Image generation method and system, vehicle and storage medium

Abstract

The invention relates to the technical field of vehicle-mounted painting, and discloses an image generation method and system, a vehicle and a storage medium. A vehicle-mounted terminal obtains a target image description text input by a user and obtains a first environment image set collected by a vehicle-mounted image collection device; extracting the feature description words of each environment image in the first environment image set, then matching the target image description text with the feature description words of each environment image, and providing the successfully matched second environment image set to the user for the user to select an environment reference image; and finally, the in-vehicle terminal sends the environment reference image selected by the user and the target image description text to the cloud, the cloud generates a target image by using an AI drawing model based on the environment reference image and the target image description text, and feeds back the target image to the in-vehicle terminal to be displayed to the user, so that the user can experience the drawing creation fun in the driving process, and the user experience is improved. And the interestingness in the driving process is increased.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle-mounted painting, and in particular to an image generation method, system, vehicle and storage medium. Background Art

[0002] Car computer refers to the abbreviation of the in-vehicle infotainment product installed in the car. The car computer can realize the information communication between people and cars, and cars and the outside world (cars and cars). At present, most car computers only provide basic services such as navigation and music, and the form of interaction with users is single. With the development of artificial intelligence (AI), AI painting large models can generate higher-quality works, and with the improvement of model performance, people can perform real-time painting, which will bring novel and interesting experiences to car computer users and provide people with emotional value. Therefore, how to apply AI painting technology to the car computer end and improve user experience has become a problem that needs to be solved urgently. Summary of the invention

[0003] In view of this, the present invention provides an image generation method, system, vehicle and storage medium to solve the problem of how to apply AI painting technology to the vehicle side to improve the user experience.

[0004] In a first aspect, the present invention provides an image generation method, which is applied to a vehicle-mounted terminal, and the method includes: obtaining a target image description text input by a user after entering an image drawing interface; obtaining a first environmental image set acquired by a vehicle-mounted image acquisition device, and extracting feature description words for each environmental image in the first environmental image set; matching the target image description text with the feature description words for each environmental image, and providing a second environmental image set that has been successfully matched to a user, so that the user can select an environmental reference image from the second environmental image set; sending the environmental reference image selected by the user and the target image description text to a cloud, so that the cloud generates a target image based on the environmental reference image and the target image description text, and feeds it back to the vehicle-mounted terminal for display to the user.

[0005] The image generation method provided by the present invention comprises the following steps: the vehicle computer end obtains a target image description text input by a user, obtains a first environmental image set collected by a vehicle-mounted image acquisition device, extracts feature description words of each environmental image in the first environmental image set, matches the target image description text with the feature description words of each environmental image, provides a second environmental image set with successful matching to the user, and allows the user to select an environmental reference image. The vehicle computer end finally sends the environmental reference image and the target image description text selected by the user to the cloud. The cloud generates the target image by using an AI painting model based on the environmental reference image and the target image description text, and feeds back to the vehicle computer end for display to the user, thereby enabling the user to experience the fun of painting creation during driving and increasing the fun of driving.

[0006] In an optional embodiment, the method also includes: in response to a user selecting an instruction to generate facial features, displaying a picture upload interface to the user, the picture upload interface including at least a local image upload option, a mobile upload option, and a shooting option, so that the user can select a corresponding image upload method to upload the image; receiving a facial image uploaded by the user based on the local image upload option or the shooting option, and sending the environment reference image, the target image description text, and the facial image to the cloud, so that the cloud generates the target image based on the environment reference image, the target image description text, and the facial image.

[0007] The image generation method designed by the present invention, in addition to collecting images inside and outside the car window, will also incorporate the user's personal image, and the reference image generation conditions are richer, so that the target image generated by AI painting can better meet the needs of users.

[0008] In an optional embodiment, the method also includes: in response to the user selecting the mobile terminal upload option, displaying a link identifier that can be connected to the mobile terminal to the user, so that the user can synchronize the facial image stored or taken in the mobile terminal to the vehicle terminal based on the link identifier; receiving the facial image uploaded by the mobile terminal, and executing the step of sending the environmental reference image, the target image description text and the facial image to the cloud.

[0009] The method of selecting facial images of the present invention not only includes the local storage of the vehicle terminal, but also can establish a connection with the mobile terminal to obtain the facial images in the mobile terminal. The acquisition method is more comprehensive, which improves the entertainment experience of the user during driving.

[0010] In an optional embodiment, the method further includes: in response to the user's modification operation on the target image, feeding back the modification operation on the target image to the cloud, so that the cloud generates a modified target image based on the modification operation on the target image, and feeds back to the vehicle end for display to the user.

[0011] The AI ​​painting module of the present invention generates and transmits back to the device according to the user's operation on the image, so that the user can get real-time feedback, which can bring great fun to the user.

[0012] In a second aspect, the present invention provides an image generation method, which is applied to a cloud, and the cloud is integrated with an AI painting model. The method includes: receiving an environment reference image and a target image description text selected by a user and sent by a vehicle terminal, and generating a target image based on the environment reference image and the target image description text, wherein the environment reference image is an environment reference image selected by the user from a second environment image set; the second environment image set is a feature description word extracted by the vehicle terminal for each environment image in a first environment image set, matching the feature description words of each environment image with the target image description text, and providing the matched second environment image set to the user; the first environment image set is a first environment image set acquired by a vehicle-mounted image acquisition device acquired by the vehicle terminal; the target image description text is a target image description text entered by the user after entering an image painting interface acquired by the vehicle terminal; and the generated target image is sent to the vehicle terminal so that the vehicle terminal displays the target image to the user.

[0013] In an optional embodiment, the cloud is integrated with an image style set, which includes multiple image styles and multiple image art forms under each image style. The method also includes: extracting target style description words from the target image description text, and determining the target image style based on the target style description words; generating target images of different image art forms based on the environment reference image, the target image description text, the target image style, and the multiple image art forms under the target image style.

[0014] The cloud of the present invention can also expand the image art form based on the target image description style input by the user, so that the generated target image is in a variety of image art forms, and the style will not be monotonous after multiple image generation, thereby improving the user's entertainment interest.

[0015] In an optional embodiment, the method further includes: if the target style description word is not extracted from the target image description text, randomly selecting a target image style from the image style set, and generating target images of different image art forms based on the environment reference image, the target image description text, the target image style and multiple image art forms under the target image style.

[0016] If the present invention fails to extract the target style description words, the target image style can also be randomly selected, and then based on a variety of image art forms under the target image style, target images of different image art forms are generated to ensure the style diversity of the target image.

[0017] In an optional embodiment, before generating the target image based on the environmental reference image and the target image description text, the method also includes: resizing the environmental reference image based on a preset size requirement; and / or, performing high-definition restoration processing on the environmental reference image; and / or, performing image quality enhancement processing on the environmental reference image.

[0018] The present invention performs image preprocessing on environmental reference images and facial images to ensure that the AI ​​painting model can generate high-quality target images based on the reference images.

[0019] In an optional embodiment, the method also includes: obtaining calendar data for the day; based on the calendar data, determining whether the day is a holiday; if the day is a holiday, generating a target image related to the holiday based on the holiday, the environmental reference image and the target image description text.

[0020] The present invention adds information related to holidays of the day to the target image based on the calendar information of the day, so that the generated target image is more in line with the driving scene of the day and better meets the needs of users.

[0021] In a third aspect, the present invention provides an image generation system, which includes a vehicle terminal and a cloud, wherein the vehicle terminal obtains a target image description text input by a user after entering an image drawing interface, obtains a first environmental image set acquired by a vehicle-mounted image acquisition device, and extracts feature description words of each environmental image in the first environmental image set; the vehicle terminal matches the target image description text with the feature description words of each environmental image, and provides a second environmental image set that successfully matches to the user, so that the user can select an environmental reference image from the second environmental image set; the vehicle terminal sends the environmental reference image selected by the user and the target image description text to the cloud; the cloud receives the environmental reference image and target image description text selected by the user sent by the vehicle terminal, generates a target image based on the environmental reference image and the target image description text, and feeds it back to the vehicle terminal for display to the user.

[0022] In a fourth aspect, the present invention provides a vehicle, comprising a vehicle terminal, wherein the vehicle terminal comprises a memory and a processor, the memory and the processor are communicatively connected to each other, computer instructions are stored in the memory, and the processor executes the image generation method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0023] In a fifth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the image generation method of the first aspect or any corresponding embodiment thereof, or to execute the image generation method of the second aspect or any corresponding embodiment thereof.

[0024] The image generation system provided by the present invention, the vehicle computer end obtains the target image description text input by the user, obtains the first environmental image set collected by the vehicle-mounted image acquisition device, and extracts the feature description words of each environmental image in the first environmental image set, and then matches the target image description text with the feature description words of each environmental image, and provides the user with a second environmental image set that matches successfully. The vehicle computer end finally sends the environmental reference image and the target image description text selected by the user to the cloud. The cloud generates the target image by using the AI ​​painting model based on the environmental reference image and the target image description text, and feeds it back to the vehicle computer end for display to the user, so that the user can experience the fun of painting creation during driving, which increases the fun of driving. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0026] Figure 1 is a schematic structural diagram of an image generation system according to an embodiment of the present invention;

[0027] Figure 2 is an example diagram of specific module composition of the image generation system according to an embodiment of this aspect;

[0028] Figure 3 is a schematic diagram of interaction between a vehicle terminal and a cloud according to an embodiment of the present invention;

[0029] Figure 4 is a flowchart of an image generating method according to an embodiment of the present invention;

[0030] Figure 5 is a structural example diagram of a vehicle according to an embodiment of the present invention;

[0031] Figure 6 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0033] The embodiment of the present invention provides an image generation system, such as Figure 1 As shown, the system includes: a vehicle terminal 1 and a cloud 2, wherein the vehicle terminal 1 obtains the target image description text input by the user after entering the image drawing interface, obtains the first environmental image set collected by the vehicle-mounted image acquisition device, and extracts the characteristic description words of each environmental image in the first environmental image set; the vehicle terminal 1 matches the target image description text with the characteristic description words of each environmental image, and provides the second environmental image set with successful matching to the user, so that the user can select an environmental reference image from the second environmental image set; the vehicle terminal 1 sends the environmental reference image and the target image description text selected by the user to the cloud 2; the cloud 2 receives the environmental reference image and the target image description text selected by the user sent by the vehicle terminal 1, generates a target image based on the environmental reference image and the target image description text, and feeds it back to the vehicle terminal 1 for display to the user.

[0034] like Figure 2 As shown, the vehicle terminal 1 of the embodiment of the present invention is integrated with an AI painting application. The user can choose to click on the AI ​​painting application on the vehicle display screen based on entertainment needs, wherein the vehicle display screen can be installed in the co-pilot or back seat of the car, etc., to provide the user with painting entertainment projects, and in response to the user clicking on the AI ​​painting application, an image painting interface is displayed on the vehicle display screen. The user can input a text description of the target image on the vehicle display screen, and can also trigger the voice input function integrated in the vehicle terminal 1 to perform a voice description of the target image. The vehicle terminal 1 responds to the voice description of the target image input by the user, and can use a voice-to-text algorithm to convert the voice description into a text description of the target image, as an example only.

[0035] The vehicle terminal 1 can obtain a first environmental image set collected by a vehicle-mounted image acquisition device, wherein the image acquisition device may include a camera installed outside the vehicle to capture images of the surrounding environment during the driving of the vehicle, such as scenery, buildings and road conditions, etc., and may also include an in-vehicle camera to capture the environment inside the vehicle, such as seat layout and accessories, etc., as an example only, and can extract feature description words for each environmental image in the first environmental image set, wherein the method for extracting the feature description words of the environmental image is not limited, such as feature extraction of the environmental image, object detection, semantic analysis of the image through a deep learning model, converting the semantic analysis result into a natural language description, or inferring the description of the environmental image through an inverse prompt word model, as an example only; then the text description of the target image input by the user and the feature description of each environmental image are combined. The feature description words are matched, and the text feature matching degree is calculated. Then, the second environment image set that has been matched successfully can be displayed on the vehicle display screen, so that the user can select the desired environment reference image, wherein the method for determining whether the match is successful can be to screen out environment images whose feature matching degree is greater than a preset matching degree threshold, or to sort the feature matching degrees from large to small, and screen out a preset number of environment images. For example, if the user's description text of the target image includes requirements such as trees, sky, color, and objects, then environment images with words such as trees, sky, and color in the feature description words can be screened out, just as an example; the vehicle terminal 1 can also directly display the environment image set collected by the vehicle-mounted image acquisition device on the vehicle display screen for the user to select the environment reference image, just as an example, and finally send the environment reference image selected by the user and the target image description text to the cloud 2.

[0036] like Figure 2 As shown, the cloud 2 of the embodiment of the present invention is integrated with an AI painting module, wherein the AI ​​painting model integrated with the AI ​​painting module can be set and updated in real time based on actual field requirements and Wenshengtu model updates, such as using an intelligent AI drawing generation tool (Stable Diffusion XL) to generate a target image based on the target image description text and the environmental reference image, just as an example. The cloud 2 generates the target image using the AI ​​drawing generation tool based on the environmental reference image and the target image description text sent by the vehicle terminal 1, and returns it to the vehicle terminal 1 on the vehicle display screen to be displayed to the user.

[0037] The image generation system provided by the present invention, the vehicle computer end obtains the target image description text input by the user, obtains the first environmental image set collected by the vehicle-mounted image acquisition device, and extracts the feature description words of each environmental image in the first environmental image set, and then matches the target image description text with the feature description words of each environmental image, and provides the user with a second environmental image set that matches successfully. The vehicle computer end finally sends the environmental reference image and the target image description text selected by the user to the cloud. The cloud generates the target image by using the AI ​​painting model based on the environmental reference image and the target image description text, and feeds it back to the vehicle computer end for display to the user, so that the user can experience the fun of painting creation during driving, which increases the fun of driving.

[0038] In an optional embodiment, the vehicle terminal 1 responds to the user's selection of an instruction to generate facial features by presenting a picture upload interface to the user, where the picture upload interface includes at least a local image upload option, a mobile upload option, and a shooting option, so that the user can select a corresponding image upload method to upload the image; the vehicle terminal 1 receives the facial image uploaded by the user based on the local image upload option or the shooting option, and sends the environmental reference image, the target image description text, and the facial image to the cloud 2, so that the cloud 2 generates the target image based on the environmental reference image, the target image description text, and the facial image.

[0039] The AI ​​painting application integrated in the vehicle terminal 1 of the embodiment of the present invention can also provide a function of generating facial features of characters. If the user wants to generate a picture in which the character has his own facial features, the user can select the instruction to generate facial features in the image painting interface. After the vehicle terminal 1 responds to the user's selection of the instruction to generate facial features, it can display the picture upload interface to the user, wherein the picture upload interface may include but is not limited to a local image upload option, a mobile upload option and a shooting option, so that the user can select a corresponding image upload method to upload the image, wherein the mobile terminal can be a mobile phone terminal, a tablet, etc. that establishes a Bluetooth connection with the vehicle terminal. For example, if the user selects the local image upload option, the image stored locally in the vehicle terminal can be displayed on the vehicle terminal display screen, so that the user can select the image to be uploaded; if the user selects the shooting option, the camera application pops up to allow the user to take a frontal photo of the face. For example, the environment reference image, the target image description text and the facial image can be sent to the cloud 2, and the cloud 2 generates a facial image based on the environment reference image, the target image description text and the facial image sent by the vehicle terminal 1.

[0040] The image generation method designed by the present invention, in addition to collecting images inside and outside the car window, will also incorporate the user's personal image, and the reference image generation conditions are richer, so that the target image generated by AI painting can better meet the needs of users.

[0041] In an optional embodiment, in response to the user selecting the mobile terminal upload option, the vehicle terminal 1 displays a link identifier that can be connected to the mobile terminal to the user, so that the user can synchronize the facial image stored or taken in the mobile terminal to the vehicle terminal 1 based on the link identifier; the vehicle terminal 1 receives the facial image uploaded by the mobile terminal, and executes the step of sending the environmental reference image, the target image description text and the facial image to the cloud 2.

[0042] In the embodiment of the present invention, if the user chooses to upload the facial image through the mobile terminal, the vehicle terminal 1 can display the link identifier that can be connected to the mobile terminal to the user, wherein the way of displaying the link identifier to the user is not limited, and can be a pop-up QR code, which contains a vehicle identification number (VIN) and a 64-bit random number. The user scans the QR code with the mobile terminal, and the mobile phone jumps to an H5 page. On this page, the user can choose to take a photo or an image in the photo gallery. After the selection is completed, click OK to synchronize the stored or photographed facial image to the vehicle terminal 1. The vehicle terminal 1 can also respond to the user selecting the mobile terminal upload option, obtain the mobile terminal identification information pre-registered with its own vehicle terminal 1, and send the link of the H5 page to the mobile terminal via Bluetooth, so that the user clicks the link to display the H5 page, and finally can choose to take a photo or an image in the photo gallery. As an example only, after the vehicle terminal 1 receives the facial image uploaded by the mobile terminal, it can send the environmental reference image, the target image description text and the facial image to the cloud 2.

[0043] The method of selecting facial images of the present invention not only includes the local storage of the vehicle terminal, but also can establish a connection with the mobile terminal to obtain the facial images in the mobile terminal. The acquisition method is more comprehensive, which improves the entertainment experience of the user during driving.

[0044] In an optional embodiment, the cloud 2 is integrated with an image style set, wherein the image style set includes multiple image styles and multiple image art forms under each image style. The cloud 2 extracts target style description words from the target image description text, and determines the target image style based on the target style description words; based on the environment reference image, the target image description text, the target image style, and the multiple image art forms under the target image style, generates target images of different image art forms.

[0045] In the embodiment of the present invention, the image style set integrated in the cloud 2 includes multiple image styles and multiple image art forms under each image style, wherein the image style and the image art form can be updated in real time based on the existing style, and the image style may include but is not limited to cartoon style, realistic style, oil painting style, simple style, etc. The multiple image art forms under the oil painting style may include impressionist painting, loose brush work, vibrant color, light and shadow play, pointillism, etc., and each image style will preset multiple expansions of different image art forms.

[0046] After receiving the target image description text sent by the vehicle terminal 1, the cloud 2 can extract the target style description words from the target image description text. For example, if a part of the target image text is "I need an image in oil painting style" or "an image with rich colors and strong three-dimensional sense", the target style description words that can be extracted from the text are "oil painting" or "rich colors and strong three-dimensional sense". Based on the target style description words, it can be determined that the target image style is oil painting style. Then, based on the environmental reference image, the target image description text, the target image style and various image art forms under the target image style, target images of different image art forms can be generated.

[0047] The cloud of the present invention can also expand the image art form based on the target image description style input by the user, so that the generated target image is in a variety of image art forms, and the style will not be monotonous after multiple image generation, thereby improving the user's entertainment interest.

[0048] In an optional embodiment, if the cloud 2 fails to extract the target style description words from the target image description text, it randomly selects the target image style from the image style set, and generates target images of different image art forms based on the environment reference image, the target image description text, the target image style, and multiple image art forms under the target image style.

[0049] In the embodiment of the present invention, if the cloud 2 fails to extract the target style description words from the target image description text, it can randomly select the target image style from a preset image style set, and then generate target images of different image art forms based on the environmental reference image, the target image description text, the target image style and multiple image art forms under the target image style.

[0050] If the present invention fails to extract the target style description words, the target image style can also be randomly selected, and then based on a variety of image art forms under the target image style, target images of different image art forms are generated to ensure the style diversity of the target image.

[0051] In an optional embodiment, before generating a target image based on the environment reference image and the target image description text, the cloud 2 resizes the environment reference image based on a preset size requirement; and / or performs high-definition restoration processing on the environment reference image; and / or performs image quality enhancement processing on the environment reference image.

[0052] In order to enable the AI ​​painting model to generate a high-quality target image based on the image sent by the vehicle terminal 1, the cloud 2 of the embodiment of the present invention may perform image preprocessing on reference images such as environmental reference images and facial images. For example, the width and height of the reference images may be adjusted to ensure that all reference images are within a suitable size range. For example, if the AI ​​painting model uses the Stable Diffusion XL model and the corresponding reference image size is close to 1024, the reference image may be resized based on the size standard of 1024. As an example only, a high-definition algorithm may be used to perform high-definition restoration on the reference image. The brightness, contrast and color balance of the reference image may also be adjusted to ensure that the quality of the reference image is suitable for subsequent AI painting processing. As an example only.

[0053] The present invention performs image preprocessing on environmental reference images and facial images to ensure that the AI ​​painting model can generate high-quality target images based on the reference images.

[0054] In an optional implementation, the cloud 2 obtains the calendar data of the day; based on the calendar data, determines whether the day is a holiday; if the day is a holiday, generates a target image related to the holiday based on the holiday, the environmental reference image and the target image description text.

[0055] The cloud 2 of the embodiment of the present invention can also obtain the calendar data of the day, and determine whether the day is a holiday based on the calendar data, such as the Dragon Boat Festival, Mid-Autumn Festival or National Day. If it is determined that the day is the Dragon Boat Festival, relevant festival description words can be added, such as "Dragon Boat Festival-related images", or the Lora model of the relevant festival can be trained in advance (which can be used in combination with the large model to interfere with the results produced by the large model), and the Lora model and the AI ​​painting model can be used together to generate target images related to holidays, as an example only.

[0056] The present invention adds information related to holidays of the day to the target image based on the calendar information of the day, so that the generated target image is more in line with the driving scene of the day and better meets the needs of users.

[0057] In an optional embodiment, the vehicle terminal 1 responds to the user's modification operation on the target image and feeds back the modification operation on the target image to the cloud 2, so that the cloud 2 generates a modified target image based on the modification operation on the target image and feeds it back to the vehicle terminal 1 for display to the user.

[0058] The target image generated by the cloud 2 of the embodiment of the present invention is returned to the real-time painting module on the vehicle computer. The module allows the user to interact with the system through a touch screen or other input devices, and provides a variety of image editing tools on the touch screen, including but not limited to brushes, erasers, palettes, shape tools, etc. The user uses the image editing tools to perform real-time modifications on the generated target image through the touch display screen, such as graffiti, adding elements, adjusting colors, etc. The vehicle computer end 1 responds to the user's modification operation on the target image and can feedback the modification operation of the target image to the cloud 2. The real-time painting module in the cloud 2 can use an AI painting model to generate a modified target image based on the modification operation on the target image. Among them, the AI ​​painting model (Stable Diffusion XL model) for real-time painting uses an acceleration model (Latent Consistency Models, LCM) to accelerate painting, so that the painting is completed within 1 second, and the modified target image is fed back to the vehicle computer end 1 for display to the user. Just as an example, in this way, every step of the user's operation can be fed back in real time until a satisfactory work is created, which increases the fun.

[0059] The AI ​​painting module of the present invention generates and transmits back to the device according to the user's operation on the image, so that the user can get real-time feedback, which can bring great fun to the user.

[0060] The embodiment of the present invention also provides an image generation method, which is applied to Figure 1 The vehicle terminal 1 and the cloud terminal 2 are shown, wherein Figure 3 As shown, the vehicle terminal 1 is used to execute steps S101 to S103, and the cloud terminal 2 is used to execute steps S201 to S202.

[0061] Step S101, obtaining a target image description text input by a user after entering an image drawing interface, obtaining a first environmental image set collected by a vehicle-mounted image acquisition device, and extracting feature description words of each environmental image in the first environmental image set.

[0062] Step S102, matching the target image description text with the characteristic description words of each environment image, and providing the second environment image set with successful matching to the user, so that the user can select an environment reference image from the second environment image set.

[0063] Step S103: Send the environment reference image and target image description text selected by the user to the cloud.

[0064] Step S201, generating a target image based on an environment reference image and a target image description text sent by the vehicle computer.

[0065] Step S202: Feedback the target image to the vehicle computer to display it to the user.

[0066] Specifically, Figure 4 As shown, after the user enters the AI ​​image painting interface on the vehicle computer, he enters the target image description text, the vehicle computer obtains the first environmental image set collected by the vehicle-mounted image acquisition device, and extracts the feature description words of each environmental image in the first environmental image set, the vehicle computer matches the feature description words of each environmental image with the target image description text entered by the user, and can provide the user with a successfully matched second environmental image set, so that the user can select an environmental reference image from the second environmental image set. If the user wants the characters in the generated target image to have their own facial features, the user can choose to upload the facial image through the local vehicle computer or mobile terminal, the vehicle computer can send the environmental reference image, the target image description text and the facial image to the cloud, the cloud can perform image preprocessing on the environmental reference image and the facial image, so as to improve the target image generated by the AI ​​painting model. Quality, and text preprocessing can also be performed on the target image description text, including extracting target style description words from the target image description text, and determining the target image style based on the target style description words, generating target images of different image art forms based on the environmental reference image, the facial image, the target image description text, the target image style and a variety of image art forms under the target image style, and returning the target image to the vehicle end, and the user can modify the target image in real time on the vehicle display screen, and the cloud can draw the target image in real time based on the user's modification operation, and return the modified target image to the display screen of the vehicle end until a satisfactory work is created, which increases the fun, and the image generation and modification process of the present invention is simple to operate, and is suitable for users of different ages. Please refer to the above embodiment for detailed description, which will not be repeated here.

[0067] The embodiment of the present invention further provides a vehicle, such as Figure 5 As shown, the vehicle includes a vehicle computer end, which includes a memory and a processor. The memory and the processor are communicatively connected to each other. Computer instructions are stored in the memory. The processor executes the image generation method of the above embodiment by executing the computer instructions.

[0068] See also Figure 6 , Figure 6 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 6As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 6 A processor 10 is taken as an example.

[0069] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0070] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0071] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0072] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0073] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 6 The example of connecting through bus is taken in the following.

[0074] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0075] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0076] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. An image generation method, characterized in that: Applied to the vehicle terminal, the method includes: Get the target image description text entered by the user after entering the image drawing interface; Acquire a first environmental image set acquired by a vehicle-mounted image acquisition device, and extract feature description words of each environmental image in the first environmental image set; matching the target image description text with the characteristic description words of each environment image, and providing a second environment image set with successful matching to a user, so that the user can select an environment reference image from the second environment image set; The environment reference image selected by the user and the target image description text are sent to the cloud, so that the cloud generates the target image based on the environment reference image and the target image description text, and feeds it back to the vehicle terminal for display to the user.

2. The method according to claim 1, characterized in that The method further comprises: In response to the user selecting the instruction to generate facial features, a picture upload interface is displayed to the user, wherein the picture upload interface includes at least a local image upload option, a mobile end upload option, and a shooting option, so that the user can select a corresponding image upload method to upload the image; Receive a facial image uploaded by a user based on the local image upload option or the shooting option, and send an environmental reference image, a target image description text, and the facial image to the cloud, so that the cloud generates a target image based on the environmental reference image, the target image description text, and the facial image.

3. The method according to claim 2, characterized in that The method further comprises: In response to the user selecting the mobile terminal upload option, a link identifier that can be connected to the mobile terminal is displayed to the user, so that the user can synchronize the facial image stored or photographed in the mobile terminal to the vehicle terminal based on the link identifier; Receive the facial image uploaded by the mobile terminal, and execute the step of sending the environment reference image, the target image description text and the facial image to the cloud.

4. The method according to claim 1, characterized in that: The method further comprises: In response to the user's modification operation on the target image, the modification operation on the target image is fed back to the cloud, so that the cloud generates a modified target image based on the modification operation on the target image, and feeds it back to the vehicle end for display to the user.

5. An image generation method, characterized in that: Applied to a cloud, the cloud is integrated with an AI painting model, and the method includes: Receiving an environment reference image and a target image description text selected by a user sent by the vehicle computer, and generating a target image based on the environment reference image and the target image description text, wherein the environment reference image is an environment reference image selected by the user from the second environment image set; The second environment image set is a vehicle-mounted computer that extracts feature description words of each environment image in the first environment image set, matches the feature description words of each environment image with the target image description text, and provides the matched second environment image set to the user; The first environment image set is a first environment image set acquired by the vehicle-mounted image acquisition device acquired by the vehicle-mounted end; the target image description text is a target image description text acquired by the vehicle-mounted end and input by the user after entering the image drawing interface; The generated target image is sent to the vehicle computer so that the vehicle computer displays the target image to the user.

6. The method according to claim 5, characterized in that The cloud is integrated with an image style set, wherein the image style set includes multiple image styles and multiple image art forms under each image style. The method further includes: Extracting target style description words from the target image description text, and determining the target image style based on the target style description words; Based on the environment reference image, the target image description text, the target image style and multiple image art forms under the target image style, target images of different image art forms are generated.

7. The method according to claim 6, characterized in that The method further comprises: If the target style description word is not extracted from the target image description text, the target image style is randomly selected from the image style set, and target images of different image art forms are generated based on the environment reference image, the target image description text, the target image style and multiple image art forms under the target image style.

8. The method according to claim 5, characterized in that Before generating the target image based on the environment reference image and the target image description text, the method further includes: Based on a preset size requirement, resizing the environment reference image; and / or, performing high-definition restoration processing on the environmental reference image; And / or, performing image quality enhancement processing on the environment reference image.

9. The method according to claim 5, characterized in that The method further comprises: Get the calendar data for the day; Based on the calendar data, determining whether the current day is a holiday; If the day is a holiday, a target image related to the holiday is generated based on the holiday, the environment reference image and the target image description text.

10. An image generation system, characterized in that: The system includes a vehicle terminal and a cloud terminal, wherein: The vehicle computer obtains the target image description text input by the user after entering the image drawing interface, obtains the first environmental image set collected by the vehicle-mounted image acquisition device, and extracts the characteristic description words of each environmental image in the first environmental image set; The vehicle computer matches the target image description text with the characteristic description words of each environmental image, and provides a second environmental image set with successful matching to the user, so that the user can select an environmental reference image from the second environmental image set; The vehicle computer sends the environment reference image selected by the user and the target image description text to the cloud; The cloud receives the environment reference image and target image description text selected by the user sent by the vehicle terminal, generates the target image based on the environment reference image and the target image description text, and feeds it back to the vehicle terminal for display to the user.

11. A vehicle, characterized in that: The vehicle includes a vehicle computer end, which includes a memory and a processor. The memory and the processor are communicatively connected to each other. Computer instructions are stored in the memory. The processor executes the image generation method described in any one of claims 1 to 4 by executing the computer instructions.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the image generation method according to any one of claims 1 to 4, or to execute the image generation method according to any one of claims 5 to 9.

Images