Vehicle model processing method, vehicle, and related apparatus

By selecting or generating images from the vehicle's local storage and performing intelligent composition and semantic segmentation, the problem of personalized customization of vehicle 3D model patterns in existing technologies has been solved, enabling personalized painting and real-time adjustments, thus improving the user experience.

WO2026237640A1PCT designated stage Publication Date: 2026-11-19HUAWEI TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/CN2026/090869
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-14
Filing Date
2026-04-15
Publication Date
2026-11-19

AI Technical Summary

Technical Problem

In existing technologies, users can only select patterns for vehicle 3D models from a limited range of system options, making personalized customization impossible. Furthermore, the operation process is cumbersome, the visual presentation is monotonous, and it is difficult to support real-time adjustments.

Method used

By selecting images from the vehicle's local storage or inputting voice commands to generate images, and performing intelligent composition and semantic segmentation, image parameters can be adjusted to achieve personalized painting of 3D car models.

Benefits of technology

It enhances the personalization and customization of car models, simplifies the operation process, supports real-time adjustments, and improves the richness and realism of visual presentation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2026090869_19112026_PF_FP_ABST
    Figure CN2026090869_19112026_PF_FP_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of terminals, and provides a vehicle model processing method, a vehicle, and a related apparatus. The method comprises: in response to a first operation for determining a first image, displaying, in a first interface, a target vehicle model having a second image rendered thereon, the first image being locally stored in the vehicle or generated on the basis of a voice instruction, an image size of the second image satisfying a preset size requirement, and the second image comprising a main object present in the first image; and subsequently, in response to a second operation on the target vehicle model, performing first adjustment processing on the second image displayed on a surface of the target vehicle model, the first adjustment processing comprising at least one of the following: moving, rotating, cropping, and zooming; and further in response to a third operation on a first adjustment control in the first interface, performing second adjustment processing on the second image displayed on the surface of the target vehicle model. In this way, the degree of personalization of the vehicle model can be effectively increased.
Need to check novelty before this filing date? Find Prior Art

Description

Model car processing methods, vehicles and related devices

[0001] This application claims priority to Chinese Patent Application No. 202510625933.8, filed on May 14, 2025, entitled “Method for Processing Car Model, Vehicle and Related Devices”, the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of terminal technology, and in particular to a method for processing car models, a vehicle, and related devices. Background Technology

[0003] With the accelerating pace of automotive intelligence, it's now possible to display a 3D virtual model of a vehicle on its screen. Users can then adjust the appearance of the 3D model through a graphical interface.

[0004] In one implementation, preset options can be provided in the settings interface, which may include color options, wheel options, texture options, and pattern options, etc. The user then selects a specific option through a menu. Subsequently, the system renders the 3D car model based on the user's selection to generate a 3D car model that reflects the corresponding effect, where the pattern selected by the user is typically displayed in a fixed preset position on the 3D car model.

[0005] However, in this implementation, users can only select from a limited range of options provided by the system, and the pattern can only be displayed in a fixed position, thus failing to effectively meet users' personalized customization needs. Summary of the Invention

[0006] This application provides a method for processing vehicle models, a vehicle, and related devices to enhance the personalization of vehicle models.

[0007] In a first aspect, embodiments of this application provide a method for processing vehicle models, applied to vehicles, the method comprising:

[0008] The response is used to determine the first image, display the target car model with the second image rendered in the first interface, the first image is determined in the local storage of the vehicle, or the first image is generated based on voice command, the image size of the second image meets the preset size requirements, and the second image contains the main object existing in the first image;

[0009] In response to a second operation applied to a target car model, a first adjustment process is performed on a second image displayed on the surface of the target car model, the first adjustment process including at least one of the following: moving, rotating, cropping, and scaling;

[0010] In response to a third operation performed on a first adjustment control in the first interface, a second adjustment process is performed on a second image displayed on the surface of the target car model. The second adjustment process is used to adjust at least one of the following image parameters: brightness, contrast, saturation, hue, color temperature, transparency, and blending mode.

[0011] This allows users to select a custom first image for vehicle painting, a significantly improved approach compared to limiting selection to preset images, enhancing the personalization of the model car's appearance. Furthermore, this application supports voice command-based painting, eliminating the need for multiple user operations and reducing complexity. Additionally, the technical solution allows users to adjust the composition or parameters of the image on the model car's surface using gestures, further enhancing the final appearance and personalization of the model.

[0012] In one possible implementation, the response for determining a first operation of the first image, displaying a target car model rendered with a second image in a first interface, includes:

[0013] The response is used to determine a first operation of the first image, to perform a compositional adjustment on the first image to obtain at least one second image;

[0014] The preset car model is rendered based on at least one second image, so as to display the target car model with the rendered second image in the first interface.

[0015] In this way, intelligent composition can be performed on the first image determined by the user to generate a second image that meets the preset size requirements. At the same time, the main object in the first image can be included in the second image, which can effectively improve the rationality and aesthetics of the painting effect.

[0016] In one possible implementation, the composition of the first image is adjusted to obtain at least one second image, including:

[0017] The first image is input into the first model so that the first model outputs the second image; wherein the first model is used to perform a first adjustment process on the first image, and the first model is a model trained based on at least one set of sample data.

[0018] Among them, intelligent composition using the first model can improve the rationality of the second image obtained after composition.

[0019] In one possible implementation, the composition of the first image is adjusted to obtain at least one second image, including:

[0020] The first image is semantically segmented according to the second model to obtain segmented sub-images corresponding to each of the at least one target object contained in the first image, wherein the segmented sub-images obtained after semantic segmentation are the second image.

[0021] Furthermore, the technical solution of this application can also support semantic segmentation of the first image to obtain the second image corresponding to each object in the first image. Then, multiple second images are rendered to their respective corresponding areas, thereby improving the personalization and diversification of the painting effect.

[0022] In one possible implementation, the method also includes:

[0023] The second control is displayed on the first interface. The second control is used to determine whether semantic segmentation needs to be performed.

[0024] The first image is semantically segmented according to the second model to obtain segmented sub-images corresponding to at least one target object contained in the second image, including:

[0025] In response to a fourth operation applied to the second control, the first image is semantically segmented according to the second model to obtain segmented sub-images corresponding to at least one target object contained in the second image.

[0026] This allows users to choose whether to perform semantic segmentation on the first image based on their actual needs, so as to achieve the painting effect they require.

[0027] In one possible implementation, a preset car model is rendered based on at least one second image to display the target car model with the rendered second image on a first interface, including:

[0028] For any segmented sub-image, determine the target region in the preset car model corresponding to the segmented sub-image;

[0029] Based on the segmented sub-images, the target region in the preset car model is rendered to generate the target car model. The surface of the target region of the target car model is covered with the segmented sub-images.

[0030] This allows different segmented sub-images to be rendered onto their respective areas within the preset car model, thereby enhancing the personalization and diversity of the final paint effect.

[0031] In one possible implementation, determining the target region in the preset car model corresponding to the segmented sub-image includes:

[0032] Obtain the first semantic type of the target object contained in the segmented sub-image;

[0033] Based on the preset mapping relationship, the vehicle model area corresponding to the first semantic type is determined as the target area. The preset mapping relationship is used to indicate the vehicle model areas corresponding to each of the multiple semantic types.

[0034] By pre-setting the mapping relationship, the rationality and accuracy of the target region determined for each segmented sub-image can be effectively improved.

[0035] In one possible implementation, in response to a second operation applied to a target vehicle model, a first adjustment process is performed on a second image displayed on the surface of the target vehicle model, including:

[0036] In response to the second operation, the second image is subjected to a first adjustment process, and based on the second image after the first adjustment process, the mapping relationship between each pixel in the second image and the texture coordinates of the preset car model is updated.

[0037] Based on the updated mapping relationship, the preset car model is re-rendered based on the second image to perform a first adjustment process on the second image displayed on the surface of the target car model; wherein, during the execution of the first adjustment process, the display effect of the second image on the surface of the target car model is adjusted in real time.

[0038] This implementation method allows for simple and efficient updates to the rendering effect based on the mapping relationship between the texture coordinates of the second image and the preset car model, enabling real-time adjustments to the display effect of the second image during user interaction. This approach eliminates the need for users to possess specialized rendering knowledge, providing the necessary technical support for real-time changes in rendering effects.

[0039] In one possible implementation, the first operation is to select an image from the vehicle's local storage; or, the first operation is to input a voice command to instruct the acquisition of a first image.

[0040] Secondly, embodiments of this application provide a vehicle including a processor and a memory, the memory for storing code instructions, and the processor for running the code instructions to perform the methods described in the first aspect or any possible implementation of the first aspect.

[0041] Thirdly, embodiments of this application provide a computer-readable storage medium storing a computer program or instructions that, when executed on a computer, cause the computer to perform the methods described in the first aspect or any possible implementation thereof.

[0042] Fourthly, embodiments of this application provide a computer program product including a computer program, which, when run on a computer, causes the computer to perform the methods described in the first aspect or any possible implementation of the first aspect.

[0043] Fifthly, this application provides a chip or chip system including at least one processor and a communication interface. The communication interface and the at least one processor are interconnected via a circuit. The at least one processor is used to run computer programs or instructions to perform the methods described in the first aspect or any possible implementation thereof. The communication interface in the chip can be an input / output interface, pins, or circuits, etc.

[0044] In one possible implementation, the chip or chip system described above in this application further includes at least one memory storing instructions. The memory can be an internal storage unit of the chip, such as a register or cache, or it can be a storage unit of the chip itself (e.g., read-only memory, random access memory, etc.).

[0045] It should be understood that the second to fifth aspects of this application correspond to the technical solutions of the first aspect of this application, and the beneficial effects achieved by each aspect and the corresponding feasible implementation are similar, and will not be repeated here. Attached Figure Description

[0046] Figure 1 is a schematic diagram of the interface of the car model processing method provided in the embodiment of this application;

[0047] Figure 2 is a schematic diagram of the interface of the car model processing method provided in the embodiment of this application;

[0048] Figure 3 is a flowchart of the car model processing method provided in an embodiment of this application;

[0049] Figure 4 is a schematic diagram of the interface of the car model processing method provided in the embodiment of this application;

[0050] Figure 5 is a schematic diagram of image processing provided in an embodiment of this application;

[0051] Figure 6 is a schematic diagram of image processing provided in an embodiment of this application;

[0052] Figure 7 is a schematic diagram of image processing provided in an embodiment of this application;

[0053] Figure 8 is a schematic diagram of the preset mapping relationship provided in the embodiments of this application;

[0054] Figure 9 is a schematic diagram of the interface of the car model processing method provided in the embodiment of this application;

[0055] Figure 10 is a schematic diagram of the interface of the car model processing method provided in the embodiment of this application;

[0056] Figure 11 is a schematic diagram of the interface of the car model processing method provided in the embodiment of this application;

[0057] Figure 12 is a schematic diagram of the interface of the car model processing method provided in the embodiment of this application;

[0058] Figure 13 is a schematic diagram of the interface of the car model processing method provided in the embodiment of this application;

[0059] Figure 14 is a schematic diagram illustrating the implementation of the image adjustment method provided in the embodiments of this application;

[0060] Figure 15 is a schematic diagram of the interface of the car model processing method provided in the embodiment of this application;

[0061] Figure 16 is a schematic diagram of transparency adjustment provided in an embodiment of this application;

[0062] Figure 17 is a schematic diagram of the interface of the car model processing method provided in the embodiment of this application;

[0063] Figure 18 is a schematic diagram of the software architecture of the electronic device provided in the embodiment of this application. Detailed Implementation

[0064] To facilitate understanding, the relevant terms and concepts involved in the embodiments of this application will be introduced below:

[0065] 1. 3D car model: In this application, 3D car model refers to a three-dimensional virtual car model, which is usually a virtual car model created by 3D modeling software. For example, a virtual 3D car model can be displayed on the screen of an electronic device to meet the corresponding user needs.

[0066] 2. Painting: Rendering and displaying 2D images on the 3D car model of the smart cockpit.

[0067] 3. UV: UV is a two-dimensional coordinate system (usually represented as the U-axis and V-axis), used to define the correspondence between the vertices of a 3D model surface and the 2D texture image.

[0068] 4. Electronic equipment

[0069] The electronic devices in this application embodiment can be referred to as user equipment (UE), terminal, etc. For example, electronic devices can be mobile phones, tablets, personal digital assistants (PDAs), handheld devices with wireless communication capabilities, computing devices, in-vehicle devices or wearable devices, virtual reality (VR) terminal devices, augmented reality (AR) terminal devices, wireless terminals in industrial control, wireless terminals in smart homes, etc. The form of the electronic devices is not specifically limited in this application embodiment.

[0070] To better understand the technical solution of this application, the relevant technologies involved in this application will be further introduced below.

[0071] As the automotive industry accelerates its intelligent transformation, smart cockpits are evolving from functional carriers to centers of emotional interaction. In one scenario, a 3D virtual model of the vehicle can be displayed on an electronic device screen. Users can then interact with the model via a graphical interface to adjust its appearance, enhancing visual immersion and emotional engagement, and strengthening the human-vehicle interaction experience.

[0072] To support users in adjusting the appearance of 3D car models, one implementation method exists, which may include the following steps:

[0073] 1) The settings interface provides preset options, which may include color options, wheel options, texture options, and pattern options, etc.

[0074] 2) Users can select basic appearance parameters such as body color, wheel style, and body texture through menu operations, and users can also choose from preset pattern options;

[0075] 3) The system renders the 3D car model based on the user's selection to generate a 3D car model that represents the corresponding body color, wheel style, body texture and the pattern selected by the user. The pattern selected by the user is usually fixedly displayed in a preset position on the 3D car model.

[0076] However, this implementation method has many limitations: First, it has low freedom, with users only able to select from a limited range of options and the pattern only displayed in a fixed position, thus failing to effectively meet the personalized customization needs of users; second, the operation process is relatively cumbersome, requiring users to configure parameters item by item through multiple menus, resulting in a high learning cost and hindering quick mastery; third, the visual presentation is relatively simple, only supporting the replacement of basic colors or simple textures, lacking processing of blending effects, and the overall visual experience is not realistic or rich enough.

[0077] In addition, there is another implementation method, which may include the following steps:

[0078] 1) Obtain real-time scene information (such as entertainment application data, vehicle speed, gear position, etc.) and basic data of 3D car model through the vehicle system;

[0079] 2) The vehicle system uploads data to the cloud-based large model platform so that the cloud-based large model platform can generate a vehicle model painting effect that matches the current scene;

[0080] 3) The vehicle's infotainment system receives the painting data generated in the cloud and then renders and displays it.

[0081] This approach also has its drawbacks, such as: First, users passively accept preset styles, where the painting effect is automatically matched by the system according to the scene, and users cannot choose or create personalized patterns on their own; Second, the interaction latency is high, where the painting effect relies on cloud computing, and a single rendering usually takes more than 1 second, making it difficult to support real-time adjustments.

[0082] To address the aforementioned problems, this application proposes a method for processing 3D car models. This method allows users to select any image from their photo album, and also allows users to generate any desired image through semantic commands. The 3D car model is then selected based on the user-selected image or the image generated by the command, thereby enhancing the personalization of the 3D car model. Furthermore, before performing rendering processing, the technical solution of this application can intelligently adjust the selected image to improve the final rendered 3D car model's performance.

[0083] Based on the above introduction, the following describes the car model processing method provided in this application in conjunction with specific embodiments.

[0084] In the technical solution of this application, users can select the required images from the local photo album of the electronic device for painting the 3D car model, or users can input instructions and then generate corresponding images based on the instructions output by the user to paint the 3D car model. For example, the operation of the user determining the image can be called the first operation, and the image selected by the user or the image generated based on the user's instructions can be called the first image.

[0085] The following describes the implementation of determining the first image based on the first operation in conjunction with Figures 1 and 2. Figure 1 is a schematic diagram of the interface of the car model processing method provided in the embodiment of this application, and Figure 2 is a schematic diagram of the interface of the car model processing method provided in the embodiment of this application.

[0086] In one implementation, the screen of the electronic device may display a user interface as shown in Figure 1(a), which can be understood as an interface for adjusting the style of the car model. Referring to Figure 1(a), the user interface may display a 3D car model 101, the style of which may be, for example, the default style, or the style after the most recent adjustment; this embodiment does not limit this.

[0087] Referring to Figure 1(a), the user interface for adjusting the model car style may also include control unit 102 and control unit 103 as shown in the figure.

[0088] The control unit 102 provides options for the color of the car model. For example, the control unit 102 may include operation controls 1021-1025 as shown in the figure. Exemplarily, operation controls 1022-1025 may each correspond to a different color, so that when the user operates these controls, the color of the car model 101 displayed on the interface is adjusted to the color corresponding to the control. Furthermore, operation control 1021 may be a control for providing more color options. For example, after detecting a user operation on operation control 1021, a color selection window can be displayed on the interface for the user to select the desired color.

[0089] Furthermore, the control unit 103 is used to provide a selection of the paint image. For example, the control unit 103 may include operation controls 1031 to 1035 as shown in the figure. Exemplarily, operation control 1031 is used to remove the paint effect applied to the car model 101, that is, to restore the car model 101 to the style of the unpainted image. Also, operation controls 1032 to 1033 may each correspond to different images. These images may be images selected by the user from the album during historical operations, or they may be preset images; this embodiment does not limit this. In this way, when the user operates these controls, the paint effect of the car model 101 displayed on the interface can be adjusted to the appearance of the image corresponding to the control, that is, the image corresponding to the control is rendered on the surface of the car model.

[0090] Furthermore, the operation control 1035 can be a control for users to select images. For example, after detecting a user operation on the operation control 1035, an image selection interface as shown in Figure 1(b) can be displayed on the screen. As shown, multiple images can be displayed in the image selection interface, and a rectangle in the figure can be understood as an image. These images can be, for example, images from the local photo album of the electronic device. For example, the interface shown in Figure 1(b) can also be the photo album interface of the electronic device, that is, the photo album interface can be pulled up and displayed on the screen through the operation control 1035.

[0091] Subsequently, for example, if a user operation on image 104 is detected, it can be determined that the current user has selected image 104 for vehicle painting. Then, the car model can be rendered based on image 104, thereby generating the car model 105 with the painting effect shown in Figure 1(c). Referring to Figure 5, it can be understood that image 104 is rendered on the side of the car model 105. In this embodiment, the user operation on the operation control 1035 and the user operation of selecting image 104 can be understood as the first operation described above. In actual implementation, the specific implementation of the first operation can be set according to actual needs, as long as the first operation is used to select an image from local storage.

[0092] This allows users to select custom images for vehicle painting, which is a more personalized approach compared to the previous method of only allowing selection from preset images.

[0093] In addition, in another scenario, users can also paint the car model using voice commands. This scenario will be described below with reference to Figure 2.

[0094] The user interface shown in Figure 2(a) is similar to that shown in Figure 1(a) above, and will not be described again here. Furthermore, the user can, for example, activate the electronic device's voice assistant and input commands to instruct the model car to be painted. As shown, for example, the user can input the voice command "Please put a pattern containing star elements on my model car".

[0095] The electronic device then responds to the user's voice command, for example, by matching relevant images from the device's local photo album. For instance, keywords can be extracted from the voice command, and then matched against images in the local photo album to retrieve relevant images and apply paint to the car model.

[0096] Alternatively, electronic devices can respond to voice commands and directly generate the image specified by the command. For example, a voice command can be input into an image generation model, causing the model to generate an image corresponding to the voice command. The image generation model can be deployed within the electronic device, allowing image generation to be achieved directly through the device. Alternatively, the image generation model can be deployed in the cloud, allowing the electronic device to upload the voice command to the cloud, enabling the cloud to generate a corresponding image based on the command. The electronic device then receives the generated image from the cloud and performs the painting process on the car model.

[0097] In one implementation, the electronic device can first respond to a voice command and perform image matching locally. If an image matching the voice command exists, the vehicle model can be painted directly based on the local image. If no image matching the voice command exists locally, a corresponding image is generated based on an image generation model for painting the vehicle model. This application does not limit the specific implementation of how the image to be painted is determined based on the voice command; it can be selected according to actual needs.

[0098] In the current embodiment, the user input of voice commands can be understood as the first operation described above. In actual implementation, the specific implementation of the first operation can be set according to actual needs, as long as the first operation is used to indicate the operation of determining the first image.

[0099] After the corresponding image is determined based on the voice command, the car model can be painted based on that image. Assuming that the image determined based on the voice command shown in the figure is the image shown in Figure 1(b), then the interface shown in Figure 2(b) can be displayed, that is, the car model 202 with the painting effect is displayed on the screen.

[0100] This allows users to input custom commands to instruct the vehicle to be painted according to the corresponding images. This approach can effectively enhance the personalization of the model car's performance. At the same time, using voice commands to instruct the model car to be painted can avoid requiring users to perform multiple operations, thus reducing the difficulty of operation for users.

[0101] Based on the above embodiments, the following describes the process of painting car models based on selected or generated images, using specific embodiments. Figure 3 is a flowchart of the car model processing method provided in this application embodiment; Figure 4 is a schematic diagram of the interface of the car model processing method provided in this application embodiment; Figure 5 is a schematic diagram of image processing provided in this application embodiment; Figure 6 is a schematic diagram of image processing provided in this application embodiment; Figure 7 is a schematic diagram of image processing provided in this application embodiment; Figure 8 is a schematic diagram of the preset mapping relationship provided in this application embodiment; and Figure 9 is a schematic diagram of the interface of the car model processing method provided in this application embodiment.

[0102] As shown in Figure 3, the method includes:

[0103] S301, a first operation for determining the first image is to acquire the first image, wherein the first image is acquired from the local storage of the electronic device, or the first image is generated based on instructions.

[0104] In this embodiment, the first image can be an image acquired locally on the electronic device as described in the above embodiments, or the first image can be an image generated based on a voice command input by the user. Implementation details regarding the first operation and the first image can be understood by referring to the above embodiments, and will not be repeated here.

[0105] S302. Adjust the composition of the first image to obtain at least one second image.

[0106] It is understandable that in the process of rendering a car model based on an image to achieve the final painting effect, there are usually certain requirements for the size of the image; otherwise, the final rendering effect will be unreasonable or the performance will be poor.

[0107] However, since the first image in this embodiment is obtained from the local storage of the electronic device or is generated based on instructions, it cannot be guaranteed that the first image will meet the image size requirements for rendering the car model. Therefore, in this embodiment, the composition of the first image can be adjusted to obtain a second image that meets the preset size requirements. The preset size requirements are the image sizes pre-set for rendering the car model. The specific implementation of the preset size requirements can be set according to actual needs, and this embodiment does not impose any restrictions on this.

[0108] In this embodiment, the composition adjustment of the first image is divided into two different cases, which will be described below. Before proceeding with the detailed description, the reason for dividing the cases will be explained first. The technical solution of this application provides a function that can perform semantic segmentation on objects in the first image, and then paint different objects on different areas of the car model to improve the performance of the car model painting. For example, the technical solution of this application can provide users with options so that users can choose whether to use this function for car model painting according to their actual needs.

[0109] For example, we can refer to Figure 4 for understanding. Figure 4(a) shows the case where the user selects image 104 as the first image, which is similar to the case described in Figure 1(b) above, and will not be repeated here. Exemplarily, after detecting a selection operation on image 104, a pop-up window 401 as shown in Figure 4(b) can be displayed on the interface. This pop-up window 401 is used to ask the user whether they need to paint by dividing the object into regions, that is, whether they need to use the semantic segmentation function described above. Exemplarily, the pop-up window 401 can include a "yes" selection control and a "no" selection control. The subsequent processing methods differ according to the user's operation on different selection controls, which will be described below.

[0110] In this application, the pop-up window 401 shown in Figure 4 can also be understood as a second control, which is used to determine whether to perform semantic segmentation. In actual implementation, the second control can be in the form of a pop-up window as shown in Figure 4, for example, it can be set as a switch button, or it can be set as an option control. This application does not limit the specific setting of the second control; it can be set according to actual needs, as long as the second control can be used to determine whether semantic segmentation needs to be performed. Furthermore, this application does not restrict the entry point for the second control. The above description shows that after the user confirms the first image, the second control is displayed on the first interface to ask the user whether semantic segmentation is needed. In other implementations, the second control can also be set at a fixed entry point (such as in the car model painting settings) so that the user can adjust whether semantic segmentation is needed according to actual needs.

[0111] Taking the pop-up window shown in Figure 4 as an example, the following will explain two different user operations on the second control.

[0112] In one implementation, if the user interacts with the "No" selection control, it means that the user has not selected the semantic segmentation function. In this case, the overall composition of the first image is adjusted to obtain the second image.

[0113] The following can be understood in conjunction with Figure 5. Referring to Figure 5, if the user does not select semantic segmentation, for example, the composition of the first image can be adjusted based on the first model. The first model can, for example, crop and / or enlarge the first image. Then, for example, the first image can be input into the first model so that the first model outputs a second image that meets the preset size requirements.

[0114] Furthermore, during the composition adjustment process, the first model in this application can, for example, retain the main subject object present in the first image. That is, the second image output by the first model includes the main subject object present in the first image. It can be understood that compared to cropping the main subject object from the first image while retaining some unimportant objects (such as background content), retaining the main subject object from the first image effectively improves the rationality of the second image obtained by adjusting the composition of the first image, and also enhances the subsequent performance of the car model painting.

[0115] The subject object can be understood as a prominent object in the image, or as the main object in the image. For example, in an image containing objects such as blue sky, white clouds, grass, and a pet, the pet is usually identified as the main object. For instance, within the first model, the subject object in the first image can be identified through methods such as saliency analysis, semantic segmentation, or object detection. Alternatively, objects occupying more than 30% of the first image and located in the central region can be identified as the subject object. This application does not limit the implementation method for determining the subject object.

[0116] For example, we can refer to Figure 6 for understanding. 601 in the figure can be understood as the first image. Referring to the illustration, we can determine that image 601 contains two triangular objects and one star-shaped object. Figure 6 illustrates several different scenarios for adjusting the composition of the first image, which will be explained below.

[0117] In one scenario, the first model can obtain a second image that meets the preset size requirements by cropping only the first image 601. Furthermore, as described above, the first model can perform intelligent composition during image cropping to preserve the main subject in the first image. Referring to Figure 6, after inputting the first image 601 into the first model, the first model can, for example, output a second image as shown in Figure 602. The second image can have a preset size, and it contains the main subject present in the first image, namely the two triangles and a pentagram shown in the figure.

[0118] In another scenario, the first model can both crop and expand the first image 601 to obtain a second image that meets the preset size requirements. During image cropping, the first model retains the main subject of the first image, and during image expansion, it intelligently expands the image based on its original content. Referring to Figure 6, after inputting the first image 601 into the first model, the first model can, for example, output a second image as shown in Figure 6 (603), where the second image has a preset size.

[0119] Referring to the illustration, it can be determined that, in order to meet the preset size, the first image 601 is both cropped vertically and expanded horizontally. Regions a and b shown in the illustration are the expanded regions. For example, expanded region a may include a triangular object, and expanded region b may include a pentagram object. This allows for image expansion based on the image content of the first image, thereby improving the image quality of the expanded image. Furthermore, the second image also includes the main objects present in the first image, namely the two triangles and one pentagram shown in the illustration.

[0120] However, it's understandable that, in order to meet the preset size requirements, parts of the two triangles are cropped during the image cropping process. In other words, although the first model performs intelligent composition to preserve the main object in the first image as much as possible, under certain image size requirements, parts of the main object will inevitably be cropped.

[0121] Furthermore, in an optional implementation, the first model may, for example, scale the first image to preserve the main subject object in the first image as much as possible. It can be understood that after scaling the first image, the main subject object in the first image can be preserved as completely as possible while meeting the preset size requirements.

[0122] For example, this can be understood by referring to Figure 6. Referring to Figure 6, after the first image 601 is input into the first model, the first model can, for example, output a second image as shown in Figure 6, 604, where the image size of the second image is a preset size. Comparing 601 and 604, it can be understood that the main object has been reduced in size; that is, the first image is reduced in size, and then, based on meeting the preset size, the reduced first image 601 is cropped vertically and expanded horizontally. Regions c and d shown in the figure are the expanded regions. The understanding is similar to the above, and will not be elaborated here.

[0123] Understandably, the area indicated by symbol 'e' in image 604 corresponds to the area in the first image. This also shows that the first model scales the first image to preserve the integrity of the main object. This allows the second image to include the main object present in the first image, namely the two triangles and one pentagram shown in the figure.

[0124] The first and second images described above in conjunction with Figure 6 are merely illustrative examples. In actual implementation, the specific image content depends on the chosen image, and this application does not impose any limitations on this. Understanding the relevant processing steps for the first image in conjunction with the above description is sufficient.

[0125] In another implementation, if the user interacts with the "Yes" selection control, it means that the user has selected the semantic segmentation function. Then, semantic segmentation is performed on the first image to obtain multiple segmented sub-images to achieve composition adjustment. These multiple segmented sub-images are also the second image in this application.

[0126] The following explanation can be aided by referring to Figure 5. Referring to Figure 5, when the user selects semantic segmentation, semantic segmentation can be performed on the first image based on a second model. This second model is the model used for semantic segmentation of the image. For example, the first image can be input into the second model, causing it to output multiple segmented sub-images after semantic segmentation. Exemplarily, after semantic segmentation by the second model, for example, segmented sub-images corresponding to at least one target object in the first image can be obtained. The segmented sub-images obtained after semantic segmentation are the second image described above.

[0127] For example, referring to Figure 7, image 700 can be understood as the first image. As shown in the figure, image 700 contains two triangular objects and one star-shaped object. After semantic segmentation of the first image 700, for example, we can obtain segmented sub-images corresponding to each of these three objects, namely images 701, 702, and 703 shown in the figure. These three segmented sub-images are also the second image described in this embodiment.

[0128] It is understood that Figure 7 is merely an exemplary introduction to semantic segmentation. In actual implementation, the specific processing of semantic segmentation can be understood with reference to the actual situation. This application does not limit the specific processing of semantic segmentation.

[0129] Furthermore, after semantic segmentation to obtain multiple segmented sub-images, the second model can, for example, perform at least one of the following operations on the multiple segmented sub-images: image cropping, image augmentation, or image scaling, so that the second image output by the second model meets the preset size requirements. Specific image adjustments can be referred to the description in the above embodiments, and will not be repeated here. In one implementation, the preset size requirements corresponding to the user selecting semantic segmentation may be different from the preset size requirements corresponding to the user not selecting semantic segmentation. Specific size requirements can be arbitrarily set according to actual needs, and this embodiment does not impose any restrictions on this.

[0130] And it is understandable, because the second model is obtained by semantic segmentation of the first image, where semantic segmentation itself is the process of segmenting the main object, so the second image naturally contains the main object that exists in the first image.

[0131] S303. Render the preset car model according to the second image to generate the target car model.

[0132] After adjusting the first image to obtain the second image, the preset car model can be rendered based on the second image. In this application, the preset car model can be understood as a car model that has already been rendered with a specific color. This specific color can be a default color or it can be selected by the user; there is no limitation on this. Then, by rendering the preset car model based on the second image, the second image can be applied to the surface of the car model, thereby generating the target car model. Correspondingly, the second image is then covered on the surface of the target car model.

[0133] The above describes two different scenarios for generating the second image. The rendering implementation for each of these scenarios will be explained below.

[0134] In one implementation, if the user does not select the semantic segmentation function, the second image is an image obtained by adjusting the complete first image. Rendering can then be performed directly on the second image to generate the target car model. For example, the target car model shown in Figure 1(c) can then be displayed on the first interface, allowing the user to see the painted car model effect in real time.

[0135] In another implementation, if the user selects the semantic segmentation function, then the second image is multiple segmented sub-images obtained by semantically segmenting the first image. For example, the rendering process can be performed as follows:

[0136] First, for any given segmented sub-image, determine the target region within the preset car model corresponding to the segmented sub-image. Then, based on the segmented sub-image, render the target region within the preset car model to generate the target car model, whose surface is covered by the segmented sub-image.

[0137] For example, in determining the target region corresponding to the segmented sub-image, a first semantic type of the target object contained in the segmented sub-image can be obtained. This first semantic type can be, for example, output by a second model, or it can be determined after image analysis of the segmented sub-image; this embodiment does not limit this. Then, based on a preset mapping relationship, the car model region corresponding to the first semantic type can be determined as the target region, where the preset mapping relationship is used to indicate the car model regions corresponding to each of the multiple semantic types.

[0138] For example, the preset mapping relationship can be understood by referring to Figure 8. As shown in Figure 8, the preset mapping relationship can pre-set various semantic types corresponding to different car model areas. For instance, the car model area corresponding to the semantic type "pet" is the hood, the car model area corresponding to the semantic type "person" is the left front door, and the car model area corresponding to the semantic type "plant" is the right front door. In this way, different semantic types can be rendered to different areas of the preset car model.

[0139] In one implementation, when multiple segmented sub-images correspond to the same car model area, these sub-images can be arranged to ensure that different semantic objects are rendered in an orderly manner within the car model area, thus avoiding overlap between semantic objects. The specific arrangement can be pre-defined according to actual needs, as long as the segmented sub-images do not overlap.

[0140] It is also important to understand that the specific division of semantic types and the specific division of the model area on the preset model can be preset according to actual needs. Similarly, the correspondence between semantic types and model areas recorded in the preset mapping relationship can also be set according to actual needs; this embodiment does not limit the specific implementation of the preset mapping relationship.

[0141] Based on this, we can determine which car model region in the preset car model each segmented sub-image corresponds to, based on the preset mapping relationship and the first semantic type corresponding to each segmented sub-image.

[0142] For example, if the first semantic type of a segmented sub-image is "pet," then based on the preset mapping relationship introduced in Figure 8, the preset region corresponding to this segmented sub-image can be determined to be the front hood. As another example, if the first semantic type of a segmented sub-image is "plant," then based on the preset mapping relationship introduced in Figure 8, the preset region corresponding to this segmented sub-image can be determined to be the right front door.

[0143] After determining the corresponding car model region for each segmented sub-image, each segmented sub-image can be rendered onto the corresponding car model region on the preset car model. The specific implementation of the rendering process can be understood by referring to the above introduction, and will not be repeated here.

[0144] For example, the target car model obtained after rendering can be understood in conjunction with Figure 9. In the example introduced in Figure 7, after semantic segmentation of the first image, three segmented sub-images corresponding to the two triangles and the pentagram are obtained. Assuming that the car model areas corresponding to these three segmented sub-images are as shown in Figure 9, then after rendering, the target car model as shown in Figure 9 can be displayed in the first interface.

[0145] In this embodiment, the user can select images from local storage for painting, and the user can also trigger the generation of corresponding images for painting based on voice commands. This implementation can effectively improve the personalization of the painting effect. Furthermore, the technical solution of this application will intelligently compose the first image determined by the user to generate a second image that meets the preset size requirements, while also including the main object in the first image in the second image. This can effectively improve the rationality and aesthetics of the painting effect. In addition, the technical solution of this application can also support semantic segmentation of the first image to obtain the second images corresponding to each object in the first image. Then, multiple second images are rendered to their respective corresponding areas, thereby improving the personalization and diversification of the painting effect.

[0146] Referring to the above embodiments, it can be determined that in the technical solution of this application, after the user determines the first image, the processing described in the above embodiments can be performed based on the first image to obtain the second image. Then, a preset car model is rendered based on the second image to generate a target car model rendered with the second image. Subsequently, the target car model rendered with the second image can be displayed on a first interface, allowing the user to view the target car model and perform corresponding operations. The implementation of displaying the target car model on the first interface can be understood with reference to Figures 1 and 9 above, where the first interface can be understood as an interface used for processing the car model's painting.

[0147] Furthermore, the car model processing method provided in this application also supports users to perform corresponding user operations on the target car model in the first interface to adjust the painting effect of the target car model. The user operations performed on the target car model are described below with reference to Figures 10 to 14. Figure 10 is a schematic diagram of the interface of the car model processing method provided in the embodiment of this application (Figure 5), Figure 11 is a schematic diagram of the interface of the car model processing method provided in the embodiment of this application (Figure 6), Figure 12 is a schematic diagram of the interface of the car model processing method provided in the embodiment of this application (Figure 7), Figure 13 is a schematic diagram of the interface of the car model processing method provided in the embodiment of this application (Figure 8), and Figure 14 is a schematic diagram of the implementation of the image adjustment method provided in the embodiment of this application.

[0148] In one implementation, the electronic device may, for example, respond to a second operation acting on the target vehicle model to perform a first adjustment process on a second image displayed on the surface of the target vehicle model. The first adjustment process in this application may include at least one of the following: movement, rotation, cropping, and scaling. Several first adjustment processes described herein are described below.

[0149] Before the formal introduction, it should be noted that the first adjustment process described in this embodiment is performed on the second image displayed on the surface of the target car model. In one implementation, for example, the current first interface can be set to the editing mode of the painting effect. In this way, the second operation described above can be executed directly to achieve the first adjustment process on the second image.

[0150] Alternatively, in another implementation, it is not necessary to set an editing mode. Instead, the second image located on the surface of the target car model is selected first through a corresponding user operation, and then the second operation described above is performed on the second image to achieve the first adjustment processing of the second image. The user operation for selecting the second image can be, for example, a long press operation on the second image, or an operation on the second image with a preset sliding trajectory, etc., and this application does not limit it in this way.

[0151] Based on this, the following explanation, in conjunction with Figure 10, will focus on the case where the first adjustment process involves movement. As shown in Figure 10, for example, the second operation can be a drag operation on the second image displayed on the surface of the target car model, which can move the second image displayed on the surface of the target car model to the position required by the user.

[0152] Referring to Figure 10(a), the second image is initially rendered on one side door of the car model (this position can be the default, or it can be determined based on a preset mapping relationship). The user can then click on the second image displayed on the target car model and drag it. Figure 10(b) shows the intermediate display state during the dragging process. Then, for example, the user drags the second image to the position on the hood as shown in Figure 10(c) and releases the finger, thus moving the second image from the side door to the hood.

[0153] Furthermore, the internal implementation of the movement process can be understood by referring to Figure 14. Referring to Figure 14(a), during the rendering process, UV unwrapping is usually performed on the preset car model. After UV unwrapping, for example, a UV unwrapping diagram can be obtained, which is the UV unwrapping diagram of the preset car model shown in the figure. For ease of explanation, the UV unwrapping diagram of the preset car model will be referred to as 1400 in the following text.

[0154] During the rendering process, for example, a mapping relationship can be established between the image to be rendered and the UV unfolded map to determine the mapping relationship between each pixel in the image to be rendered and the texture coordinates of the preset car model. In this way, the model rendering can be achieved based on the mapping relationship. For example, we can refer to Figure 14(b) for understanding. For example, the mapping relationship between the second image 1401 and the UV unfolded map 1400 of the preset car model in this embodiment is shown in the figure. For example, the position covered by the second image 1401 corresponds to the side door area on the preset car model. In this way, the car model rendering effect shown in Figure 14(b) can be achieved.

[0155] Therefore, in the technical solution of this application, for example, in response to the second operation, a corresponding first adjustment process can be performed on the second image. For instance, in the current embodiment, the moving direction and moving distance of the second image can be determined based on the second operation, thereby performing corresponding moving processing on the second image. Subsequently, based on the moved second image, the mapping relationship between each pixel in the second image and the texture coordinates of the preset car model is updated, and then rendering is re-executed based on the updated mapping relationship, so that the second image displayed by the target car model can also complete the corresponding rendering processing.

[0156] For example, we can refer to Figure 14(b) and Figure 14(c) for understanding. For instance, based on the second operation, we can determine the direction and distance of movement of the second image, thereby moving the mapping relationship between the second image and the UV unfolded map from the position shown in Figure 14(b) to the position shown in Figure 14(c). Subsequently, we can re-render based on the mapping relationship shown in Figure 14(c) to achieve the target car model after moving the second image as shown in Figure 10.

[0157] Let's now understand the case where the first adjustment process is rotation, referring to Figure 11. As shown in Figure 11, for example, the second operation can be a two-finger rotation operation on the second image displayed on the surface of the target car model, which can rotate the second image displayed on the surface of the target car model by a certain angle.

[0158] Referring to Figure 11(a), assuming the second image is currently rendered on the hood of the car model, the user can then rotate the second image displayed on the target car model by two fingers, for example, to rotate the second image by a certain angle. The direction of rotation is indicated by arrows in the figure. The target car model after rotating the second image can then be displayed, for example, as shown in Figure 11(b). In the current implementation, for example, the angle of rotation by the user's two fingers can be detected to determine the required rotation angle of the second image displayed on the surface of the target car model.

[0159] Furthermore, the internal implementation of the rotation process can be understood by referring to Figure 14. The understanding of Figure 14(a) and Figure 14(b) is similar to that described above, and will not be repeated here.

[0160] The electronic device, for example, can respond to the second operation to perform a corresponding first adjustment process on the second image. For instance, in the current embodiment, the rotation angle of the second image can be determined based on the second operation, thereby performing a corresponding rotation process on the second image. Subsequently, based on the rotated second image, the mapping relationship between each pixel in the second image and the texture coordinates of the preset car model is updated, and then rendering is re-executed based on the updated mapping relationship, so that the second image displayed for the target car model can also complete the corresponding rendering process.

[0161] For example, we can refer to Figure 14(c) and Figure 14(d) for understanding. For instance, based on the second operation, the rotation angle of the second image can be determined, thereby moving the mapping relationship between the second image and the UV unfolded image from the position shown in Figure 14(c) to the position shown in Figure 14(d). Subsequently, based on the mapping relationship shown in Figure 14(d), we can re-render to achieve the target car model after rotating the second image as shown in Figure 11.

[0162] Furthermore, in implementing the rotation, besides rotating the second image using two fingers, multi-finger operations (such as three-finger rotation) can also be used to rotate the car model. This means that in this operation mode, the second image remains in its original position while the preset car model is rotated, thus achieving the corresponding rendering processing of the second image displayed on the target car model's surface. The specific methods for triggering the car model rotation and the specific operations during the rotation process can be expanded according to actual needs; this embodiment does not limit these.

[0163] The underlying logic of this implementation is similar to that described above. For example, the UV unfolded image of the preset car model can be rotated accordingly. Based on the rotated UV unfolded image, the mapping relationship between each pixel in the second image and the texture coordinates of the preset car model can be updated. Then, based on the updated mapping relationship, the rendering can be re-executed, so that the second image displayed by the target car model can also be rendered accordingly.

[0164] Let's now understand the first adjustment process as scaling, referring to Figure 12. As shown in Figure 12, for example, the second operation can be a two-finger scaling operation on the second image displayed on the surface of the target car model, which can adjust the size of the second image displayed on the surface of the target car model.

[0165] Referring to Figure 12(a), assuming the second image is currently rendered on the hood of a car model, the user can then zoom in and out of the displayed second image on the target car model using, for example, by pinching two fingers. This zooms in on the second image, and then, for example, displays the target car model with the second image enlarged, as shown in Figure 12(b). In the current implementation, for example, the distance of the user's pinch-finger zoom can be detected to determine the required scaling ratio of the second image displayed on the surface of the target car model.

[0166] Furthermore, the internal implementation of the rotation process can be understood by referring to Figure 14. The understanding of Figure 14(a) and Figure 14(b) is similar to that described above, and will not be repeated here.

[0167] The electronic device, for example, can respond to the second operation to perform a corresponding first adjustment process on the second image. For instance, in the current embodiment, the scaling ratio of the second image can be determined based on the second operation, and the second image can be scaled accordingly. Subsequently, based on the scaled second image, the mapping relationship between each pixel in the second image and the texture coordinates of the preset car model is updated, and then rendering is re-executed based on the updated mapping relationship, so that the second image displayed for the target car model can also complete the corresponding rendering process.

[0168] For example, we can refer to (c) and (e) in Figure 14 for understanding. For instance, based on the second operation, the scaling ratio of the second image can be determined, thereby moving the mapping relationship between the second image and the UV unfolded image from the position shown in (c) in Figure 14 to the position shown in (e) in Figure 14. Subsequently, based on the mapping relationship shown in (e) in Figure 14, we can re-render to achieve the target car model after scaling the second image as shown in Figure 12.

[0169] Let's now understand the first adjustment process as cropping in conjunction with Figure 13. As shown in Figure 13, for example, the second operation can be a sliding operation on the second image displayed on the surface of the target car model, which can crop the second image displayed on the surface of the target car model.

[0170] Referring to Figure 12(a), assuming the second image is rendered on the hood of a car model, the user can, for example, select a specific area of ​​the second image displayed on the target car model by sliding their finger. The selected area is then cropped. The box in the figure represents the selected area described here. The target car model after cropping the second image can then be displayed, for example, as shown in Figure 12(b). In the current implementation, for example, the cropping box defined by the user's sliding operation can be detected to determine the area of ​​the second image that needs to be retained on the surface of the target car model.

[0171] Furthermore, the internal implementation of the rotation process can be understood by referring to Figure 14. The understanding of Figure 14(a) and Figure 14(b) is similar to that described above, and will not be repeated here.

[0172] The electronic device, for example, can respond to the second operation to perform a corresponding first adjustment process on the second image. For instance, in the current embodiment, the second operation can be used to determine the area of ​​the second image to be retained, thereby performing a corresponding cropping process on the second image. Subsequently, based on the cropped second image, the mapping relationship between each pixel in the second image and the texture coordinates of the preset car model is updated, and then rendering is re-executed based on the updated mapping relationship, so that the second image displayed for the target car model can also complete the corresponding rendering process.

[0173] For example, we can refer to (c) and (f) in Figure 14 for understanding. For instance, based on the second operation, the area to be retained in the second image can be determined, thereby moving the mapping relationship between the second image and the UV unfolded image from the position shown in (c) in Figure 14 to the position shown in (f) in Figure 14. Subsequently, based on the mapping relationship shown in (f) in Figure 14, we can re-render to achieve the target car model after cropping the second image as shown in Figure 13.

[0174] It is also understood that in the technical solution of this application, during the first adjustment process, the above-described update mapping relationship and re-rendering process will be performed in real time. Accordingly, the display effect of the second image displayed on the surface of the target car model can be adjusted in real time on the first interface. This allows the user to quickly observe the specific adjustment situation and adjustment effect, thereby improving the speed and efficiency of paint effect adjustment.

[0175] Furthermore, it should be noted that the above illustrations all use a single image as an example for illustrating the second image. In another implementation, when semantic segmentation is performed, the second image can include multiple segmented sub-images, each corresponding to its own car model region. The specific implementation of the first adjustment process for the second image as a segmented sub-image is similar to that described above; the user simply selects a segmented sub-image of a specific region, and then the aforementioned processing is performed. This scenario will not be elaborated upon further here.

[0176] As can be understood from the above description, the technical solution of this application supports users to adjust the second image rendered on the surface of the target car model using specific gestures. This effectively improves the flexibility and convenience of adjusting the car model paint, and allows users to arbitrarily adjust the paint effect they need, thereby enhancing the user experience. In actual implementation, the specific gesture settings for the first operation corresponding to each first adjustment process can be set according to actual needs and are not limited to the technical solution described above.

[0177] The above embodiments describe the implementation of adjusting the position and / or size of the second image displayed on the surface of the target car model. Furthermore, the technical solution of this application can also support users in adjusting the image parameters of the second image displayed on the surface of the target car model, which will be described below. Figure 15 is a schematic diagram of the interface of the car model processing method provided in this application embodiment (Figure 9), Figure 16 is a schematic diagram of transparency adjustment provided in this application embodiment, and Figure 17 is a schematic diagram of the interface of the car model processing method provided in this application embodiment (Figure 10).

[0178] In one implementation, a first adjustment control can be displayed on a first interface, and then a second adjustment process can be performed on the second image displayed on the surface of the target car model in response to a third operation acting on the first adjustment control on the first interface. The second adjustment process is used to adjust at least one of the following image parameters: brightness, contrast, saturation, hue, color temperature, transparency, and blending mode.

[0179] As can be understood, the first adjustment control is the control used to adjust the image parameters described above, which can be understood in conjunction with Figure 15.

[0180] As shown in Figure 15(a), a color adjustment control 1501 may be displayed in the first interface. In response to user actions on the color adjustment control 1501, a color adjustment panel 1502 may be displayed, as shown in Figure 15(b). Referring to the figures, it can be understood that the color adjustment panel 1502 may contain first adjustment controls corresponding to each image parameter. For example, 1503 in Figure 15 represents the first adjustment control for the image parameter of transparency. The first adjustment control may include a movable slider. The user can then move the slider in the first adjustment control to adjust the corresponding image parameter. The understanding of the adjustment controls for each image parameter is similar and will not be repeated here.

[0181] Figure 15 shows one possible implementation of the first adjustment control. In actual implementation, the specific settings, operation methods, and trigger display methods of the first adjustment control can be arbitrarily set according to actual needs. This embodiment does not impose any restrictions on this.

[0182] Similar to the above description, in the process of implementing the second adjustment processing of the second image displayed on the surface of the target car model, for example, a third operation is performed on the first adjustment control to obtain the corresponding adjusted image parameters, and then the second image is adjusted based on the adjusted image parameters. Subsequently, the second image is rendered based on the second image with adjusted image parameters to realize the corresponding second adjustment processing of the second image displayed on the surface of the target car model.

[0183] Furthermore, during the second adjustment process, the image parameters of the updated image and the re-rendering process described above will be executed in real time. Consequently, the display effect of the second image on the surface of the target car model can be adjusted in real time on the first interface. This allows users to quickly observe the specific adjustment situation and effect, thereby improving the speed and efficiency of paint effect adjustment.

[0184] The following section provides additional information regarding transparency and blending modes in the image parameters.

[0185] As can be understood from the above description, in this embodiment, after determining the target model, the image parameters of the second image can be further adjusted according to the first adjustment control to improve the performance of the subsequent car model rendering based on the second image and to achieve the performance effect required by the user. The image parameters may include transparency parameters and / or blending modes.

[0186] The transparency parameter is used to adjust the transparency of the second image. It is understood that if the car model is rendered directly based on the original second image, the original color and texture of the car model will be obscured. Therefore, this application adjusts the transparency of the second image so that the color and texture of the car model can still be displayed after rendering, thus improving the rendering effect of the car model.

[0187] In the process of adjusting the transparency of the second image, for example, the transparency of the entire second image can be adjusted. For example, if the transparency is set to 30%, then the transparency of the entire second image can be adjusted to 30%. This way, after the second image is covered on the surface of the car model, the original color and texture of the car model can be effectively preserved.

[0188] Alternatively, the transparency of a preset area in the second image can be adjusted. The preset area can be, for example, the edge area of ​​the second image, or the center area of ​​the second image, or the area corresponding to a range preset for the image. This application does not limit this.

[0189] The following description uses the example of the preset area being the edge region of the second image, with reference to Figure 16. The second image 603 described above is still used as an example for illustration.

[0190] As shown in Figure 16, for example, the length direction of image 603 can be considered as the interval from 0% to 100%. Then, the regions corresponding to the intervals of 0% to 20% and 80% to 100% in the length direction can be understood as the preset regions described here. Specifically, the region corresponding to the 0% to 20% interval in the length direction is the region indicated by 1601 in the figure, and the region corresponding to the 80% to 100% interval in the length direction is the region indicated by 1602 in the figure. Regions 1601 and 1602 can be understood as corresponding to the left and right edges of the image, respectively.

[0191] The above description, in conjunction with Figure 16, illustrates the implementation of setting the left and right edge regions as preset regions. For example, the top and bottom edge regions can also be set as preset regions. The implementation for the top and bottom edge regions is similar to the above description, simply by adjusting the length direction to the width direction. Furthermore, the four edges of the image can also be set as preset regions; that is, the edge regions are determined by the sum of the length and width dimensions. This embodiment does not impose any restrictions on the specific selection of the preset regions.

[0192] Furthermore, when adjusting the transparency of a preset area, you can adjust the overall transparency of the preset area. Alternatively, you can adjust the transparency of the preset area gradually.

[0193] The following explanation, also in conjunction with Figure 16, assumes the current transparency parameter is set to 30%. In one implementation, the overall transparency of regions 1601 and 1602 can be set to 30%. Alternatively, in another implementation, the transparency of preset regions can be set to gradually change. For example, the transparency of a preset region can gradually change from the side furthest from the non-preset region to the side closest to the non-preset region. Referring to Figure 16, for example, the transparency of region 1601 can be set to transition from 30% to 0% from left to right, and the transparency of region 1602 can be set to transition from 30% to 0% from right to left. This achieves a smooth transition between the transparency of preset and non-preset regions, improving the final effect of the image painted on the car model. It is understood that the above descriptions, both the range values ​​corresponding to the preset regions and the transparency settings, are merely illustrative examples and can be adjusted according to actual needs during implementation.

[0194] In one implementation, the transparency parameter can be adjusted using a first adjustment control corresponding to the transparency parameter. For example, the transparency parameter can be adjusted using the first adjustment control 1503 described in Figure 15 above. For instance, if the transparency parameter is adjusted to 40%, then the implementation described in Figure 16 above will apply a 40% transparency parameter. For example, the overall transparency of regions 1601 and 1602 can be set to 40%. Alternatively, the transparency of region 1601 can be set to transition from 40% to 0% from left to right, and the transparency of region 1602 can be set to transition from 40% to 0% from right to left. The understanding is similar to that described above and will not be repeated here.

[0195] Furthermore, the technical solution of this application can also support users to adjust the preset area corresponding to the second image through specific gesture operations. Referring to the above description, for example, the length direction of image 603 can be considered as a range of 0% to 100%. Then, the areas corresponding to the 0% to 20% and 80% to 100% ranges in the length direction can be understood as the preset area described here. Such a preset area can be pre-set, and the user can then adjust the range of the preset area through gesture operations.

[0196] Referring to Figure 17(a), for example, a user can perform a sliding operation along the length of the second image on the target car model to adjust a preset area of ​​the second image to which the transparency parameter will be applied. For example, the adjustment mode for the current preset area can be set first, and then the range of the preset area can be adjusted based on the sliding operation. Alternatively, a preset gesture or preset operation can be used to first trigger the adjustment operation of the preset area, and then the range of the preset area can be adjusted based on the sliding operation. The specific operation method for adjusting the range of the preset area can be set according to actual needs.

[0197] For example, the adjustment method for a preset area can be determined based on the user's sliding direction and distance along the length direction. For instance, if the sliding direction is from the image edge towards the image center, the preset area can be expanded. Conversely, if the sliding direction is from the image center towards the image edge, the preset area can be reduced. Furthermore, the proportion of expansion or reduction of the preset area can be determined based on the user's sliding distance to achieve adjustment of the preset area.

[0198] For example, you can refer to Figure 17 for understanding. Suppose that the user has slid a certain distance from the edge of the image toward the center of the image. Then, for example, the preset area can be adjusted to the area corresponding to the interval 1701 of 0% to 30% and the interval 1702 of 70% to 100% in the length direction as shown in Figure 17(b). And suppose that the current transparency parameter can be 40%. The understanding is similar to that described above, and will not be repeated here.

[0199] As can be seen from the illustration, in order to ensure the aesthetics of the final coating effect, the preset areas on both sides of the second image can be symmetrically adjusted based on the user's operation. Alternatively, the preset area on only the side operated by the user can be adjusted; this embodiment does not limit this.

[0200] As can be understood from the above description, by setting the transparency of a preset area, the display effect of the image on the car model can be improved on the one hand, and the color and texture of the car model itself can be displayed on the other hand.

[0201] In addition, image parameters may also include a blending mode, which indicates how the colors of the image and the car model are blended. For example, a blending mode may include at least one of the following: Darken mode, Multiply, Color Burn mode, Linear Burn mode, Lighten mode, etc. Specific options for the blending mode can be understood by referring to relevant implementations; this embodiment does not limit this.

[0202] In the technical solution of this application, the second image can be adjusted based on a selected blending mode so that when the adjusted second image is rendered on the surface of the car model, it can exhibit a corresponding blending effect. In one implementation, since the blending mode usually requires the mixing calculation of two colors, the calculation corresponding to the selected blending mode can be performed based on the preset color of the car model (which can be the default color or a color selected by the user) and the second image, thereby adjusting the image parameters of the second image so that when the adjusted second image is rendered on the surface of the car model, it can directly exhibit the effect of the corresponding blending mode. The image parameters may include at least one of the following: color, brightness, and contrast.

[0203] The above embodiments describe the implementation of rendering a preset car model based on a second image to display the target car model with the rendered second image on the first interface, and then making corresponding adjustments to the second image displayed on the surface of the target car model. In another implementation, the technical solution of this application can also provide the user with the first adjustment control described above before displaying the target car model, so that the user can adjust the image parameters of the second image according to actual needs before performing the rendering process. This ensures that the target car model obtained in the first rendering is the effect presented after the user adjusts the colors, thereby improving the flexibility of user operation.

[0204] The implementation is similar to that described above. For example, after the user selects the first image, the interface can display the first adjustment controls corresponding to each image parameter described above. Subsequently, the user adjusts the image parameters based on the first adjustment controls. Then, the electronic device will adjust the second image accordingly based on the user's operation on the first adjustment controls, and then perform rendering processing. The implementation method is similar to that described above, and will not be repeated here.

[0205] The electronic device described in the above embodiments can be a vehicle infotainment system or a mobile phone; this application does not limit this. In one implementation, after a user performs the above operation on the vehicle infotainment system to paint the vehicle model, the system can, for example, synchronize the painted target vehicle model to the mobile phone via the cloud, so that the vehicle model displayed in the mobile phone's application also exhibits the corresponding painting effect. Furthermore, after a user performs the above operation on the mobile phone to paint the vehicle model, the mobile phone can, for example, synchronize the painted target vehicle model to the vehicle infotainment system via the cloud, so that the vehicle model displayed on the system also exhibits the corresponding painting effect. Besides vehicle infotainment systems and mobile phones, any electronic device capable of displaying a vehicle model can use the implementation described herein to synchronize the target vehicle model, so that the painting effect determined by the user can be synchronized across various devices.

[0206] Based on the above introduction, the car model processing method provided in this application will be described in general with reference to Figure 18 and the software architecture of the electronic device. Figure 18 is a schematic diagram of the software architecture of the electronic device provided in the embodiment of this application.

[0207] As shown in Figure 18, an electronic device may include, for example, an interaction layer, an algorithm processing layer, and a system layer.

[0208] The interaction layer may include a gallery application and an interaction unit. For example, an electronic device can access a gallery application to determine the first image described above. During access to the gallery application, the device can access its local storage or cloud storage to obtain the first image. The interaction unit parses user-input gestures, which may include various user operations described in the embodiments above. The interaction unit can also determine adjusted image parameters based on user operations. Gesture parsing and parameter adjustment can be performed before or after rendering the target car model; this application does not limit this. In one implementation, user gesture operations and parameter adjustments can be supported both before and after rendering the target car model to adjust the corresponding painting effect.

[0209] Furthermore, the algorithm processing layer may include an image processing unit and a fusion unit. The image processing unit may include, for example, a first model and a second model. Referring to the description in the above embodiments, the first model and the second model are models used for compositional adjustments to the first image. Specific implementation details can be understood by referring to the above embodiments and will not be repeated here. The fusion unit is used to render the car model based on the second image, the 3D car model, and user-supported adjustments to obtain the painted target model, and can also adjust the painted target model.

[0210] It should be noted that both the first model and the second model in this application are models obtained after training. The training process of the first model may include, for example:

[0211] The first model is trained based on multiple sets of sample data. Each set of sample data includes: sample images and label images. The image size of the label images meets the preset size requirements, and the label images contain the main objects that exist in the sample images.

[0212] In the specific training process, for example, sample images can be input into the first model so that the first model outputs training output images. Then, based on the training output images and label images, a loss value is determined. Finally, the model parameters of the first model are updated based on the loss value to obtain the trained first model.

[0213] Furthermore, the training process for the second model is similar, and will not be elaborated here.

[0214] Furthermore, the system layer can include a GPU and a rendering engine, which can provide underlying support for the image processing and vehicle model rendering described above. After rendering is complete, the display unit can be invoked to display the target vehicle model on the screen of the electronic device. In the case where the electronic device is an in-vehicle infotainment system, the screens of the electronic device can include, for example, an instrument panel screen, a central control screen, and a passenger-side screen. In this case, multi-screen display of these screens can be achieved through the display unit.

[0215] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation portals are provided for users to choose to authorize or refuse.

[0216] This application provides an electronic device, comprising: a processor and a memory; the memory storing computer-executable instructions; the processor executing the computer-executable instructions stored in the memory, causing the electronic device to perform the aforementioned method. The electronic device may be, for example, a vehicle.

[0217] This application provides a chip. The chip includes a processor, which is used to call a computer program in memory to execute the technical solutions in the above embodiments. Its implementation principle and technical effects are similar to those in the related embodiments described above, and will not be repeated here.

[0218] This application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, it implements the methods described above. The methods described in the above embodiments can be implemented wholly or partially by software, hardware, firmware, or any combination thereof. If implemented in software, the functionality can be stored as one or more instructions or code on or transmitted over the computer-readable medium. The computer-readable medium can include computer storage media and communication media, and can also include any medium that can transfer a computer program from one place to another. The storage medium can be any target medium accessible by a computer.

[0219] In one possible implementation, a computer-readable medium may include random access memory (RAM), read-only memory (ROM), compact disc read-only memory (CD-ROM) or other optical disc storage, magnetic disk storage or other magnetic storage devices, or any other medium intended to carry or store required program code in the form of instructions or data structures, and accessible by a computer. Furthermore, any connection is appropriately referred to as a computer-readable medium. For example, if software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. As used herein, disks and optical discs include optical discs, laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray discs, where disks typically reproduce data magnetically, while optical discs optically reproduce data using lasers. Combinations of the above should also be included within the scope of computer-readable media.

[0220] This application provides a computer program product, which includes a computer program that, when run, causes a computer to perform the above-described method.

[0221] It should be noted that the modules or components described in the above embodiments can be one or more integrated circuits configured to implement the above methods, such as one or more application-specific integrated circuits (ASICs), one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs), etc. Furthermore, when a module is implemented through processing element scheduler code, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processors capable of calling program code, such as a controller. Additionally, these modules can be integrated together to implement a system-on-a-chip (SOC).

[0222] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., a solid-state disk (SSD)).

[0223] The term "multiple" in this document refers to two or more. The term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Furthermore, the character " / " in this document generally indicates an "or" relationship between the preceding and following related objects; in formulas, " / " indicates a "division" relationship. Additionally, it should be understood that in the description of this application, words such as "first" and "second" are used only for descriptive purposes and should not be construed as indicating or implying relative importance or order.

[0224] It is understood that the various numerical designations used in the embodiments of this application are merely for descriptive convenience and are not intended to limit the scope of the embodiments of this application.

[0225] It is understood that, in the embodiments of this application, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

Claims

1. A vehicle model processing method characterized by comprising: Applied to vehicles, the method includes: The response is a first operation to determine the first image, displaying a target car model with a second image rendered in a first interface, wherein the first image is determined in the local storage of the vehicle or the first image is generated based on a voice command, the image size of the second image meets a preset size requirement, and the second image contains the main object present in the first image; In response to a second operation applied to the target car model, a first adjustment process is performed on the second image displayed on the surface of the target car model, the first adjustment process including at least one of the following: moving, rotating, cropping, and scaling; In response to a third operation applied to the first adjustment control in the first interface, a second adjustment process is performed on the second image displayed on the surface of the target car model. The second adjustment process is used to adjust at least one of the following image parameters: brightness, contrast, saturation, hue, color temperature, transparency, and blending mode.

2. The method of claim 1, wherein, The response is used to determine a first operation of the first image, displaying a target car model rendered with a second image in a first interface, including: The response is used to determine a first operation of the first image, to perform a compositional adjustment on the first image to obtain at least one second image; The preset car model is rendered based on the at least one second image, so as to display the target car model with the rendered second image in the first interface.

3. The method of claim 2, wherein, The step of adjusting the composition of the first image to obtain at least one second image includes: The first image is input into the first model so that the first model outputs the second image; wherein the first model is used to perform the first adjustment process on the first image, and the first model is a model trained based on at least one set of sample data.

4. The method of claim 2, wherein, The step of adjusting the composition of the first image to obtain at least one second image includes: The first image is semantically segmented according to the second model to obtain segmented sub-images corresponding to at least one target object contained in the first image, wherein the segmented sub-images obtained after semantic segmentation are the second image.

5. The method of claim 4, wherein, The method further includes: A second control is displayed on the first interface, the second control being used to determine whether the semantic segmentation operation needs to be performed; The step of performing semantic segmentation on the first image according to the second model to obtain segmented sub-images corresponding to at least one target object contained in the second image includes: In response to a fourth operation applied to the second control, the first image is semantically segmented according to the second model to obtain segmented sub-images corresponding to at least one target object contained in the second image.

6. The method according to claim 4 or 5, characterized in that, The step of rendering a preset car model based on the at least one second image to display the target car model with the rendered second image on a first interface includes: For any one of the segmented sub-images, determine the target region in the preset car model corresponding to the segmented sub-image; Based on the segmented sub-image, the target region in the preset car model is rendered to generate the target car model, and the surface of the target region of the target car model is covered with the segmented sub-image.

7. The method of claim 6, wherein, Determining the target region in the preset car model corresponding to the segmented sub-image includes: Obtain the first semantic type of the target object contained in the segmented sub-image; According to the preset mapping relationship, the vehicle model area corresponding to the first semantic type is determined as the target area. The preset mapping relationship is used to indicate the vehicle model area corresponding to each of the multiple semantic types.

8. The method according to any one of claims 1 to 7, characterized in that, The response to the second operation acting on the target car model involves performing a first adjustment process on the second image displayed on the surface of the target car model, including: In response to the second operation, the first adjustment process is performed on the second image, and based on the second image after the first adjustment process, the mapping relationship between each pixel in the second image and the texture coordinates of the preset car model is updated; Based on the updated mapping relationship, the preset car model is re-rendered based on the second image to perform a first adjustment process on the second image displayed on the surface of the target car model; wherein, during the execution of the first adjustment process, the display effect of the second image on the surface of the target car model is adjusted in real time.

9. The method according to any one of claims 1 to 8, characterized in that, The first operation is to select an image in the vehicle's local storage; or, the first operation is to input a voice command to instruct the acquisition of the first image.

10. A vehicle characterized by comprising: The vehicle includes: one or more processors and memory; The memory is coupled to the one or more processors, the memory being used to store computer program code, the computer program code including computer instructions, the one or more processors invoking the computer instructions to cause the vehicle to perform the method as described in any one of claims 1-9.

11. A chip system, characterized by The chip system is applied to a vehicle, and the chip system includes one or more processors, the one or more processors being used to invoke computer instructions to cause the vehicle to perform the method as described in any one of claims 1-9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes computer instructions that, when executed on a vehicle, cause the vehicle to perform the method as described in any one of claims 1-9.

13. A computer program product, characterised in that, The computer program product includes computer program code that, when run on a vehicle, causes the vehicle to perform the method as described in any one of claims 1-9.