Display method and device, equipment and storage medium

Abstract

The invention provides a display method and device, equipment and a medium. The method comprises the following steps: monitoring multi-dimensional scene information of a vehicle; the vehicle multi-dimensional scene information comprises vehicle state information, navigation information, user operation information and / or user driving habits; predicting a target service scene required by the user according to the vehicle multi-dimensional scene information; determining a target function module from the three-dimensional vehicle-mounted model according to the target service scene; and loading the target function module to a memory, and rendering and displaying a model interface corresponding to the loaded target function module. According to the method, only the target function module corresponding to the target service scene required by the user is loaded, the three-dimensional vehicle-mounted model does not need to be loaded completely, on-demand loading of the three-dimensional vehicle-mounted model resources is achieved, delay during application starting or function switching is improved, and the user experience is improved by actively predicting the service requirement of the user. And the visual angle of the three-dimensional vehicle-mounted model matched with the service demand is displayed, so that the intelligence of vehicle-machine interaction is improved.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to a display method, apparatus, device, and computer-readable storage medium. Background Technology

[0002] Currently, in-vehicle systems introduce three-dimensional (3D) vehicle models, or 3D vehicle models for short, allowing users to intuitively see the vehicle's status and feel the control feedback of the in-vehicle system on the vehicle.

[0003] Typically, after receiving a user's instruction (e.g., to launch a 3D vehicle application or view the interior), the in-vehicle system loads a high-precision 3D vehicle model into memory. Only after the 3D vehicle model is fully loaded does the in-vehicle system render and display the corresponding viewpoint of the 3D vehicle according to the user's initial instruction (e.g., displaying the exterior viewpoint of the 3D vehicle model). However, this method is time-consuming, resulting in significant delays when launching applications or switching functions. Summary of the Invention

[0004] In order to overcome the technical problems existing in the related technologies, embodiments of this application provide a display method, apparatus, device and computer-readable storage medium.

[0005] In a first aspect, embodiments of this application provide a display method, including:

[0006] Monitor vehicle multi-dimensional scene information; wherein, the vehicle multi-dimensional scene information includes vehicle status information, navigation information, user operation information and / or user driving habits;

[0007] Predict the target service scenario required by the user based on the multi-dimensional scenario information of the vehicle;

[0008] Based on the target service scenario, target functional modules are determined from the 3D vehicle model; wherein, the 3D vehicle model includes multiple independent functional modules divided according to vehicle functional components;

[0009] Load the target functional module into memory, render and display the model interface corresponding to the loaded target functional module.

[0010] Secondly, embodiments of this application provide a display device, comprising:

[0011] The monitoring module is used to monitor multi-dimensional scene information of the vehicle; wherein, the multi-dimensional scene information of the vehicle includes vehicle status information, navigation information, user operation information and / or user driving habits;

[0012] The processing module is used to predict the target service scenario required by the user based on the multi-dimensional scene information of the vehicle; determine the target functional module from the three-dimensional vehicle model based on the target service scenario; load the target functional module into memory; render and display the model interface corresponding to the loaded target functional module; wherein, the three-dimensional vehicle model includes multiple independent functional modules divided according to the vehicle's functional components.

[0013] Thirdly, embodiments of this application provide an electronic device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the display method provided in the first aspect of embodiments of this application.

[0014] Fourthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the display method provided in the first aspect of embodiments of this application.

[0015] Fifthly, embodiments of this application provide a computer program product having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the display method provided in the first aspect of embodiments of this application.

[0016] The technical solution provided in this application monitors multi-dimensional scene information of the vehicle and actively predicts the target service scene required by the user based on the multi-dimensional scene information of the vehicle. Only the target functional module corresponding to the target service scene is loaded, without loading all three-dimensional vehicle models. This realizes "on-demand loading" of three-dimensional vehicle model resources, reduces memory occupation, improves the latency when starting up or switching functions of 3D visualization applications, and enhances the intelligence of vehicle-machine interaction by actively predicting the user's service needs and displaying the perspective of the three-dimensional vehicle model that matches the service needs. Attached Figure Description

[0017] Figure 1 A schematic flowchart of a display method provided in an embodiment of this application;

[0018] Figure 2 A flowchart illustrating the prediction model training process provided in this application embodiment;

[0019] Figure 3 A schematic diagram of the structure of a display device provided in an embodiment of this application;

[0020] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be further described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for explaining this application and are not intended to limit this application. Those skilled in the art can make adjustments as needed to suit specific application scenarios. Furthermore, it should be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not the entire structure.

[0022] Figure 1 This is a schematic flowchart illustrating a display method provided in an embodiment of this application. Figure 1 As shown, the method may include:

[0023] S101, monitor multi-dimensional scene information of vehicles.

[0024] Among them, the multi-dimensional vehicle scenario information includes vehicle status information, navigation information, user operation information and / or user driving habits.

[0025] Vehicle status information refers to signals indicating the vehicle's own operation and component status, which may include, but are not limited to, vehicle speed, acceleration, gear position signal, door open / close status, key position, charging port cover lock status, battery level, driving range, charging gun connection status, sunroof status, etc. Specifically, vehicle status information can be obtained by continuously monitoring and parsing data frames from the vehicle network (such as the CAN bus) through the vehicle's built-in bus communication interface (such as the CAN controller) and driver program.

[0026] Navigation information refers to information related to trip planning that comes from the vehicle's built-in navigation system or mobile devices connected to the vehicle. It may include the type of destination (such as "charging station", "shopping mall", "residential area", "company", "highway service area", etc.), the current driving route, the remaining driving distance and time, the estimated arrival time, and road congestion conditions.

[0027] User operation information can refer to the information corresponding to the real-time operation commands issued by the user through the in-vehicle and out-of-vehicle interaction interfaces, including but not limited to: user operation events on the vehicle's touch screen (such as clicking the "View Exterior" control on the 3D vehicle model interface), user voice command events, and remote control events issued to the vehicle through mobile device applications.

[0028] User driving habits can refer to statistical information that reflects a user's personalized preferences and behavioral patterns, obtained by analyzing historical driving data. This includes, but is not limited to: the user's common settings preferences (such as default driving mode), behavioral patterns (such as habitually checking the exterior when approaching the vehicle, or habitually checking tire information before long-distance driving), and frequently occurring usage scenarios (such as charging at a fixed time each week).

[0029] S102. Predict the target service scenario required by the user based on multi-dimensional vehicle scenario information.

[0030] The system matches multi-dimensional vehicle information with a pre-configured scenario rule base to determine the user's desired service scenario. This pre-configured scenario rule base includes the correspondence between multi-dimensional vehicle information and service scenarios. For example, if the vehicle's speed is greater than 0 and it's in drive (D), it can be predicted that the user is in a driving state, meaning the desired service scenario is a driving scenario. Similarly, if the vehicle's battery level is below a preset threshold or the navigation destination is a charging station, the predicted desired service scenario is a charging scenario.

[0031] S103. Based on the target service scenario, determine the target functional modules from the 3D vehicle model.

[0032] The 3D vehicle model comprises multiple independent functional modules, divided according to the vehicle's functional components. In other words, the 3D vehicle model is composed of a combination of these independent functional modules. During the model preparation stage, based on the actual functions of the vehicle (e.g., using doors, seats, charging ports, and tires as independent units), a dedicated, independently executable 3D model data file is created for each functional component, thus forming a functional module. There are no mandatory real-time data dependencies between the functional modules, allowing them to be individually loaded or released by the resource manager during vehicle operation. The logical collection of all functional modules constitutes the complete 3D vehicle model.

[0033] After identifying the target service scenario, one or more target functional modules associated with the target service scenario are determined by querying a pre-defined scenario module mapping table or parsing the scenario tag metadata carried by the functional modules. The scenario module mapping table includes the correspondence between scenarios and modules. For example, when the target service scenario is a "charging scenario," querying the pre-defined scenario module mapping table can determine that the target functional module associated with the target service scenario is the "charging port module."

[0034] S104. Load the target functional module into memory, render and display the model interface corresponding to the loaded target functional module.

[0035] By calling the operating system kernel's file system interface, the module file corresponding to the target functional module is retrieved from the storage device and written into memory. The loaded target functional module is then rendered, and the rendered image is displayed. Optionally, before loading the target functional module, the already loaded functional modules in memory can be retrieved. If the target functional module is not among the already loaded functional modules, it is loaded into memory. This method avoids repeatedly loading the same functional module, reducing the consumption of system resources.

[0036] For example, if the system detects that the battery is low or the navigation destination is a charging station, the system predicts that the user's target service scenario is a charging scenario. Then, it determines the target functional modules corresponding to the charging scenario from the 3D vehicle model as the charging port module and the charging animation module. The system then loads the charging port module and the charging animation module into memory, and plays the charging animation when it receives the user's charging command.

[0037] For example, if the system detects that a user is approaching the vehicle via a Bluetooth key or ultra-wideband sensor, and predicts that the user needs to view the vehicle's exterior, then it determines from the 3D vehicle model that the target functional modules matching the target service scenario are the appearance module and the lighting animation module. The system then loads the appearance module and the lighting animation module into memory, and displays the vehicle's exterior and plays welcome lighting effects when it detects that the user has arrived at the vehicle.

[0038] In this embodiment, by monitoring multi-dimensional scene information of the vehicle and actively predicting the target service scene required by the user based on the multi-dimensional scene information of the vehicle, only the target functional module corresponding to the target service scene is loaded, without loading all 3D vehicle models. This realizes "on-demand loading" of 3D vehicle model resources, reduces memory usage, improves the latency when starting up or switching functions of 3D visualization applications, and enhances the intelligence of vehicle-machine interaction by actively predicting the user's service needs and displaying the perspective of the 3D vehicle model that matches the service needs.

[0039] Optionally, the method further includes controlling the working state of the corresponding vehicle component in response to operations on the 3D model interface.

[0040] For example, in response to a user's action of turning on the air conditioner in the corresponding model interface, the air conditioner can be turned on through the corresponding application. Similarly, in response to a user's action of raising the temperature in the corresponding model interface, the air conditioner can be raised through the corresponding application.

[0041] In this way, through steps such as user interaction, application logic processing, vehicle hardware control, and feedback and real-time updates, the conversion and execution of user intentions into actual operations of vehicle components are realized.

[0042] Optionally, the method further includes: obtaining loaded functional modules in memory; determining the functional module to be uninstalled based on the target functional module and the loaded functional modules; and uninstalling the functional module to be uninstalled from memory.

[0043] In one example, the remaining functional modules in the loaded functional modules, excluding the target functional module, can be identified as functional modules to be uninstalled. A memory release function (such as `free()`) is called to release the memory space occupied by these modules, and they are then removed from the loaded module list, thus completing the uninstallation. In this way, the system can promptly reclaim idle resources and reduce memory usage.

[0044] As an optional implementation, the process of determining the unloadable functional module based on the target functional module and the loaded functional modules may include: determining the unloadable functional module based on the difference between the target functional module and the loaded functional modules; determining the unloading priority of each unloadable functional module based on the user's driving habits and the memory space occupied by each unloadable functional module; and determining the unloadable functional module from the unloadable functional modules according to the unloading priority.

[0045] Specifically, the difference between the target functional module and the loaded functional modules can be calculated to identify the uninstallable functional modules. Then, considering user driving habits and the memory space occupied by each uninstallable functional module, the uninstallation priority of each module can be calculated. For example, based on user driving habits, the probability of each uninstallable functional module being requested to be loaded again in the future can be predicted. This probability, combined with the memory space occupied by each module, can determine its uninstallation priority. For instance, a lower probability of being requested and a larger memory space occupied result in a higher uninstallation priority for that module; conversely, a higher probability of being requested and a smaller memory space occupied result in a lower uninstallation priority. After obtaining the uninstallation priorities of each uninstallable functional module, one or more modules with the highest uninstallation priority can be identified as modules to be uninstalled. The specific number of modules to be uninstalled can be dynamically adjusted based on the real-time pressure of the system's available memory. This method allows functional modules that may be requested and loaded again in a short period of time to remain in memory, while functional modules that are not requested or are unlikely to be requested in a short period of time and occupy a large amount of memory space are unloaded. This reduces the number of times functional modules are repeatedly loaded, and improves the responsiveness of the model interface while ensuring efficient memory utilization.

[0046] It should be noted that when rendering each frame, the rendering engine only processes and outputs the model interface corresponding to the target functional module. For other modules that are loaded into memory but do not belong to the target functional module, the rendering engine will ensure that they do not appear in the final rendered screen by ignoring their drawing calls or setting them to an invisible state.

[0047] In one embodiment, optionally, the above S102 may include: predicting the target service scenario currently required by the user based on the vehicle's multi-dimensional scene information and a preset prediction model; wherein the prediction model is trained in the cloud based on teacher model distillation and deployed to the vehicle.

[0048] The teacher model is a large, pre-trained model. In this embodiment, the prediction model can be trained in the cloud. That is, the cloud can train the student model through distillation of the teacher model, and then deploy the distilled student model (i.e., the prediction model) on the vehicle. The lightweight nature of the prediction model is more suitable for the resource-constrained vehicle, enabling the prediction of user intent / service scenarios to be realized on the vehicle, improving prediction efficiency and meeting the needs of real-time applications.

[0049] Optionally, such as Figure 2 As shown, the prediction model can be trained in the cloud through the following process:

[0050] S201. Obtain training data.

[0051] The training data includes multi-dimensional scene information of sample vehicles and sample prediction results corresponding to the multi-dimensional scene information of sample vehicles (the sample prediction results here are the sample service scenarios required by the sample users determined by combining the multi-dimensional scene information of sample vehicles, which are real label data); the multi-dimensional scene information of sample vehicles includes sample vehicle status information, sample navigation information, sample user operation information and / or sample user driving habits.

[0052] After obtaining multi-dimensional scene information and sample prediction results of the sample vehicles, the sample prediction results are used as real label data for supervised training of the student model. The teacher model predicts the service scenarios required by the sample users associated with the sample vehicles based on the multi-dimensional scene information of the sample vehicles.

[0053] S202. Input the multi-dimensional scene information of the sample vehicles into the student model and the teacher model.

[0054] In deep learning, the student model is a model that has not yet been trained or optimized and is ready to learn knowledge from the teacher model. The student model is typically a simplified version designed to mimic the behavior of a more complex and accurate teacher model.

[0055] The teacher model has a larger network size than the initial student model. The teacher model is a neural network model that has been trained on a large amount of data and has achieved a certain level of performance, enabling it to accurately predict the service scenarios required by users. In applications such as knowledge distillation, the teacher model transfers the knowledge it has learned to the initial student model.

[0056] S203. Obtain the first intermediate layer features and the first prediction result of the teacher model.

[0057] The multi-dimensional scene information of the sample vehicles is input into the student model and the teacher model. The input multi-dimensional scene information of the sample vehicles is used for feature extraction through the input layer and intermediate layer of each model. Based on the extracted features, the final prediction result is output through the output layer.

[0058] The first intermediate layer feature of the teacher model is the feature extracted from the intermediate layer of the network structure of the teacher model, and the first prediction result is the prediction result of the service scenario obtained by the teacher model based on the multi-dimensional scene information of the sample vehicles.

[0059] S204. Use the first prediction result as the soft label, the sample prediction result as the hard label, and the first intermediate layer features as feature-level knowledge.

[0060] S205. Distill the student model based on soft labels, hard labels, and feature-level knowledge to obtain a prediction model.

[0061] The process of distilling the student model using hard labels is essentially a supervised training process based on sample prediction results. Soft labels represent the first prediction output of the already trained teacher model. Using these soft labels for supervised training of the student model, this knowledge distillation process guides the student model to learn from the teacher model's knowledge, thereby improving its prediction accuracy. Simultaneously, the first intermediate layer features output from the teacher model's intermediate layers are used to guide the student model's training, allowing it to learn richer feature representations, thus improving the overall distillation performance of the student model.

[0062] Optionally, the method further includes: obtaining the second intermediate layer features and the second prediction result of the student model. Correspondingly, optionally, S205 above may include: constructing a first loss function based on the soft label and the second prediction result; constructing a second loss function based on the hard label and the second prediction result; constructing a third loss function based on the first intermediate layer features and the second intermediate layer features; and performing distillation training on the student model based on the first loss function, the second loss function, and the third loss function to obtain a prediction model.

[0063] After inputting the multi-dimensional scene information of the sample vehicles into the student model and the teacher model, we can also obtain the features extracted by the intermediate layer of the student model, i.e., the second intermediate layer features, and the second prediction result output by the output layer of the student model. Based on the soft label output by the teacher model and the second prediction result of the student model, we construct a first loss function. The first loss function can be the KL divergence, expressed as follows:

[0064] ;

[0065] in, Denotes KL divergence, This represents the soft labels output by the teacher model. This represents the second prediction result output by the student model. This represents the multi-dimensional scene information of the input sample vehicle.

[0066] Hard labels refer to the sample prediction results corresponding to the multi-dimensional scene information of the sample vehicle, i.e., the true label data. The second loss function can be the cross-entropy loss function, as shown below:

[0067] ;

[0068] in, Represents the cross-entropy loss function. It is the sample size. It refers to the number of service scenario categories. Indicates the first The sample belongs to the first Real labels for each service scenario category Represents the student model for the first The sample belongs to the first The predicted probability of each service scenario category (i.e., the second prediction result).

[0069] The first intermediate layer features output by the teacher model are used as knowledge to supervise the learning of the student model's intermediate layers, allowing the student model to more closely mimic the teacher model's feature extraction methods. The third loss function can be the mean squared error function, minimizing the feature differences between the student model's intermediate layer features and the teacher model's features. The third loss function can be expressed as follows:

[0070] ;

[0071] in, This represents the third loss function. This represents the first intermediate layer features of the teacher model. This represents the second intermediate layer features of the student model.

[0072] For example, a total loss function can be constructed based on the first, second, and third loss functions. The student model is then trained using distillation based on this total loss function. If the loss value of the total loss function is greater than a preset convergence threshold, the model parameters of the student model are adjusted, and the student model continues to be trained based on the above distillation process until the loss value of the total loss function is less than or equal to the preset convergence threshold, thus obtaining a well-trained prediction model. The constructed total loss function serves as the joint optimization objective to supervise the learning of the student model. The total loss function can be expressed as follows:

[0073] ;

[0074] in, Represents the total loss function. This represents the loss between the soft label and the second prediction output by the student model (i.e., the first loss function). This represents the loss between the hard labels and the second prediction output by the student model (i.e., the second loss function). This represents the loss between the first intermediate layer features output by the teacher model and the second intermediate layer features output by the student model (i.e., the third loss function). and These are hyperparameters used to control... and exist The proportion of hyperparameters in the equation and their specific values ​​can be set based on actual needs.

[0075] In this embodiment, hard labels, soft labels, and feature-level knowledge are combined to supervise the training of the student model through distillation. That is, in addition to supervising the student model with hard labels in the training data, the prediction results of the teacher model on multi-dimensional scene information of sample vehicles are also used to supervise the training of the student model. At the same time, the feature-level knowledge output by the intermediate layer of the teacher model is also used to supervise the training of the student model. This not only makes the output distribution of the student model close to that of the teacher model, but also allows the student model to imitate the feature extraction method of the teacher model, thereby improving the prediction ability of the trained prediction model for service scenarios and improving the prediction accuracy of the service scenarios.

[0076] Figure 3 This is a schematic diagram of a display device provided in an embodiment of this application. Figure 3 As shown, the device may include a monitoring module 301 and a processing module 302.

[0077] Specifically, the monitoring module 301 is used to monitor multi-dimensional scene information of the vehicle; wherein, the multi-dimensional scene information of the vehicle includes vehicle status information, navigation information, user operation information and / or user driving habits;

[0078] The processing module 302 is used to predict the target service scenario required by the user based on the multi-dimensional scene information of the vehicle; determine the target functional module from the 3D vehicle model based on the target service scenario; load the target functional module into memory, render and display the model interface corresponding to the loaded target functional module; wherein, the 3D vehicle model includes multiple independent functional modules divided according to the vehicle functional components.

[0079] Optionally, based on the above embodiments, it also includes an acquisition module.

[0080] Specifically, the acquisition module is used to acquire the loaded functional modules in memory;

[0081] The processing module 302 is also used to determine the functional module to be uninstalled based on the target functional module and the loaded functional modules; and to uninstall the functional module to be uninstalled from memory.

[0082] Based on the above embodiments, optionally, the processing module 302 is further configured to determine the uninstallable functional modules based on the difference between the target functional module and the loaded functional modules; determine the uninstallation priority of each uninstallable functional module based on the user's driving habits and the memory space occupied by each uninstallable functional module; and determine the functional modules to be uninstalled from each uninstallable functional module according to the uninstallation priority.

[0083] Optionally, based on the above embodiments, the processing module 302 is further configured to predict the target service scenario currently required by the user based on the vehicle's multi-dimensional scene information and a preset prediction model; wherein the prediction model is obtained by training on a teacher model distillation in the cloud and deployed to the vehicle.

[0084] Based on the above embodiments, optionally, the training process of the prediction model includes: acquiring training data; inputting multi-dimensional scene information of sample vehicles into the student model and the teacher model; acquiring the first intermediate layer features and the first prediction result of the teacher model; using the first prediction result as a soft label, the sample prediction result as a hard label, and the first intermediate layer features as feature-level knowledge; performing distillation training on the student model based on the soft label, hard label, and feature-level knowledge to obtain the prediction model; wherein, the training data includes multi-dimensional scene information of sample vehicles and sample prediction results corresponding to the multi-dimensional scene information of sample vehicles; the multi-dimensional scene information of sample vehicles includes sample vehicle status information, sample navigation information, sample user operation information, and / or sample user driving habits; the teacher model predicts the service scenario required by the sample user based on the multi-dimensional scene information of sample vehicles.

[0085] Based on the above embodiments, optionally, the training process of the above prediction model further includes: obtaining the second intermediate layer features and the second prediction result of the student model; constructing a first loss function based on the soft label and the second prediction result; constructing a second loss function based on the hard label and the second prediction result; constructing a third loss function based on the first intermediate layer features and the second intermediate layer features; and performing distillation training on the student model based on the first loss function, the second loss function, and the third loss function to obtain the prediction model.

[0086] Based on the above embodiments, optionally, the processing module 302 is also used to control the working state of the corresponding vehicle component in response to the operation on the three-dimensional model interface.

[0087] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of this application, such as... Figure 4As shown, the device includes a processor 40, a memory 41, an input device 42, and an output device 43; the number of processors 40 in the device can be one or more. Figure 4 Taking a processor 40 as an example; the processor 40, memory 41, input device 42, and output device 43 in this device can be connected via a bus or other means. Figure 4 Taking the example of a connection between China and Israel via a bus.

[0088] The memory 41, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as program instructions / modules corresponding to the display method in the embodiments of this application (e.g., monitoring module 301 and processing module 302 in the display device). The processor 40 executes various functional applications and data processing of the device by running the software programs, instructions, and modules stored in the memory 41, thereby realizing the above-described display method.

[0089] The memory 41 may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a given function; the data storage area may store data created during the display process. Furthermore, the memory 41 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory, or other non-volatile solid-state storage device. In some instances, the memory 41 may further include memory remotely located relative to the processor 40, which can be connected to a device / terminal / server via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0090] Input device 42 can be used to receive input digital or character information, and to generate key signal inputs related to user settings and function control of the electronic device. Output device 43 may include display devices such as a display screen.

[0091] In one embodiment, a computer-readable storage medium is also provided, on which a computer program is stored, the computer program performing the following steps when executed by a processor:

[0092] Monitor vehicle multi-dimensional scene information; wherein, the vehicle multi-dimensional scene information includes vehicle status information, navigation information, user operation information and / or user driving habits;

[0093] Predict the target service scenario required by the user based on the multi-dimensional scenario information of the vehicle;

[0094] Based on the target service scenario, target functional modules are determined from the 3D vehicle model; wherein, the 3D vehicle model includes multiple independent functional modules divided according to vehicle functional components;

[0095] Load the target functional module into memory, render and display the model interface corresponding to the loaded target functional module.

[0096] In one embodiment, a computer program product is also provided, on which a computer program is stored, which, when executed by a processor, performs the following steps:

[0097] Monitor vehicle multi-dimensional scene information; wherein, the vehicle multi-dimensional scene information includes vehicle status information, navigation information, user operation information and / or user driving habits;

[0098] Predict the target service scenario required by the user based on the multi-dimensional scenario information of the vehicle;

[0099] Based on the target service scenario, target functional modules are determined from the 3D vehicle model; wherein, the 3D vehicle model includes multiple independent functional modules divided according to vehicle functional components;

[0100] Load the target functional module into memory, render and display the model interface corresponding to the loaded target functional module.

[0101] The display device, electronic device, computer-readable storage medium, and computer program product provided in the above embodiments can execute the display method provided in any embodiment of this application, and have the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in the above embodiments can be found in the display method provided in any embodiment of this application.

[0102] Based on the above description of the implementation methods, those skilled in the art can clearly understand that this application can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0103] It is worth noting that the units and modules included in the above embodiments are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of this application.

[0104] Note that the above description is merely a preferred embodiment and the technical principles employed in this application. Those skilled in the art will understand that this application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments. Many other equivalent embodiments may be included without departing from the concept of this application, and the scope of this application is determined by the scope of the appended claims.

Claims

1. A display method, characterized in that, include: Monitor vehicle multi-dimensional scene information; wherein, the vehicle multi-dimensional scene information includes vehicle status information, navigation information, user operation information and / or user driving habits; Predict the target service scenario required by the user based on the multi-dimensional scenario information of the vehicle; Based on the target service scenario, target functional modules are determined from the 3D vehicle model; wherein, the 3D vehicle model includes multiple independent functional modules divided according to vehicle functional components; Load the target functional module into memory, render and display the model interface corresponding to the loaded target functional module.

2. The method according to claim 1, characterized in that, Also includes: Retrieve the loaded functional modules in the memory; Based on the target functional module and the loaded functional modules, determine the functional modules to be uninstalled; Unload the functional module to be uninstalled from the memory.

3. The method according to claim 2, characterized in that, Based on the target functional module and the loaded functional modules, determine the functional modules to be uninstalled, including: Based on the difference between the target functional module and the loaded functional module, the unloadable functional module is determined; Based on user driving habits and the amount of memory space occupied by each uninstallable functional module, the uninstallation priority of each uninstallable functional module is determined. Based on the uninstallation priority, the functional modules to be uninstalled are determined from the uninstallable functional modules.

4. The method according to claim 1, characterized in that, The step of predicting the user's current target service scenario based on the vehicle's multi-dimensional scene information includes: Based on the vehicle's multi-dimensional scene information and a preset prediction model, the target service scene currently required by the user is predicted; wherein, the prediction model is trained in the cloud based on teacher model distillation and deployed to the vehicle.

5. The method according to claim 4, characterized in that, The training process of the prediction model includes: Acquire training data; wherein, the training data includes multi-dimensional scene information of sample vehicles and sample prediction results corresponding to the multi-dimensional scene information of sample vehicles; the multi-dimensional scene information of sample vehicles includes sample vehicle status information, sample navigation information, sample user operation information and / or sample user driving habits; the teacher model predicts the service scenario required by the sample user based on the multi-dimensional scene information of sample vehicles. The multi-dimensional scene information of the sample vehicles is input into the student model and the teacher model; Obtain the first intermediate layer features and the first prediction result of the teacher model; The first prediction result is used as a soft label, the sample prediction result is used as a hard label, and the first intermediate layer feature is used as feature-level knowledge. The student model is trained by distillation based on the soft labels, the hard labels, and the feature-level knowledge to obtain the prediction model.

6. The method according to claim 5, characterized in that, Also includes: Obtain the second intermediate layer features and the second prediction result of the student model; The student model is trained by distillation based on the soft labels, the hard labels, and the feature-level knowledge to obtain the prediction model, including: A first loss function is constructed based on the soft label and the second prediction result; A second loss function is constructed based on the hard label and the second prediction result; A third loss function is constructed based on the features of the first intermediate layer and the features of the second intermediate layer; The student model is trained by distillation based on the first loss function, the second loss function, and the third loss function to obtain the prediction model.

7. The method according to claim 1, characterized in that, Also includes: In response to operations on the 3D model interface, the operating status of the corresponding vehicle components is controlled.

8. A display device, characterized in that, include: The monitoring module is used to monitor multi-dimensional scene information of the vehicle; wherein, the multi-dimensional scene information of the vehicle includes vehicle status information, navigation information, user operation information and / or user driving habits; The processing module is used to predict the target service scenario required by the user based on the multi-dimensional scene information of the vehicle; determine the target functional module from the three-dimensional vehicle model based on the target service scenario; load the target functional module into memory; render and display the model interface corresponding to the loaded target functional module; wherein, the three-dimensional vehicle model includes multiple independent functional modules divided according to the vehicle's functional components.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images