure 7 A schematic diagram of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present disclosure.

[0021] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0022] For ease of understanding, the specific process of the embodiment of the present disclosure is described below. Figure 1 , an embodiment of the method for generating a model in the embodiment of the present disclosure includes:

[0023] Step S10, generating a target image by using a preset prompt word, wherein the target image includes a target texture;

[0024] It is understandable that prompt is a concept in the field of artificial intelligence, especially natural language processing (NLP). In prompt engineering, the description of the task will be embedded in the input, and the language model will be trained through "prompt-based learning" to obtain a large language model. Among them, the prompt can be understood as a description of the task that the model needs to complete, which is used to input the large language model.

[0025] In this embodiment, a target image is first generated by preset prompt words, wherein the target image includes a target texture. For example, assuming that the target model is a person model and the target texture is freckles, then by generating an image of a person with freckles on his face as the target image, the target model can obtain the freckles in the target image, so that the required texture can be obtained without freckle making for the texture map of the target model, thereby improving the generation efficiency of the model.

[0026] In one embodiment, the target texture is a local texture on the surface of the target model, a texture that does not affect the structure of the target model, and a texture that can conform to the structure of the target model. As an example but not a limitation, when the target model is a person model, the target texture can be local texture features such as freckles, wrinkles, scars, pores, etc. on the face of the person, or it can be a texture that conforms to the structure of the target model, such as eyeballs, lips, eyebrows, etc., which have the same morphology / structure as the target model but a different single color, and are not specifically limited here.

[0027] In one embodiment, the object in the target image has the same structure as the target model. For example, if the target model is a human model, then the object in the target image is also a human, and the target texture is the texture of the human. In a more preferred case, the target model is a female model, and the target image is also a female image, which can achieve a better texture adaptation effect. For another example, assuming that the target model is an animal, such as a rabbit, then the object in the target image is a rabbit of the same species / structure as the target model, which is not specifically limited here.

[0028] It should be noted that when generating a target image, there is no need to limit the angle of the object in the target image, as long as the target texture can be fully presented. For example, if the target object is a person and the target texture is freckles on the person's face, then the face of the person in the target image can be facing any angle, as long as the freckles on the person's face can be fully presented.

[0029] Step S20, converting the target image into a first texture map suitable for the target model;

[0030] It can be understood that a map is an image used to add texture to the surface of a three-dimensional model, which can simulate the appearance of the material, such as skin color, texture, and details, etc., to make the model look more realistic. The map is associated with the surface coordinates of the model through UV mapping to ensure that the map is correctly covered on the model, where U and V represent the horizontal and vertical axes of the map respectively.

[0031] Based on this, this embodiment converts the target image into a first texture map suitable for the target model, which actually establishes a UV mapping relationship between the target image and the target model or a model with the same structure / architecture / skeleton structure as the target model, and records it in the texture map, thereby obtaining the first texture map.

[0032] In one embodiment, the target image can be converted into a first texture map suitable for the target model through an artificial intelligence model. The training samples of the artificial intelligence model include a two-dimensional image of the target object and a texture map of the target object model. The artificial intelligence model trained by these training samples can be used to convert the two-dimensional image into a texture map. The specific training process is not described here.

[0033] In one implementation, the first texture map corresponding to the target image may also be generated by a texture generation tool, for example, a texture generator (Texture Builder) in the film and television synthesis software Nuke, which is not specifically limited here.

[0034] Step S30, merging the target texture in the first texture map into the original texture map of the target model to obtain a target texture map;

[0035] In this embodiment, after obtaining the first texture map suitable for the target model, the target texture in the first texture map can be fused into the original texture map of the target model, thereby obtaining the target texture map containing the target texture, which is used as the texture rendering of the target model. It should be noted that the target model itself corresponds to a texture map, that is, the original texture map. The embodiment of the present disclosure fuses the target texture on the basis of the original texture map of the target model, making the method of adding / modifying the local texture of the model more convenient.

[0036] It is understandable that, since the first texture map is already a texture map suitable for the target model, that is, the first texture map is a texture map that can fit the target model, and has the same or similar UV mapping coordinates as the original texture map of the target model, the position of the target texture in the first texture map is the same as the position of the original texture map that needs to be merged into the target model. In one embodiment, according to the position of the target texture in the first texture map, the target texture in the first texture map can be transplanted to the same position in the original texture map of the target model, and then the edge blurring and color fusion of the target texture in the original texture map can be performed to obtain the target texture map.

[0037] By way of example and not limitation, Figure 2 FIG. 1 is a schematic diagram of fusing freckles on a person's face in a first texture map to an original texture map of a target model, wherein the image on the left is a first texture map, and the image on the right is an original texture map of a target model. The target texture in the first texture map is the freckles on the left face of the person in the map, that is, Figure 2 The image pointed by the arrow on the left is fused with freckles onto the left face of the face image on the right, and the target texture map pointed by the arrow on the right can be obtained.

[0038] Step S40: Project the target texture map onto the target model to obtain a target model including the target texture.

[0039] In this embodiment, after obtaining the target texture map containing the target texture, the target texture map can be projected onto the target model according to the original mapping relationship between the target model and its original texture map, thereby obtaining the target model containing the target texture. It should be noted that the above steps S10-S30 do not change the UV mapping coordinates of the texture map of the target model. Therefore, this step only needs to accurately map the target texture map to the target model according to the original mapping relationship, making the addition or refinement of the model texture more efficient.

[0040] The model generation method provided in the above embodiment only needs to generate a target image containing the required target texture through the prompt words of the large language model. The target image can be at any angle, and the target model containing the target texture can be obtained through texture map conversion and texture fusion, so that the efficiency of adding, refining and modifying the texture of the target model is higher.

[0041] Next, the specific method of converting an image into a texture map is described.

[0042] In one embodiment, when converting a target image into a first texture map suitable for a target model, a first model corresponding to the target image is generated, and an original texture map of the first model is obtained; wherein the original texture map of the first model includes the target texture; the original texture map of the first model is adapted to the original texture map of the target model to obtain a first texture map suitable for the target model.

[0043] In this embodiment, the target image is first converted into a first model, wherein the original texture map of the first model is generated by the target image, and then the original texture map of the first model is obtained and adapted to the original texture map of the target model to obtain a first texture map suitable for the target model. It is understandable that there are many ways to model images, such as deep learning algorithms, which can be used to generate the first model corresponding to the target image, and the specifics are not limited here.

[0044] Specifically, in one embodiment, if the target model is a human head model, then a first model corresponding to the target image can be generated by a head model building tool. For example, the head model building tool facebuilder in three-dimensional production software such as film and television synthesis software Nuke, or three-dimensional graphic image software Blender, can be used to generate a model corresponding to the head image, thereby obtaining a first model corresponding to the target image.

[0045] It should be noted that the first model corresponding to the target image already includes a texture map corresponding to the target image, and the original texture map of the first model can be obtained by performing UV unfolding on the first model corresponding to the target image.

[0046] By way of example and not limitation, Figure 3 FIG. 1 is a schematic diagram of a target image, a first model corresponding to the target image, and an original texture map of the first model. Figure 3 In the figure, (a) is a target image, (b) is a first model corresponding to the target image, and (c) is an original texture map corresponding to the first model, which is not limited to the specific details.

[0047] In one embodiment, when a first model corresponding to a target image is generated and an original texture map of the first model is obtained, a three-dimensional model of the target image is generated to obtain the first model; the target image is projected onto the first model to obtain the first model including the texture map; and the texture map of the first model including the texture map is expanded to obtain the original texture map of the first model.

[0048] In this embodiment, a three-dimensional model of the target image can be generated first. The three-dimensional model is used to indicate the structure / architecture / skeleton structure of the target object in the target image and does not contain texture. The target image is then projected onto the first model so that the first model obtains texture. Finally, the first model is UV unfolded to obtain the original texture map of the first model.

[0049] By way of example and not limitation, Figure 4 Shown is Figure 3 The schematic diagram of the three-dimensional model corresponding to the target image shown is that the three-dimensional model does not contain texture but contains the structure of the target person in the target image. The structure of the model is represented by the vertices of the model, which is not specifically limited here.

[0050] In one embodiment, when the original texture map of the first model is adapted to the original texture map of the target model to obtain a first texture map suitable for the target model, at least one common feature area in the original texture map of the target model and the original texture map of the first model is determined; wherein each common feature area corresponds to a feature common to the first model and the target model; the original texture map of the first model is positionally matched with at least one common feature area to obtain a first texture map suitable for the target model.

[0051] It should be noted that the common features between the first model and the target model refer to the same features that the first model and the target model have in structure / architecture / skeleton structure. For example, when the target model is a human head model, the common features may include human facial features. The common feature area refers to the area where the facial features are located, specifically the area of ​​the facial features in the original texture map of the target model and the original texture map of the first model. Each common feature area includes corresponding common features, such as eyes, nose, and mouth, which are not specifically limited here.

[0052] In one embodiment, when determining at least one common feature area in the original texture map of the target model and the original texture map of the first model, the common features of the original texture map of the target model and the original texture map of the first model can be identified respectively through a pre-trained artificial intelligence model, and the identified common features are marked as areas / ranges in the form of image coordinates to indicate the positions of the common feature areas in the original texture map of the target model and the original texture map of the first model, respectively.

[0053] In this embodiment, after determining at least one common feature area in the original texture map of the target model and the original texture map of the first model, the common feature areas of the original texture map of the first model are positionally matched with the common feature areas of the original texture map of the target model, so that each common feature area of ​​the original texture map of the first model is aligned with the common feature area corresponding to the original texture map of the target model in its entirety or at least one edge is aligned, thereby obtaining a first texture map suitable for the target model.

[0054] Specifically, when performing position matching of at least one common feature area in the original texture map of the first model, one or more operations such as translation, stretching, compression, distortion, and transformation can be performed on the common feature areas in the original texture map of the first model according to the coordinate information of the common feature areas, so that each common feature area of ​​the original texture map of the first model is aligned with the common feature area corresponding to the original texture map of the target model in all or at least one edge is aligned, thereby obtaining a first texture map suitable for the target model.

[0055] In one embodiment, when adapting the original texture map of the first model to the original texture map of the target model to obtain a first texture map suitable for the target model, a target area where the target texture needs to be projected is determined; wherein the target area is used to indicate an area in the original texture map of the target model; the area where the target texture is located in the original texture map of the first model is positionally matched with the target area to obtain a first texture map suitable for the target model.

[0056] It is understandable that when adapting the original texture map of the first model to the original texture map of the target model, in addition to the method of position matching of the common feature area, the method of target texture area matching can also be adopted. In this embodiment, the target area of ​​the target model that the target texture needs to be projected to is first determined. The target area is represented in the original texture map of the target model. For example, assuming that the target texture is a mole point on the face of a person, the mole point needs to be projected to the nose tip area of ​​the target model (human head model), then the nose tip area of ​​the original texture map of the target model is the target area, and the specific details are not limited here.

[0057] In this embodiment, after determining the target area where the target texture needs to be projected, the area where the target texture is located in the original texture map of the first model can be positionally matched with the target area in the original texture map of the target model. Specifically, one or more operations such as translation, stretching, compression, distortion, and transformation can be performed to perform position matching, so as to obtain a first texture map suitable for the target model.

[0058] By way of example and not limitation, Figure 5 Shown is a schematic diagram of a first texture map (a), an original texture map of a target model (b), and a partial target texture map (c), wherein the white cross in (a) is the target texture, the white rectangular box in (b) is the target area to which the target texture is to be projected, and (c) is the partial target texture map after the target texture is projected. It can be seen that the white cross is projected to the position defined by the white rectangular box, which is not specifically limited here.

[0059] Next, the target texture fusion method is described in detail.

[0060] In one embodiment, when the target texture in the first texture map is fused to the original texture map of the target model to obtain the target texture map, the color of the first texture map is adjusted according to the color of the original texture map of the target model to obtain a second texture map; the area where the target texture is located in the second texture map is superimposed with the original texture map of the target model to obtain the target texture map.

[0061] It should be noted that, in order to make the target texture present a more realistic effect on the target model, this embodiment adjusts the color of the first texture map to be close to the color of the original texture map of the target model according to the color of the original texture map of the target model, thereby obtaining the second texture map. Specifically, when the color of the first texture map is adjusted, at least one color value of the hue, saturation, contrast, brightness, exposure, hue, etc. of the first texture map is adjusted so that the range of the corresponding color value is within a certain range of the corresponding color value of the original texture map of the target model, thereby obtaining the second texture map.

[0062] In this embodiment, after obtaining the second texture map, the image area where the target texture in the second texture map is located can be superimposed on the original texture map of the target model to obtain the target texture map. In one embodiment, the image area where the target texture in the second texture map is located can be directly copied to the same area in the original texture map of the target model to obtain the target texture map.

[0063] In one embodiment, when the target texture map is obtained by superimposing the area where the target texture is located in the second texture map with the original texture map of the target model, the second texture map is aligned and placed under the layer of the original texture map of the target model; wherein the upper layer is the original texture map of the target model; the area where the target texture is located in the upper layer is selected as the mask area by a brush; the mask area and the upper layer are pattern-superimposed to obtain the target texture map.

[0064] In this embodiment, the second texture map is first aligned and stacked under the original texture map of the target model, so that the second texture map and the original texture map of the target model form an upper and lower layer relationship, wherein the second texture map is at the bottom, which is the lower layer, and the original texture map of the target model is at the top, which is the upper layer. Next, a mask is applied to the upper layer to create a mask area of ​​the target texture, so that the mask area of ​​the lower layer appears in the upper layer, and then the mask area is pattern-overlaid with the original texture map of the upper layer, thereby obtaining the target texture map. The pattern overlay may be a normal pattern overlay, which is not specifically limited here.

[0065] Corresponding to the above method embodiment, see Figure 6 A schematic diagram of a model generation device is shown, the device comprising: an acquisition module 601, used to generate a target image through a preset prompt word, the target image including a target texture; a conversion module 602, used to convert the target image into a first texture map suitable for the target model; a fusion module 603, used to fuse the target texture in the first texture map to the original texture map of the target model to obtain a target texture map; a projection module 604, used to project the target texture map to the target model to obtain a target model including the target texture.

[0066] The generation device of the above-mentioned model only needs to generate a target image containing the required target texture through the prompt words of the large language model. The target image can be at any angle, and the target model containing the target texture can be obtained through texture map conversion and texture fusion, so that the efficiency of adding, refining and modifying the texture of the target model is higher.

[0067] Optionally, the above-mentioned conversion module 602 includes: a generation unit, used to generate a first model corresponding to the target image, and obtain an original texture map of the first model; wherein the original texture map of the first model includes the target texture; an adaptation unit, used to adapt the original texture map of the first model to the original texture map of the target model, and obtain a first texture map suitable for the target model.

[0068] Optionally, the above-mentioned generation unit is specifically used to: generate a three-dimensional model of the target image to obtain a first model; project the target image onto the first model to obtain a first model including a texture map; and perform texture expansion on the first model including the texture map to obtain the original texture map of the first model.

[0069] Optionally, the adaptation unit is specifically used to: determine at least one common feature area in the original texture map of the target model and the original texture map of the first model; wherein each common feature area corresponds to a feature common to the first model and the target model; and perform position matching on at least one common feature area of ​​the original texture map of the first model to obtain a first texture map suitable for the target model.

[0070] Optionally, the above-mentioned adaptation unit is specifically used to: determine a target area where the target texture needs to be projected; wherein the target area is used to indicate an area in the original texture map of the target model; positionally match the area where the target texture is located in the original texture map of the first model with the target area to obtain a first texture map suitable for the target model.

[0071] Optionally, the above-mentioned fusion module 603 includes: an adjustment unit, used to adjust the color of the first texture map according to the color of the original texture map of the target model to obtain a second texture map; an overlay unit, used to overlay the area where the target texture in the second texture map is located with the original texture map of the target model to obtain a target texture map.

[0072] Optionally, the above-mentioned overlay unit is specifically used to: align the second texture map and place it under the layer of the original texture map of the target model; wherein the upper layer is the original texture map of the target model; select the area where the target texture is located in the upper layer as the mask area by brush; perform pattern overlay on the mask area and the upper layer to obtain the target texture map.

[0073] This embodiment also provides an electronic device, including a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the above-mentioned model generation method. The electronic device can be a server or a terminal device.

[0074] See also Figure 7 As shown, the electronic device includes a processor 100 and a memory 101, wherein the memory 101 stores machine executable instructions that can be executed by the processor 100, and the processor 100 executes the machine executable instructions to implement the above-mentioned model generation method.

[0075] Further, Figure 7The electronic device shown further includes a bus 102 and a communication interface 103 , and the processor 100 , the communication interface 103 and the memory 101 are connected via the bus 102 .

[0076] The memory 101 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 103 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used. The bus 102 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0077] The processor 100 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 100. The above processor 100 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present disclosure can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present disclosure can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module may be located in a storage medium mature in the art, such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 101, and the processor 100 reads the information in the memory 101 and completes the steps of the method of the above embodiment in combination with its hardware, for example:

[0078] A target image is generated by preset prompt words, and the target image includes a target texture; the target image is converted into a first texture map suitable for a target model; the target texture in the first texture map is merged into the original texture map of the target model to obtain a target texture map; the target texture map is projected onto the target model to obtain a target model including the target texture.

[0079] In this method, it is only necessary to generate a target image containing the required target texture through the prompt words of the large language model. The target image can be at any angle, and then the target model containing the target texture can be obtained through texture map conversion and texture fusion, making the addition, refinement and modification of the texture of the target model more efficient.

[0080] Optionally, the step of converting the target image into a first texture map suitable for the target model includes: generating a first model corresponding to the target image, and obtaining an original texture map of the first model; wherein the original texture map of the first model includes the target texture; and adapting the original texture map of the first model to the original texture map of the target model to obtain a first texture map suitable for the target model.

[0081] Optionally, the steps of generating a first model corresponding to the target image and obtaining the original texture map of the first model include: generating a three-dimensional model of the target image to obtain the first model; projecting the target image onto the first model to obtain the first model including the texture map; and performing texture expansion on the first model including the texture map to obtain the original texture map of the first model.

[0082] Optionally, the step of adapting the original texture map of the first model to the original texture map of the target model to obtain a first texture map suitable for the target model includes: determining at least one common feature area in the original texture map of the target model and the original texture map of the first model; wherein each common feature area corresponds to a feature common to the first model and the target model; and performing position matching of at least one common feature area in the original texture map of the first model to obtain a first texture map suitable for the target model.

[0083] Optionally, the step of adapting the original texture map of the first model to the original texture map of the target model to obtain a first texture map suitable for the target model includes: determining a target area where the target texture needs to be projected; wherein the target area is used to indicate an area in the original texture map of the target model; and positionally matching an area in the original texture map of the first model where the target texture is located with the target area to obtain a first texture map suitable for the target model.

[0084] Optionally, the step of fusing the target texture in the first texture map to the original texture map of the target model to obtain the target texture map includes: adjusting the color of the first texture map according to the color of the original texture map of the target model to obtain a second texture map; and superimposing the area where the target texture is located in the second texture map with the original texture map of the target model to obtain the target texture map.

[0085] Optionally, the step of superimposing the area where the target texture in the second texture map is located with the original texture map of the target model to obtain the target texture map includes: aligning the second texture map and placing it under the layer of the original texture map of the target model; wherein the upper layer is the original texture map of the target model; selecting the area where the target texture is located in the upper layer as the mask area by using a brush; and pattern-superimposing the mask area and the upper layer to obtain the target texture map.

[0086] This embodiment further provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement the above-mentioned model generation method, for example:

[0087] A target image is generated by preset prompt words, and the target image includes a target texture; the target image is converted into a first texture map suitable for a target model; the target texture in the first texture map is merged into the original texture map of the target model to obtain a target texture map; the target texture map is projected onto the target model to obtain a target model including the target texture.

[0088] In this method, it is only necessary to generate a target image containing the required target texture through the prompt words of the large language model. The target image can be at any angle, and then the target model containing the target texture can be obtained through texture map conversion and texture fusion, making the addition, refinement and modification of the texture of the target model more efficient.

[0089] Optionally, the step of converting the target image into a first texture map suitable for the target model includes: generating a first model corresponding to the target image, and obtaining an original texture map of the first model; wherein the original texture map of the first model includes the target texture; and adapting the original texture map of the first model to the original texture map of the target model to obtain a first texture map suitable for the target model.

[0090] Optionally, the steps of generating a first model corresponding to the target image and obtaining the original texture map of the first model include: generating a three-dimensional model of the target image to obtain the first model; projecting the target image onto the first model to obtain the first model including the texture map; and performing texture expansion on the first model including the texture map to obtain the original texture map of the first model.

[0091] Optionally, the step of adapting the original texture map of the first model to the original texture map of the target model to obtain a first texture map suitable for the target model includes: determining at least one common feature area in the original texture map of the target model and the original texture map of the first model; wherein each common feature area corresponds to a feature common to the first model and the target model; and performing position matching of at least one common feature area in the original texture map of the first model to obtain a first texture map suitable for the target model.

[0092] Optionally, the step of adapting the original texture map of the first model to the original texture map of the target model to obtain a first texture map suitable for the target model includes: determining a target area where the target texture needs to be projected; wherein the target area is used to indicate an area in the original texture map of the target model; and positionally matching an area in the original texture map of the first model where the target texture is located with the target area to obtain a first texture map suitable for the target model.

[0093] Optionally, the step of fusing the target texture in the first texture map to the original texture map of the target model to obtain the target texture map includes: adjusting the color of the first texture map according to the color of the original texture map of the target model to obtain a second texture map; and superimposing the area where the target texture is located in the second texture map with the original texture map of the target model to obtain the target texture map.

[0094] Optionally, the step of superimposing the area where the target texture in the second texture map is located with the original texture map of the target model to obtain the target texture map includes: aligning the second texture map and placing it under the layer of the original texture map of the target model; wherein the upper layer is the original texture map of the target model; selecting the area where the target texture is located in the upper layer as the mask area by using a brush; and pattern-superimposing the mask area and the upper layer to obtain the target texture map.

[0095] The computer program product of the model generation method, device, electronic device and storage medium provided in the embodiments of the present disclosure includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the previous method embodiments. The specific implementation can be found in the method embodiments, which will not be repeated here.

[0096] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0097] In addition, in the description of the embodiments of the present disclosure, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present disclosure can be understood according to specific circumstances.

[0098] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present disclosure, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0099] In the description of the present disclosure, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present disclosure and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present disclosure. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0100] Finally, it should be noted that the above embodiments are only specific implementation methods of the present disclosure, which are used to illustrate the technical solutions of the present disclosure, rather than to limit them. The protection scope of the present disclosure is not limited thereto. Although the present disclosure is described in detail with reference to the above embodiments, those skilled in the art should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above embodiments within the technical scope disclosed in the present disclosure, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A method for generating a model, characterized in that: The method comprises: Generate a target image by using a preset prompt word, wherein the target image includes a target texture; Converting the target image into a first texture map suitable for a target model; Merging the target texture in the first texture map into the original texture map of the target model to obtain a target texture map; The target texture map is projected onto the target model to obtain a target model including the target texture.

2. The method according to claim 1, characterized in that: The step of converting the target image into a first texture map suitable for the target model comprises: Generate a first model corresponding to the target image, and obtain an original texture map of the first model; wherein the original texture map of the first model includes the target texture; The original texture map of the first model is adapted to the original texture map of the target model to obtain a first texture map suitable for the target model.

3. The method according to claim 2, characterized in that The step of generating a first model corresponding to the target image and obtaining an original texture map of the first model includes: Generate a three-dimensional model of the target image to obtain a first model; Projecting the target image onto the first model to obtain a first model including a texture map; The texture map of the first model including the texture map is expanded to obtain the original texture map of the first model.

4. The method according to claim 2, characterized in that The step of adapting the original texture map of the first model to the original texture map of the target model to obtain the first texture map suitable for the target model comprises: Determine at least one common feature region in the original texture map of the target model and the original texture map of the first model; wherein each common feature region corresponds to a feature common to the first model and the target model; The original texture map of the first model is subjected to position matching of the at least one common feature region to obtain a first texture map suitable for the target model.

5. The method according to claim 2, characterized in that: The step of adapting the original texture map of the first model to the original texture map of the target model to obtain the first texture map suitable for the target model comprises: Determine a target area where the target texture needs to be projected; wherein the target area is used to indicate an area in the original texture map of the target model; The region where the target texture is located in the original texture map of the first model is matched with the target region to obtain a first texture map suitable for the target model.

6. The method according to claim 1, characterized in that The step of fusing the target texture in the first texture map to the original texture map of the target model to obtain the target texture map comprises: According to the color of the original texture map of the target model, adjusting the color of the first texture map to obtain a second texture map; The area where the target texture is located in the second texture map is superimposed with the original texture map of the target model to obtain a target texture map.

7. The method according to claim 6, characterized in that The step of superimposing the area where the target texture is located in the second texture map with the original texture map of the target model to obtain the target texture map comprises: Aligning and placing the second texture map below the layer of the original texture map of the target model; wherein the upper layer is the original texture map of the target model; Use the brush to select the area where the target texture is located in the upper layer as the mask area; The mask area and the upper layer are pattern-superimposed to obtain a target texture map.

8. A model generation device, characterized in that: The device comprises: An acquisition module, used to generate a target image by using a preset prompt word, wherein the target image includes a target texture; A conversion module, used for converting the target image into a first texture map suitable for a target model; A fusion module, used for fusing the target texture in the first texture map to the original texture map of the target model to obtain a target texture map; The projection module is used to project the target texture map onto the target model to obtain a target model including the target texture.

9. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the method for generating a model according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement the model generation method described in any one of claims 1 to 7.

Images