Pbr material library parameter adjustment method and device, storage medium and computer device

Abstract

The application provides a PBR material library parameter adjustment method and device, a storage medium and a computer device. The method comprises the following steps: receiving collected vehicle-mounted multi-dimensional data, wherein the vehicle-mounted multi-dimensional data comprises environmental data, driving data and hardware data; performing weighted fusion on the vehicle-mounted multi-dimensional data to generate scene feature information; adjusting PBR material library parameters according to the scene feature information, and updating the rendering effect according to the adjusted PBR material library parameters, so that the material rendering can adapt to the exclusive requirements of the vehicle-mounted scene.

Description

Technical Field

[0001] This invention relates to the field of interdisciplinary technology of automotive 3D HMI and computer graphics, and in particular to a method, apparatus, storage medium and computer device for adjusting PBR material library parameters. Background Technology

[0002] With the development of automotive 3D (Three-Dimensional) HMI (Human-Machine Interface) technology, PBR (Physically Based Rendering) material libraries have become the core for enhancing the interactive immersion. However, related technologies only focus on material library construction, resulting in material rendering that cannot adapt to the specific needs of automotive scenarios. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a method, apparatus, storage medium, and computer device for adjusting PBR material library parameters, enabling material rendering to adapt to the specific needs of automotive scenarios.

[0004] In a first aspect, embodiments of the present invention provide a method for adjusting parameters of a PBR material library, the method comprising: Receive collected vehicle-mounted multidimensional data, which includes environmental data, driving data, and hardware data; The vehicle-mounted multidimensional data is weighted and fused to generate scene feature information (also referred to as scene feature signal in this article). Adjust the PBR material library parameters based on the scene feature information, and update the rendering effect based on the adjusted PBR material library parameters.

[0005] Optionally, adjusting the PBR material library parameters based on the scene feature information includes: Based on the scene feature information, the parameters of the PBR material library are adjusted using a basic strategy; the basic strategy is the minimum guarantee strategy for ensuring frame rate stability and driving safety in vehicle scenarios.

[0006] Optionally, adjusting the PBR material library parameters using a basic strategy based on the scene feature information includes: When the GPU utilization rate exceeds the first threshold, the non-core layer, including the special effects particle layer, is shut down. When the GPU utilization rate is less than or equal to the second threshold, full-layer rendering is enabled. In close-up shots, if the GPU utilization rate is less than or equal to the second threshold, full-layer high-precision rendering is performed; if the GPU utilization rate is greater than the first threshold, the non-core layer is turned off. In the foreground, if the GPU utilization rate is greater than the first threshold, the non-core layer is turned off; if the GPU utilization rate is less than or equal to the second threshold, full-layer rendering is enabled. When the light intensity is greater than or equal to the fourth threshold, the reflectivity of the clear coat is reduced by the first percentage. When the light intensity is less than or equal to the third threshold, the brightness of the metal reflective layer is increased by a second percentage.

[0007] Optionally, adjusting the PBR material library parameters based on the scene feature information further includes: When a vehicle is detected to be in the first scene, the PBR material library parameters are adjusted using an extended strategy based on the scene feature information. The extended strategy is an optimization strategy that balances performance, image quality, safety and personalized adaptation on the basis of the basic strategy. The step of adjusting the PBR material library parameters using an extended strategy based on the scene feature information includes: The scene feature information is input into the prediction model to obtain the predicted PBR material library parameters; Read the priority parameters set by the user; The PBR material library parameters are adjusted based on the predicted PBR material library parameters and the priority parameters.

[0008] Optionally, the method further includes: In response to a user's operation, a user operation instruction is generated and the user operation instruction is parsed to obtain adjustment information, which is parameter adjustment information or user switching information; The PBR material library parameters are adjusted according to the adjustment information, and the rendering effect is updated according to the adjusted PBR material library parameters.

[0009] Optionally, the vehicle-mounted multidimensional data further includes physiological data; the method further includes: When the vehicle-mounted multidimensional data is weighted and fused, the driver's physiological state information (also referred to as physiological state signal or physiological characteristic information in this article) is also generated. The PBR material library parameters are adjusted based on the physiological state information, and the rendering effect is updated based on the adjusted PBR material library parameters.

[0010] Optionally, the method further includes: Record the adjustment and extension data of the PBR material library parameters; Upload the adjusted and expanded data to the cloud; Receive model parameter update information sent from the cloud; The parameters of the prediction model are updated based on the model parameter update information.

[0011] Optionally, adjusting the PBR material library parameters based on the physiological state information includes: When the physiological state information indicates a state of fatigue, the reflectivity of all materials is reduced by a third percentage, and the contrast of the core information area is increased by a fourth percentage. When the physiological state information indicates a state of distraction, the transparency of the core area material is enhanced by a fifth percentage, and a border highlight layer is added by a sixth percentage.

[0012] On the other hand, embodiments of the present invention provide a PBR material library parameter adjustment device, the device comprising: The transceiver module is used to receive collected vehicle-mounted multidimensional data, which includes environmental data, driving data, and hardware data. The processing module is used to perform weighted fusion of the vehicle-mounted multidimensional data to generate scene feature information; adjust the PBR material library parameters according to the scene feature information; and update the rendering effect according to the adjusted PBR material library parameters.

[0013] On the other hand, embodiments of the present invention provide a storage medium including a stored program, wherein the program controls the device where the storage medium is located to execute the above-described method when it is running.

[0014] On the other hand, embodiments of the present invention provide a computer device including a memory and a processor. The memory is used to store information including program instructions, and the processor is used to control the execution of the program instructions, wherein the program instructions, when loaded and executed by the processor, implement the steps of the above-described method.

[0015] The technical solution of the PBR material library parameter adjustment method, apparatus, storage medium, and computer device provided in the embodiments of the present invention includes: receiving collected vehicle-mounted multi-dimensional data, the vehicle-mounted multi-dimensional data including environmental data, driving data, and hardware data; weightedly fusing the vehicle-mounted multi-dimensional data to generate scene feature information; adjusting the PBR material library parameters according to the scene feature information; and updating the rendering effect according to the adjusted PBR material library parameters, so that the material rendering can adapt to the specific needs of the vehicle-mounted scene. Attached Figure Description

[0016] Figure 1 A flowchart illustrating a method for adjusting PBR material library parameters provided in an embodiment of the present invention; Figure 2 This is an architecture diagram of the PBR material library parameter adjustment system in an embodiment of the present invention; Figure 3 This is a signaling interaction diagram of the PBR material library parameter adjustment scheme in an embodiment of the present invention; Figure 4This is a schematic diagram of the layered PBR material structure and parameter baseline in an embodiment of the present invention; Figure 5 This is a schematic diagram of the profile switching process in an embodiment of the present invention; Figure 6 This is a schematic diagram comparing the optimization effects of profile switching in an embodiment of the present invention; Figure 7 This is a schematic diagram of a PBR material library parameter adjustment device provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0017] To better understand the technical solution of the present invention, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0018] It should be understood that the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0019] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0020] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0021] The relevant technologies suffer from the following core defects, resulting in a failure to meet the specific needs of automotive scenarios: 1. Lack of adaptation to in-vehicle scenes: Only focusing on the construction of general material libraries without taking into account the characteristics of in-vehicle scenes (such as highway / city road conditions, in-vehicle chip GPU (Graphics Processing Unit) / memory constraints, and user interaction restrictions during driving), resulting in material rendering being incompatible with in-vehicle hardware and large frame rate fluctuations; 2. Limited Layered Material Scheduling: The lack of dynamic linkage between layered materials and the failure to intelligently schedule layered precision based on the load of the vehicle scene (camera distance, number of models, road condition complexity) result in performance waste due to "consistent rendering precision for near and far scenes". 3. Insufficient user interaction and self-optimization: It relies on manual touch adjustment, lacks multimodal interaction (voice + quick touch) adapted to in-vehicle scenarios, and does not record high-frequency user adjustment behaviors, making it impossible to achieve self-optimization and update of material parameters; 4. Lack of linkage with in-vehicle environment: The material rendering effect is not dynamically adjusted in combination with in-vehicle environment factors (light intensity, in-vehicle temperature). For example, the material reflection is too strong under strong light, which affects driving safety, and the power consumption of material rendering is not optimized in low temperature environment.

[0022] To address the aforementioned technical issues, this invention provides a method for adjusting parameters in a PBR material library.

[0023] Figure 1 A flowchart of a PBR material library parameter adjustment method provided in an embodiment of the present invention is shown below. Figure 1 As shown, the method includes: Step 101: Receive the collected vehicle-mounted multidimensional data, which includes environmental data, driving data, and hardware data.

[0024] The PBR material library parameter adjustment method provided in this embodiment of the invention can be implemented using a PBR material library parameter adjustment system.

[0025] For example, the components of the PBR material library parameter adjustment system are shown in Table 1.

[0026]

[0027] In Table 1, "profile" usually refers to a "configuration file" or "user configuration profile," which is a set of pre-defined data used to define the network connection parameters, security policies, or personalized settings of terminal devices (such as mobile phones, computers, and IoT devices).

[0028] For example, Figure 2 This is an architecture diagram of the PBR material library parameter adjustment system in an embodiment of the present invention. Figure 3 This is a signaling interaction diagram of the PBR material library parameter adjustment scheme in an embodiment of the present invention. Figure 2 and Figure 3 As shown, the vehicle-mounted multi-dimensional scene perception module collects vehicle-mounted multi-dimensional data in parallel at a frequency of greater than or equal to 10Hz, and can complete scene perception and feature fusion in 0-600ms.

[0029] For example, in-vehicle multidimensional data may include data from the following scenarios: Environmental scenarios: light intensity (e.g., light intensity ≥ 8000 lux is strong light, light intensity ≤ 500 lux is weak light, light intensity greater than 500 lux and less than 8000 lux is normal), vehicle interior temperature (e.g., -10~45 degrees Celsius, accuracy ±1 degree Celsius); Driving scenarios include: road conditions (e.g., speed ≥ 60 km / h is highway, speed > 30 km / h and < 60 km / h is city, speed ≤ 30 km / h is congestion), camera distance (e.g., camera distance ≤ 1.5m is close-up, camera distance > 1.5m and < 3m is medium shot, camera distance ≥ 3m is long shot), and driving safety status (e.g., using a threshold judgment + weighted summation vehicle lightweight algorithm to assign weights to the three types of indicators, such as driver status weight 0.4, vehicle driving status weight 0.4, and driving behavior weight 0.2; each indicator is assigned a value according to risk level: low risk = 0 points, moderate risk = 1 point, high risk = 2 points; weighted summation to calculate the total score: total score = driver status score × 0.4 + vehicle status score × 0.4 + driving behavior score × 0.2; total score mapping: total score < 1 is low risk, 1 ≤ total score < 2 is medium risk, and total score ≥ 2 is high risk). Hardware load scenarios: GPU utilization (e.g., GPU utilization ≤ 50% is light load, GPU utilization > 50% and < 70% is medium load, GPU utilization > 70% is high load), memory utilization (e.g., memory utilization ≤ 60% is smooth, memory utilization > 60% and < 80% is normal, memory utilization > 80% is stuttering). Step 102: Weighted fusion of vehicle-mounted multidimensional data to generate scene feature information.

[0030] For example, the vehicle-mounted multi-dimensional scene perception module generates scene feature information through a weighted fusion algorithm and sends it to the layered PBR material library and dynamic scheduling module.

[0031] For example, scene feature information could be "high speed + strong light + GPU usage 60% + close-up + low risk".

[0032] Step 103: Adjust the PBR material library parameters according to the scene feature information, and update the rendering effect according to the adjusted PBR material library parameters.

[0033] For example, Figure 4 This is a schematic diagram of the layered PBR material structure and parameter baseline in an embodiment of the present invention, as shown below. Figure 4 As shown, the baseline structure and physical parameters of the basic material library storing the core materials for vehicles are as follows: Car paint: base color layer (red, green and blue brightness 80%-95%) + metallic reflective layer (reflective intensity 40%-60%) + clear coat layer (reflective intensity 30%-45%) + special effect particle layer (particle density 10-30 particles / square millimeter).

[0034] Glass: Basic transparent layer (light transmittance 85%-90%) + reflective layer (reflection coefficient 20%-35%) + anti-glare layer (glare suppression rate 50%-70%).

[0035] In some possible embodiments, the PBR material library parameters are adjusted according to scene feature information, including: adjusting the PBR material library parameters using a basic strategy based on scene feature information; the basic strategy is the minimum guarantee strategy to ensure frame rate stability and driving safety in vehicle scenarios.

[0036] In this embodiment of the invention, the PBR material library parameters are adjusted using a basic strategy based on scene feature information, including: When the GPU utilization rate exceeds the first threshold, the non-core layer, including the special effects particle layer, is turned off. When the GPU utilization is less than or equal to the second threshold, enable full-layer rendering; In close-up shots, if the GPU utilization is less than or equal to the second threshold, full-layer high-precision rendering is performed; if the GPU utilization is greater than the first threshold, non-core layers are turned off. When rendering in the foreground, if the GPU utilization rate is greater than the first threshold, the non-core layer is turned off; if the GPU utilization rate is less than or equal to the second threshold, full-layer rendering is enabled. When the light intensity is greater than or equal to the fourth threshold, the reflectivity of the clear coat is reduced by the first percentage. When the light intensity is less than or equal to the third threshold, the brightness of the metal reflective layer is increased by a second percentage.

[0037] For example, the basic strategy may include multiple sub-strategies such as hardware load strategy, lens distance strategy, and ambient light strategy.

[0038] For example, a hardware load strategy could be: disable non-core layers when GPU utilization is >70%, and enable full-layer rendering when GPU utilization is ≤50%.

[0039] The core layer is the essential layer, ensuring basic visual appeal and functionality. It includes: a base color layer (ensuring basic color), a metallic reflective layer (ensuring metallic texture), and a clear coat layer (ensuring gloss and basic reflectivity). Non-core layers are enhancement layers, which do not affect basic recognition when turned off. These include: special effect particle layers, such as pearlescent / glitter particles, which only enhance the visual effects of details.

[0040] For example, a camera distance strategy could be: full-layer high-precision rendering (200,000 triangles) for close-ups, and disabling non-core layers (≤5,000 triangles) for distant views. Here, triangles are parameters of a 3D model, referring to the smallest constituent unit of the 3D model rendering. A higher number of triangles results in richer model details, but also increases the GPU computational load. For example, an ambient light strategy could be to reduce the reflectivity of the clear coat layer by 30% in strong light and increase the brightness of the metal reflective layer by 20% in weak light.

[0041] Among them, the sub-strategies in the basic strategy are superimposed relationships where each is judged in parallel and takes effect independently, rather than mutually exclusive "or" or "and" where all are forced to take effect. The final execution combination is determined by the real-time scene conditions, thereby realizing the dynamic linkage of layered materials.

[0042] Each sub-strategy is triggered independently: the hardware load strategy, camera distance strategy, and ambient light strategy are determined separately by their respective scene conditions and do not interfere with each other.

[0043] When multiple conditions are met, the sub-strategies are executed in combination: If multiple conditions are met simultaneously (e.g., GPU > 70% + distant view + strong light), the corresponding sub-strategies are executed simultaneously (e.g., turn off non-core layers + compress triangles to ≤5000 + reduce clear coat reflection by 30%); if only some conditions are met (e.g., GPU = 50% + close view + strong light), the corresponding sub-strategy is executed only (e.g., enable full layering + 200,000 triangles + reduce reflection by 30%).

[0044] For example, if the hardware load strategy and the lens distance strategy are mutually exclusive, the hardware load strategy will be executed with the highest priority for optimization.

[0045] For example, when the GPU is greater than 70% for close-up shots, high-precision rendering with full layering should not be performed as is for close-up shots. Instead, non-core layers should be turned off when the GPU is greater than 70%.

[0046] For example, when GPU=50%+ for distant views, non-core layers should not be turned off as for distant views; instead, full layer rendering should be enabled as for GPU≤50%.

[0047] In some possible embodiments, the PBR material library parameters are adjusted according to scene feature information, including: when a vehicle is detected to be in a first scene, the PBR material library parameters are adjusted according to scene feature information using an extended strategy; the extended strategy is an optimization strategy that balances performance, image quality, security and personalized adaptation on the basis of the basic strategy.

[0048] In this embodiment of the invention, when the vehicle is in the first scenario, an extended strategy is executed on top of the basic strategy. The extended strategy does not replace the basic strategy, but only enhances or fine-tunes it.

[0049] For example, the first scenario includes any of the following conditions: 1. Users can manually enable the "Performance / Image Quality / Security Priority" configuration mode; 2. GPU utilization is between 50% and 70% (performance and image quality need to be balanced); 3. Changes in the driver's physiological state (such as increased fatigue and inattention); 4. The system detects scenarios that require AI prediction and optimization (such as continuous high-speed driving and frequent camera switching).

[0050] In some possible embodiments, the PBR material library parameters are adjusted using an extended strategy based on scene feature information, including: inputting scene feature information into the prediction model to obtain predicted PBR material library parameters; reading the priority parameters set by the user; and adjusting the PBR material library parameters based on the predicted PBR material library parameters and the priority parameters.

[0051] For example, the layered PBR material library and dynamic scheduling module can execute a dual scheduling logic of "basic strategy + extended strategy" based on scene feature information.

[0052] For example, the prediction model can be an AI (Artificial Intelligence) prediction model, such as an LSTM (Long Short-Term Memory) model or a CNN (Convolutional Neural Network) (adapted to complex road condition image features).

[0053] For example, the layered PBR material library and dynamic scheduling module can include a prediction submodule. The prediction submodule calls the AI ​​prediction model to predict the optimal parameters, that is, to predict the PBR material library parameters; the layered PBR material library and dynamic scheduling module reads the user-set priority parameters (such as performance / image quality / safety priority) / driving safety status adaptation parameters and executes them; during high-risk driving, the material complexity of non-core areas is reduced and the rendering priority of core areas is increased; the latency of outputting material rendering signals to the vehicle display terminal is ≤100ms.

[0054] The core area refers to the screen / 3D scene space where the driver's visual attention is highly focused and directly linked to driving safety or core driving information. Typical areas include: Screen interface layer: upper part of the instrument panel / central control screen, central area (such as speedometer, battery level display, navigation route, ADAS warning pop-up, emergency function access); 3D vehicle model layer: Core viewing areas for daily observation / interaction, such as the front of the vehicle body and the driver's side, as well as component areas that are highly related to the vehicle's status (such as the door and headlight status display areas).

[0055] Non-core areas refer to screen / 3D scene spaces with low driver visual attention and no direct connection to driving safety or core information. Typical areas include: Screen interface layer: screen edge, bottom decoration area, secondary function icon area (such as wallpaper, non-emergency settings entry), and pure visual decorative element area; 3D car model layer: The rear of the car body, the roof, the side detail areas away from the driver's view, and the areas where special effects particles and decorative materials are located for purely decorative purposes.

[0056] The core area refers to the rendered 3D scene / screen space. The core layer refers to the internal structure of the rendered material (such as the base color layer, metallic reflection layer, and varnish layer).

[0057] For example, the prediction submodule could be an AI prediction submodule.

[0058] like Figure 3 As shown, the layered PBR material library and dynamic scheduling module call the AI ​​prediction submodule to predict parameters, and combine user priority / physiological state information to adapt the PBR material library parameters. Then, the material rendering signal is sent to the vehicle display terminal. The time delay from receiving the material rendering signal to the display terminal completing the display adjustment is less than or equal to 100ms.

[0059] For example, user priority is determined by a combination of user-configured settings and automatic scenario adaptation.

[0060]

[0061] For example, proactive configuration refers to the user triggering proactive configuration by operating the central control screen or entering a voice command, and then manually selecting one of the three categories: "performance priority," "image quality priority," and "security priority." For example, the three priorities of "performance priority," "image quality priority," and "security priority" are shown in Table 2.

[0062] For example, automatic scenario adaptation means that in high-risk driving conditions, the system automatically and forcibly switches to safety priority, while other scenarios continue to use the priority actively configured by the user.

[0063] For details on how the layered PBR material library and dynamic scheduling module adapt PBR material library parameters in conjunction with physiological state information, please refer to the relevant descriptions in "Physiological Data" and "Physiological State Information" below.

[0064] For example, the prediction model can also be a rule tree algorithm (adapted to simple scenarios, i.e., pre-defined, rule-based finite scenarios, as opposed to real-time scenarios). The rule tree algorithm can reduce computational requirements.

[0065] Optionally, in response to user operations, this embodiment of the invention can also generate parameter adjustment instructions and parse the parameter adjustment instructions to obtain parameter adjustment information; adjust the PBR material library parameters according to the parameter adjustment information, and update the rendering effect according to the adjusted PBR material library parameters.

[0066] This invention supports multimodal interactive adjustment of PBR material library parameters. Users can initiate adjustment requests through multimodal commands, such as voice commands ("enhance metallic luster", "switch performance priority mode"), touch commands (switching styles on steering wheel shortcut buttons, manually adjusting parameters on the central control screen), and priority configuration commands.

[0067] like Figure 3 As shown, the in-vehicle multimodal interaction module parses parameter adjustment instructions and maps them to specific material layers, such as "enhance metallic luster" → locates the metal reflection layer of the car paint → reflection intensity + 25%; outputs parameter adjustment information, synchronously updates the rendering effect and records it to the behavior feedback and self-optimization module.

[0068] Optionally, in response to user operations, this embodiment of the invention can also generate user switching instructions and parse user switching instructions to obtain user switching information; adjust PBR material library parameters according to user switching information, and update rendering effects according to the adjusted PBR material library parameters.

[0069] For example, user switching information includes user profile parameters.

[0070] like Figure 3 As shown, the in-vehicle multimodal interaction module parses the user switching command and outputs the user switching information. After switching users, the corresponding profile parameters are automatically loaded, the rendering effect is updated synchronously, and the information is recorded in the behavior feedback and self-optimization module.

[0071] For example, such as Figure 3 As shown, the in-vehicle multimodal interaction module can respond to user commands in real time, with a response time of ≤0.8 seconds.

[0072] For example, multi-user identity recognition can use fingerprint recognition, facial recognition, voiceprint recognition, or in-vehicle Bluetooth (bound to mobile phone ID) to achieve automatic profile switching; it is compatible with both physical button cockpit designs and buttonless cockpit designs.

[0073] For example, Figure 5 This is a schematic diagram of the profile switching process in an embodiment of the present invention. Figure 6 This is a schematic diagram comparing the optimization effects of profile switching in an embodiment of the present invention, as shown below. Figure 5As shown, the user first initiates a switching command (such as fingerprint / voice / touch command). The user priority and multi-user management module verifies the user's identity based on the switching command. If the authentication fails, the user is prompted to re-authenticate. If the authentication succeeds, the corresponding user profile is loaded (each user profile has independent parameter configuration), and the user profile signal (including personalized material parameters) is output to the layered PBR material library and dynamic scheduling module. The layered PBR material library and dynamic scheduling module apply the personalized parameters to send a rendering update signal, and the vehicle display terminal updates the rendering effect. The optimization effect of the profile switching process in this embodiment of the invention is compared to... Figure 6 As shown, compared to before the optimization, the number of monthly adjustments for drivers has decreased significantly, and the scene matching degree has improved significantly.

[0074] Optionally, the in-vehicle multidimensional data also includes physiological data. In this embodiment of the invention, when weighted and fused with the in-vehicle multidimensional data, driver physiological state information is also generated; the PBR material library parameters are adjusted based on the physiological state information, and the rendering effect is updated based on the adjusted PBR material library parameters.

[0075] For example, physiological data may include eye data and head position data. Eye data may include blink frequency, gaze position, PERCLOS (Percentage of Eyelid Closure), etc.

[0076] For example, head data can be collected using a cockpit camera, eye data can be collected using the cockpit camera's eye-tracking function, or eye data can be obtained by processing head data using a visual attention algorithm, in order to replace the eye-tracking function.

[0077] For example, the driver's physiological state information is obtained based on the correspondence between physiological data and physiological state information.

[0078] For example, the correspondence between physiological data and physiological state information can be shown in Table 3.

[0079]

[0080] Optionally, physiological data may also include heart rate data, which can be collected using a steering wheel heart rate sensor. For example, a heart rate variability (HRV) of less than 50 ms may indicate fatigue.

[0081] like Figure 2 and Figure 3As shown, when the in-vehicle multi-dimensional scene perception module performs weighted fusion of in-vehicle multi-dimensional data, it also generates the driver's physiological state information and sends it to the in-vehicle environment-material linkage module. Based on the physiological state information, the in-vehicle environment-material linkage module sends a physiological adaptation signal to the layered PBR material library and the dynamic scheduling module, causing the layered PBR material library and the dynamic scheduling module to adjust the PBR material library parameters. Based on the adjusted PBR material library parameters, it then sends a material update signal to the in-vehicle display terminal to update the rendering effect.

[0082] For example, when the physiological state information is fatigue, the corresponding physiological adaptation signals include: reducing the reflectivity of all materials by 40% and increasing the contrast of the core information area by 20%.

[0083] For example, when the physiological state information is that attention is distracted, the corresponding physiological adaptation signals include: enhancing the material recognition of the core area (such as increasing transparency by 10% + adding a highlight layer to the border).

[0084] For example, when the physiological state information is that the person is awake, the corresponding physiological adaptation signals include: execute according to the conventional environmental adaptation strategy, that is, the system does not make any special changes and executes according to the original strategy.

[0085] In this embodiment of the invention, adjusting the PBR material library parameters based on physiological state information includes: When the physiological state information indicates a state of fatigue, the reflectivity of all materials is reduced by the third percentage, and the contrast of the core information area is increased by the fourth percentage. When the physiological state information indicates a state of distraction, the transparency of the core area material is increased by the fifth percentage, and the border highlight layer is added by the sixth percentage.

[0086] For example, the third percentage is 40%, the fourth percentage is 20%, the fifth percentage is 10%, and the sixth percentage is 30%.

[0087] For example, adjusting PBR material library parameters based on physiological state information can be responsive in real time.

[0088] Optionally, embodiments of the present invention may also record adjustment data and extended data of PBR material library parameters; upload the adjustment data and extended data to the cloud; receive model parameter update information sent from the cloud; and update the parameters of the prediction model according to the model parameter update information.

[0089] In this embodiment of the invention, basic data (such as high-frequency adjustment parameters and scene-material preferences) and extended data (priority configuration records, multi-user switching frequency, and high-risk scene adjustment records) for adjusting PBR material library parameters are recorded. Optimization is performed through machine learning, such as setting high-frequency user adjustments as default values, generating optimization baselines independently for multi-user profiles and supporting cloud synchronization, updating prediction model parameters monthly, and improving prediction accuracy to over 95%. Optimized parameters are fed back to the hierarchical PBR material library and dynamic scheduling module, forming a closed loop that can iterate over a long period of 7 days or more.

[0090] This invention relates to the interdisciplinary field of automotive 3D HMI and computer graphics, specifically to a method for dynamic adaptation and self-optimization of a layered PBR (Physically Based Rendering) material library based on in-vehicle scene perception, user behavior feedback, and driver physiological state adaptation. Through a collaborative process encompassing "deep scene perception - layered intelligent scheduling - multimodal interaction adjustment - behavior feedback self-optimization - physiological state linkage," it achieves a deep integration of material rendering realism, performance stability, personalized experience, and driving safety in an in-vehicle environment, suitable for 3D interactive interfaces of L2-L4 level intelligent driving vehicles.

[0091] This invention addresses the gaps not covered by related technologies by proposing an integrated solution of "vehicle scene perception - hierarchical dynamic scheduling - multimodal interaction - behavior self-optimization", filling the technical gap in vehicle-specific PBR material libraries.

[0092] This invention aims to solve the technical problems of "insufficient adaptation to vehicle scenes, single hierarchical scheduling, limited interaction methods, lack of self-optimization mechanism, and insufficient extended adaptation" in related technologies. By integrating four extended functions—dynamic scheduling of prediction models, user priority configuration, linkage of driver physiological state, and self-optimization of multi-user profiles—the software layer takes "scene perception → hierarchical scheduling → interaction adjustment → self-optimization → physiological adaptation" as the core sequence to achieve full-process collaboration and ensure the deep integration of multiple objectives of material rendering in the vehicle environment.

[0093] This invention employs in-vehicle multi-dimensional scene fusion scheduling technology, integrating four types of data: environment, driving, hardware, and physiology, and dynamically adjusting material layering accuracy using AI prediction; it adopts multi-user profile and priority adaptation technology, supporting independent parameter configuration and identity recognition switching to balance personalization and performance; it employs driver physiological state-material linkage technology, adjusting material parameters in real time based on fatigue / attention data to improve driving safety; and it adopts behavioral feedback and safety-related self-optimization technology, combining iterative parameters of driving safety status to adapt to the needs of high-risk scenarios.

[0094] This invention includes a method for dynamic scheduling of in-vehicle layered PBR materials, comprising closed-loop logic of scene feature input, model prediction, and parameter output; a self-optimization system for in-vehicle PBR materials supporting multiple user profiles, comprising identity recognition, independent storage, and cloud synchronization technologies; a method for linkage and adaptation between driver physiological state and in-vehicle materials, comprising real-time response strategies for state detection and parameter adjustment; and a user priority scheduling system for in-vehicle PBR materials, comprising parameter adaptation rules for three priorities: performance, image quality, and safety.

[0095] The PBR material library parameter adjustment method provided in this embodiment of the invention has at least the following beneficial effects: 1. Enhanced Scene Adaptability and Safety: Physiological state linkage reduces visual interference in fatigue scenarios by 60%, lowering driving safety risks by 35%; core information recognition efficiency is improved by 25% in high-risk scenarios. 2. Performance stability and scheduling efficiency optimization: Layered dynamic scheduling reduces GPU utilization by an average of 30%, memory usage by 25%, and ensures a stable frame rate of ≥45fps at 2.5K resolution; AI prediction reduces scheduling response time to within 300ms. 3. Enhanced Personalized Experience for Multiple Users: Multi-user profiles support independent parameter configuration, with switching loading time ≤ 0.5 seconds and repeated adjustment operations reduced by 80%; the self-optimization mechanism improves the matching degree between scenarios and user preferences to 92%. 4. Enhanced Interaction Flexibility and Driving-Friendliness: Multimodal interaction response time ≤ 0.8 seconds, and strong light information recognition accuracy reaches 98% in safety-priority mode; 5. Outstanding technical scalability and exclusive adaptability: Supports hardware and algorithm alternatives, adapts to different vehicle configurations; the vehicle-specific adaptation rate reaches 100%, which is different from general material library technology.

[0096] The technical solution of the PBR material library parameter adjustment method provided by the present invention includes: receiving collected vehicle multidimensional data, the vehicle multidimensional data including environmental data, driving data, and hardware data; weightedly fusing the vehicle multidimensional data to generate scene feature information; adjusting the PBR material library parameters according to the scene feature information; and updating the rendering effect according to the adjusted PBR material library parameters, so that the material rendering can adapt to the specific needs of the vehicle scene.

[0097] Figure 7 This is a schematic diagram of a PBR material library parameter adjustment device provided in an embodiment of the present invention, as shown below. Figure 7 As shown, the device includes: The transceiver module 11 is used to receive collected vehicle-mounted multidimensional data, which includes environmental data, driving data, and hardware data. The processing module 12 is used to perform weighted fusion of the vehicle-mounted multidimensional data to generate scene feature information; adjust the PBR material library parameters according to the scene feature information; and update the rendering effect according to the adjusted PBR material library parameters.

[0098] In some possible embodiments, the processing module 12 is specifically used to: adjust the PBR material library parameters according to the scene feature information using a basic strategy; the basic strategy is the minimum guarantee strategy for ensuring frame rate stability and driving safety in vehicle scenarios.

[0099] In some possible embodiments, the processing module 12 is specifically configured to: shut down the non-core layer, which includes the special effects particle layer, when the GPU utilization rate is greater than a first threshold; and enable full-layer rendering when the GPU utilization rate is less than or equal to a second threshold. In close-up shots, if the GPU utilization rate is less than or equal to the second threshold, full-layer high-precision rendering is performed; if the GPU utilization rate is greater than the first threshold, the non-core layer is turned off. In the foreground, if the GPU utilization rate is greater than the first threshold, the non-core layer is turned off; if the GPU utilization rate is less than or equal to the second threshold, full-layer rendering is enabled. When the light intensity is greater than or equal to the fourth threshold, the reflectivity of the clear coat is reduced by the first percentage; when the light intensity is less than or equal to the third threshold, the brightness of the metal reflective layer is increased by the second percentage.

[0100] In some possible embodiments, the processing module 12 is further specifically used to: when the vehicle is detected to be in a first scene, adjust the PBR material library parameters according to the scene feature information using an extended strategy; the extended strategy is an optimization strategy that balances performance, image quality, security and personalized adaptation on the basis of the basic strategy. The processing module 12 is specifically used for: inputting the scene feature information into the prediction model to obtain the predicted PBR material library parameters; reading the priority parameters set by the user; and adjusting the PBR material library parameters according to the predicted PBR material library parameters and the priority parameters.

[0101] In some possible embodiments, the processing module 12 is further configured to: in response to a user's operation, generate a user operation instruction and parse the user operation instruction to obtain adjustment information, wherein the adjustment information is parameter adjustment information or user switching information; adjust the PBR material library parameters according to the adjustment information, and update the rendering effect according to the adjusted PBR material library parameters.

[0102] In some possible embodiments, the vehicle-mounted multidimensional data also includes physiological data; the processing module 12 is further configured to: generate driver physiological state information when weighted and fused the vehicle-mounted multidimensional data; adjust the PBR material library parameters according to the physiological state information; and update the rendering effect according to the adjusted PBR material library parameters.

[0103] In some possible embodiments, the processing module 12 is further configured to: record the adjustment data and extended data of the PBR material library parameters; the transceiver module 11 is further configured to: upload the adjustment data and extended data to the cloud, and receive model parameter update information sent by the cloud; the processing module 12 is further configured to: update the parameters of the prediction model according to the model parameter update information.

[0104] In some possible embodiments, the processing module 12 is specifically used to: when the physiological state information is a fatigue state, reduce the reflectivity of all materials according to a third percentage, and increase the contrast of the core information area according to a fourth percentage; when the physiological state information is a distracted state, enhance the transparency of the core area material according to a fifth percentage, and add a border highlight layer according to a sixth percentage.

[0105] In the technical solution provided by the embodiments of the present invention, the vehicle-mounted multi-dimensional data is received, including environmental data, driving data, and hardware data; the vehicle-mounted multi-dimensional data is weighted and fused to generate scene feature information; the PBR material library parameters are adjusted according to the scene feature information, and the rendering effect is updated according to the adjusted PBR material library parameters, so that the material rendering can adapt to the specific needs of the vehicle-mounted scene.

[0106] This application provides a storage medium that includes a stored program, wherein the program controls the device where the storage medium is located to execute the above-described method when it is running.

[0107] This application provides a computer device including a memory and a processor. The memory is used to store information including program instructions, and the processor is used to control the execution of the program instructions. When the program instructions are loaded and executed by the processor, they implement the steps of the above-described method.

[0108] Figure 8 This application provides a schematic diagram of the structure of a computer device, as shown in the embodiment of the present application. Figure 8 As shown, the computer device 20 includes a processor 21, a memory 22, and a computer program 23 stored in the memory 22 and executable on the processor 21. When the computer program 23 is executed by the processor 21, it implements the PBR material library parameter adjustment method in the embodiment. To avoid repetition, it will not be described in detail here.

[0109] Computer device 20 includes, but is not limited to, processor 21 and memory 22. Those skilled in the art will understand that... Figure 8 This is merely an example of computer device 20 and does not constitute a limitation on computer device 20. It may include more or fewer components than shown, or combine certain components, or different components. For example, computer device 20 may also include input / output devices, network access devices, buses, etc.

[0110] The processor 21 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0111] The memory 22 can be an internal storage unit of the computer device 20, such as a hard disk or RAM of the computer device 20. The memory 22 can also be an external storage device of the computer device 20, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the computer device 20. Furthermore, the memory 22 can include both internal and external storage units of the computer device 20. The memory 22 is used to store computer programs and other programs and data required by the computer device 20. The memory 22 can also be used to temporarily store data that has been output or will be output.

[0112] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0113] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0114] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0115] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in a combination of hardware and software functional units.

[0116] The integrated units implemented as software functional units described above can be stored in a computer-readable storage medium. These software functional units, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0117] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for adjusting parameters in a PBR material library, characterized in that, The method includes: Receive collected vehicle-mounted multidimensional data, which includes environmental data, driving data, and hardware data; The vehicle-mounted multidimensional data is weighted and fused to generate scene feature information; Adjust the PBR material library parameters based on the scene feature information, and update the rendering effect based on the adjusted PBR material library parameters.

2. The method according to claim 1, characterized in that, The step of adjusting the PBR material library parameters based on the scene feature information includes: Based on the scene feature information, the parameters of the PBR material library are adjusted using a basic strategy; the basic strategy is the minimum guarantee strategy for ensuring frame rate stability and driving safety in vehicle scenarios.

3. The method according to claim 2, characterized in that, The step of adjusting the PBR material library parameters according to the scene feature information using a basic strategy includes: When the GPU utilization rate exceeds the first threshold, the non-core layer, including the special effects particle layer, is shut down. When the GPU utilization rate is less than or equal to the second threshold, full-layer rendering is enabled. In close-up shots, if the GPU utilization rate is less than or equal to the second threshold, full-layer high-precision rendering is performed; if the GPU utilization rate is greater than the first threshold, the non-core layer is turned off. In the foreground, if the GPU utilization rate is greater than the first threshold, the non-core layer is turned off; if the GPU utilization rate is less than or equal to the second threshold, full-layer rendering is enabled. When the light intensity is greater than or equal to the fourth threshold, the reflectivity of the clear coat is reduced by the first percentage. When the light intensity is less than or equal to the third threshold, the brightness of the metal reflective layer is increased according to the second percentage.

4. The method according to claim 2, characterized in that, The step of adjusting the PBR material library parameters according to the scene feature information also includes: When a vehicle is detected to be in the first scene, the PBR material library parameters are adjusted using an extended strategy based on the scene feature information. The extended strategy is an optimization strategy that balances performance, image quality, safety and personalized adaptation on the basis of the basic strategy. The step of adjusting the PBR material library parameters using an extended strategy based on the scene feature information includes: The scene feature information is input into the prediction model to obtain the predicted PBR material library parameters; Read the priority parameters set by the user; The PBR material library parameters are adjusted based on the predicted PBR material library parameters and the priority parameters.

5. The method according to claim 1, characterized in that, The method further includes: In response to a user's operation, a user operation instruction is generated and the user operation instruction is parsed to obtain adjustment information, which is parameter adjustment information or user switching information; The PBR material library parameters are adjusted according to the adjustment information, and the rendering effect is updated according to the adjusted PBR material library parameters.

6. The method according to claim 1, characterized in that, The in-vehicle multidimensional data also includes physiological data; the method further includes: When the vehicle-mounted multidimensional data is weighted and fused, the driver's physiological state information is also generated. The PBR material library parameters are adjusted based on the physiological state information, and the rendering effect is updated based on the adjusted PBR material library parameters.

7. The method according to claim 4, characterized in that, The method further includes: Record the adjustment and extension data of the PBR material library parameters; Upload the adjusted and expanded data to the cloud; Receive model parameter update information sent from the cloud; The parameters of the prediction model are updated based on the model parameter update information.

8. The method according to claim 6, characterized in that, The step of adjusting the PBR material library parameters based on the physiological state information includes: When the physiological state information indicates a state of fatigue, the reflectivity of all materials is reduced by a third percentage, and the contrast of the core information area is increased by a fourth percentage. When the physiological state information indicates a state of distraction, the transparency of the core area material is enhanced by a fifth percentage, and a border highlight layer is added by a sixth percentage.

9. A PBR material library parameter adjustment device, characterized in that, include: The transceiver module is used to receive collected vehicle-mounted multidimensional data, which includes environmental data, driving data, and hardware data. The processing module is used to perform weighted fusion of the vehicle-mounted multidimensional data to generate scene feature information; Adjust the PBR material library parameters based on the scene feature information, and update the rendering effect based on the adjusted PBR material library parameters.

10. A storage medium, characterized in that, The storage medium includes a stored program, wherein, when the program is executed, it controls the device containing the storage medium to perform the method described in any one of claims 1 to 8.

11. A computer device, characterized in that, The method includes a memory and a processor, wherein the memory is used to store information including program instructions, and the processor is used to control the execution of the program instructions, wherein when the program instructions are loaded and executed by the processor, they implement the steps of the method according to any one of claims 1 to 8.

Images