Seat adjusting method and device, storage medium, controller, vehicle and product
By acquiring occupant images on the in-vehicle camera, determining three-dimensional feature information and predicting body posture information, the seats are accurately adjusted, and the problems of high cost and low accuracy in the prior art are solved, and the riding experience is improved.
Patent Information
- Application Number
- CN202510138454.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-07
- Publication Date
- 2025-04-25
AI Technical Summary
The existing methods of automatically adjusting the seat position in the car are costly and have low accuracy, and need to be improved.
The target image of the target occupant is obtained by the camera in the target vehicle, the three-dimensional characteristic information of the key points on the occupant is determined based on the image, the body posture information of the occupant is predicted, and the seat is adjusted according to the body posture information.
While reducing costs, improving the accuracy of seat adjustment and improving the user's riding experience.
Smart Images
Figure CN119872358A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of vehicles, and in particular to a seat adjustment method, device, storage medium, controller, vehicle and product. Background Art
[0002] With the development of science and technology, vehicles are becoming more and more popular among users, and they have more and more functions. For example, vehicles can automatically adjust the position of seats in the car to provide a better riding experience for people in the car.
[0003] However, the current methods for automatically adjusting the position of seats in a car are either costly or inaccurate, and therefore, the methods for automatically adjusting the position of seats in a car need to be improved. Summary of the invention
[0004] The embodiment of the present application provides a seat adjustment method, which can reduce costs while improving accuracy, so as to at least partially solve the above technical problems.
[0005] In order to achieve the above object, according to a first aspect of the present application, a seat adjustment method is provided, comprising:
[0006] Acquiring a target image of a target occupant in the target vehicle through a camera in the target vehicle;
[0007] Based on the target image, determine the three-dimensional feature information of key points on the target occupant, and based on the three-dimensional feature information of the key points, predict the body posture information of the target occupant;
[0008] Based on the body posture information of the target occupant, the seat in which the target occupant sits is adjusted.
[0009] Optionally, the body posture information includes body shape information, and the body shape information of the target occupant is predicted based on the three-dimensional feature information of the key points, including:
[0010] Based on the three-dimensional feature information of the key points, predict the volume information of the target occupant;
[0011] Based on the volume information, the body shape information of the target occupant is predicted.
[0012] Optionally, predicting the body shape information of the target occupant based on the volume information includes:
[0013] Based on the volume information, predict the weight information of the target occupant;
[0014] Based on the weight information and the three-dimensional feature information of the key points, the body shape information of the target occupant is predicted.
[0015] Optionally, the predicting of the body shape information of the target occupant based on the weight information and the three-dimensional feature information of the key points includes:
[0016] Predicting the initial body shape information of the target occupant based on the three-dimensional feature information of the key points through human factor engineering method;
[0017] Based on the weight information, the initial body shape information is adjusted to obtain the body shape information of the target occupant.
[0018] Optionally, the key points are key points of the upper body of the target occupant, the initial body shape information includes initial height information, and the method of predicting the initial body shape information of the target occupant based on the three-dimensional feature information of the key points by using the human factors engineering method includes:
[0019] Predicting the height information of the upper body of the target occupant based on the three-dimensional feature information of the key points through human factor engineering method;
[0020] Based on the height information of the upper body, predicting the height information of the lower body of the target occupant;
[0021] Based on the height information of the upper body and the height information of the lower body, the initial height information of the target occupant is determined.
[0022] Optionally, this embodiment also includes:
[0023] Based on the target image, determining attribute information of the target occupant;
[0024] The initial body shape information is adjusted based on the weight information to obtain the body shape information of the target passenger, including:
[0025] Based on the weight information and the attribute information, the initial body shape information is adjusted to obtain the body shape information of the target occupant.
[0026] Optionally, determining the attribute information of the target occupant based on the target image includes:
[0027] The attribute information of the target occupant is determined based on the target image through the attribute recognition model.
[0028] Optionally, the attribute information includes at least one of the age and gender of the target occupant.
[0029] Optionally, determining the three-dimensional feature information of key points on the target occupant based on the target image includes:
[0030] Determine the depth feature information of key points on the target occupant based on the target image by using a depth recognition model;
[0031] Determine the two-dimensional position feature information of the key points based on the target image through a position recognition model;
[0032] Based on the depth feature information and the two-dimensional position feature information, the three-dimensional feature information of the key point is determined.
[0033] Optionally, the depth feature information is used to indicate the absolute depth of the key point.
[0034] Optionally, the above-mentioned posture information includes at least one of the body shape information and sitting posture information of the above-mentioned target occupant.
[0035] Optionally, adjusting the seat on which the target occupant sits based on the body posture information of the target occupant includes:
[0036] Determining target configuration information of a seat for the target occupant based on the body posture information of the target occupant;
[0037] determining adjustment information based on the current configuration information of the seat in which the target occupant is seated and the target configuration information;
[0038] Based on the adjustment information, the seat in which the target occupant sits is adjusted.
[0039] Optionally, the camera is a monocular camera.
[0040] According to a second aspect of the present application, a seat adjustment device is provided, comprising:
[0041] An acquisition module, used for acquiring a target image of a target occupant in the target vehicle through a camera in the target vehicle;
[0042] a determination module, configured to determine three-dimensional feature information of key points on the target occupant based on the target image, and predict body posture information of the target occupant based on the three-dimensional feature information of the key points;
[0043] The adjustment module is used to adjust the seat occupied by the target occupant based on the body posture information of the target occupant.
[0044] According to a third aspect of the present application, a computer-readable storage medium is also provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the embodiments of the present application are implemented.
[0045] According to a fourth aspect of the present application, a controller is also provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps in the embodiments of the present application are implemented.
[0046] According to a fifth aspect of the present application, a vehicle is also provided, comprising the controller in an embodiment of the present application.
[0047] According to the sixth aspect of the present application, a computer program product is also provided, including a computer program or instructions, which implement the steps in the embodiments of the present application when executed by a processor.
[0048] To summarize, in the embodiments of the present application, since the target image of the target occupant in the target vehicle is obtained through the camera in the target vehicle, the cost can be reduced. Based on the target image, the three-dimensional feature information of the key points on the target occupant is determined, and based on the three-dimensional feature information of the key points, the body posture information of the target occupant is predicted; based on the body posture information of the target occupant, the seat for the target occupant is adjusted, thereby improving the accuracy of seat adjustment. Therefore, the present application can improve the accuracy of seat adjustment while reducing costs, thereby further improving the user's riding experience.
[0049] Other features and advantages of the present application will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application, and those skilled in the art can obtain other drawings based on these drawings without creative work.
[0051] In order to more completely understand the present application and its beneficial effects, the following description will be given in conjunction with the accompanying drawings, wherein the same figure numbers represent the same parts in the following description.
[0052] Figure 1 is a flowchart of the steps of a seat adjustment method provided in an exemplary embodiment of the present application;
[0053] Figure 2 is a schematic diagram of another seat adjustment method provided in an exemplary embodiment of the present application;
[0054] Figure 3 is a schematic diagram of another seat adjustment method provided in an exemplary embodiment of the present application;
[0055] Figure 4 is a schematic diagram of another seat adjustment method provided in an exemplary embodiment of the present application;
[0056] Figure 5 is a schematic diagram of a seat adjustment device provided in an exemplary embodiment of the present application. DETAILED DESCRIPTION
[0057] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.
[0058] This application provides a method for seat adjustment. Figure 1 The seat adjustment method provided in the embodiment of the present application includes steps 100 to 300, which are described in detail below.
[0059] Step 100, obtaining a target image of a target occupant in the target vehicle through a camera in the target vehicle.
[0060] The camera includes at least one, and when there is one camera, the camera is a monocular camera, and the monocular camera refers to a single camera in the target vehicle. When the camera is a monocular camera, the cost can be further reduced.
[0061] The target occupant may be any occupant of the target vehicle, for example, the target occupant may be a driver or a passenger of the target vehicle. The target image of the target occupant may at least include upper body information of the target occupant.
[0062] It is understandable that a monocular camera can be used to capture the target image of one target occupant, or a monocular camera can also capture the target images of multiple target occupants, which is not limited in this embodiment.
[0063] This embodiment obtains the target image of the target occupant in the target vehicle through a monocular camera in the target vehicle. There is no need to obtain the relevant information of the target occupant through a camera outside the target vehicle, nor is there any need to obtain the relevant information of the target occupant through multiple cameras. Therefore, the cost of seat adjustment can be reduced.
[0064] Optionally, when the target vehicle receives an adjustment instruction, it can obtain a target image of the target occupant in the target vehicle through a monocular camera in the target vehicle, or the target vehicle can also obtain a target image of the target occupant on the seat in response to detecting that the seat is occupied, which is not limited to this embodiment.
[0065] Step 200, based on the target image, determine the three-dimensional feature information of the key points on the target occupant, and based on the three-dimensional feature information of the key points, predict the body posture information of the target occupant.
[0066] Among them, the key points refer to specific position points on the target occupant. The key points may include the key points of the upper body of the target occupant, or the key points may include the key points of the upper body and the key points of the lower body of the target occupant. At this time, based on the target image, the key points of the upper body of the target occupant can be determined first, and then based on the key points of the upper body, the key points of the lower body of the target occupant can be determined.
[0067] When the key points are key points on the target occupant's body, the key points may be, for example, at least one of the key points on the eyes, the key points on the nose, the key points on the ears, the key points on the shoulders, the key points on the hands, the key points on the thighs, the key points on the knees, and the key points on the torso.
[0068] The three-dimensional feature information of the key point may refer to the three-dimensional position information of the key point. The body posture information of the target occupant may include at least one of body shape information and sitting posture information of the target occupant.
[0069] The body shape information of the target occupant is used to describe the external shape and outline of the target occupant, which can be set according to actual conditions. For example, the body shape information of the target occupant may include at least one of the target occupant's height information, arm length information, and leg length information.
[0070] Sitting posture information refers to the position and posture of each part of the target occupant's body when sitting.
[0071] In this embodiment, the target image is identified to obtain three-dimensional feature information of key points on the target occupant, and the target occupant's posture information is predicted based on the three-dimensional feature information of the key points so that the seat can be adjusted based on the target occupant's posture information, thereby improving the accuracy of seat adjustment.
[0072] In some embodiments, based on the target image, the method for determining the three-dimensional feature information of the key points on the target occupant can be set according to actual conditions. For example, the three-dimensional feature information of the key points on the target occupant can be determined based on the target image through a histogram of oriented gradients (HOG) or a trained neural network model. This embodiment is not limited here.
[0073] When determining the three-dimensional feature information of the key points on the target occupant based on the target image through the trained neural network model, determining the three-dimensional feature information of the key points on the target occupant based on the target image includes:
[0074] Determine the depth feature information of key points on the target occupant based on the target image through the depth recognition model;
[0075] Through the position recognition model, based on the target image, the two-dimensional position feature information of the key points is determined;
[0076] Based on the depth feature information and the two-dimensional position feature information, the three-dimensional feature information of the key point is determined.
[0077] Among them, the depth feature information is used to indicate the depth of the key point, that is, the depth feature information is used to represent the third-dimensional position feature information of the key point, so that the depth feature information and the two-dimensional position feature information can be combined to form the three-dimensional feature information of the key point.
[0078] The depth recognition model and the position recognition model are both trained neural network models, and their types can be set according to actual conditions. For example, the depth recognition model and the position recognition model can be a YOLOv7-pose model or a convolutional neural network model, which is not limited in this embodiment.
[0079] In this embodiment, the depth feature information of key points on the target occupant is determined based on the target image through the depth recognition model, the two-dimensional position feature information of the key points is determined based on the target image through the position recognition model, and the three-dimensional feature information of the key points is determined based on the depth feature information and the two-dimensional position feature information. The three-dimensional feature information is obtained through a trained neural network model, the accuracy of the three-dimensional feature information is improved, and thus the accuracy of seat adjustment based on the three-dimensional feature information is improved.
[0080] It is understandable that before determining the depth feature information of the key points on the target occupant based on the target image through the depth recognition model and determining the two-dimensional position feature information of the key points based on the target image through the position recognition model, the depth recognition model and the position recognition model need to be trained. The process of training the depth recognition model and the position recognition model can be:
[0081] Acquire a first training sample set, where the first training sample set includes at least one first sample image;
[0082] Determine sample depth feature information of sample key points based on the first sample image by using an initial depth recognition model;
[0083] Determine sample two-dimensional position feature information of sample key points based on the first sample image by using an initial position recognition model;
[0084] Based on the sample depth feature information and the sample two-dimensional position feature information, the sample three-dimensional feature information of the sample key point is obtained;
[0085] Determine a first loss value based on the sample three-dimensional feature information and label information corresponding to the first sample image;
[0086] Based on the first loss value, the initial depth recognition model and the position recognition model are trained to obtain a depth recognition model and a position recognition model.
[0087] The first training sample set may be a sample set that has been subjected to sample enhancement processing, and the sample enhancement processing may be, for example, at least one of rotation, scaling, and translation.
[0088] The first sample image may include at least one of images captured of sample occupants in different vehicles, images captured of sample occupants in vehicles in different environments, and images captured of sample occupants at different viewing angles.
[0089] The first loss value may be a value of a first loss function, or the first loss value may include a value of the first loss function and a first regularization value. The type of the first loss function may be set according to actual conditions, for example, the first loss function may be a mean square error (MSE) or a cross entropy function, which is not limited in this embodiment.
[0090] The method for determining the first regularization value may be set according to actual conditions. For example, the first regularization value may be determined by Dropout, L1, or L2, which is not limited in this embodiment.
[0091] When the loss value includes the value of the first loss function and the regularization value, the overfitting phenomenon can be alleviated.
[0092] In some embodiments, during the training of the initial depth recognition model and the position recognition model, a learning rate (LR) decay mechanism may be used to gradually reduce the learning rate as the training progresses, so as to improve the stability and generalization ability of the model.
[0093] In some embodiments, after the depth recognition model and the position recognition model are obtained, the depth recognition model and the position recognition model may be subjected to model quantization processing and / or pruning processing to reduce the storage space and inference space of the model.
[0094] In some embodiments, the depth feature information is used to indicate the absolute depth of the key point. In this embodiment, the depth feature information is used to indicate the absolute depth of the key point, so that the posture information obtained based on the three-dimensional feature information of the key point is more accurate, thereby making the adjustment of the seat based on the posture information more accurate.
[0095] In some embodiments, the method of predicting the body posture information of the target occupant based on the three-dimensional feature information of the key points can be set according to actual conditions, and this embodiment does not limit this.
[0096] For example, the body posture information of the target occupant can be predicted based on the three-dimensional feature information of key points through a preset body posture prediction model.
[0097] For example, posture information includes body shape information. Based on the three-dimensional feature information of key points, the body shape information of the target occupant is predicted, including:
[0098] Based on the three-dimensional feature information of key points, the volume information of the target occupant is predicted;
[0099] Based on the volume information, the body shape information of the target occupant is predicted.
[0100] Among them, the target vehicle can be modeled based on the three-dimensional feature information of key points to obtain the volume information of the target occupant, and then the body shape information of the target occupant is predicted based on the volume information.
[0101] Specifically, the body shape information of the target occupant can be predicted based on the volume information through the prediction mapping strategy. The type of the preset mapping strategy can be set according to actual conditions, for example, the preset mapping strategy can be a preset mapping table or a preset function, which is not limited in this embodiment.
[0102] In this embodiment, the volume information of the target occupant is predicted based on the three-dimensional feature information of the key points, and the body shape information of the target occupant is predicted based on the volume information, thereby predicting the body shape information based on the volume information, improving the accuracy of the body shape information, and further improving the accuracy of adjusting the seat based on the body shape information.
[0103] In some embodiments, predicting the body size information of the target occupant based on the volume information includes:
[0104] Based on the volume information, predict the weight information of the target occupant;
[0105] Based on the weight information and the three-dimensional feature information of key points, the body shape information of the target occupant is predicted.
[0106] The weight information of the target occupant can be predicted based on the volume information and density information. The target vehicle predicts the initial body shape information of the target occupant based on the three-dimensional feature information of the key points, and then adjusts the initial body shape information based on the weight information to obtain the body shape information of the target occupant.
[0107] In this embodiment, the weight information of the target occupant is predicted based on the volume information, and the body shape information of the target occupant is predicted based on the weight information and the three-dimensional feature information of the key points, thereby further improving the accuracy of the body shape information and further improving the accuracy of adjusting the seat based on the body shape information.
[0108] In some embodiments, the method of predicting the initial body shape information of the target occupant based on the three-dimensional feature information of the key points can be selected according to the actual situation. For example, the initial body shape information of the target occupant can be predicted based on the three-dimensional feature information of the key points through human factors engineering method, or the initial body shape information of the target occupant can be predicted based on the three-dimensional feature information of the key points through a body shape neural network model. This embodiment is not limited here.
[0109] When the initial body shape information of the target occupant is predicted based on the three-dimensional feature information of the key points by the human factor engineering method, the process of predicting the body shape information of the target occupant based on the weight information and the three-dimensional feature information of the key points can be:
[0110] Through human factor engineering method, the initial body shape information of the target occupant is predicted based on the three-dimensional feature information of key points;
[0111] Based on the weight information, the initial body shape information is adjusted to obtain the body shape information of the target occupant.
[0112] Among them, human factors engineering, also known as ergonomics or ergonomics, is a technology that studies the interaction between people and other elements (other elements such as tools, equipment, machines or environment, etc.).
[0113] The method of adjusting the initial body shape information based on the weight information can be set according to the actual situation. For example, by presetting a first mapping strategy, the weight information and the initial body shape information are mapped to obtain the body shape information.
[0114] The type of the preset first mapping strategy may be selected according to actual conditions. For example, the preset first mapping strategy may be a first mapping table or a function, which is not limited in this embodiment.
[0115] In this embodiment, the initial body shape information of the target occupant is predicted based on the three-dimensional feature information of key points through the ergonomics method, and the initial body shape information is adjusted based on the weight information to obtain the body shape information of the target occupant. The body shape information is obtained through the ergonomics method, and the accuracy of the body shape information is further improved, thereby further improving the accuracy of adjusting the seat based on the body shape information.
[0116] In some embodiments, the key points are key points of the upper body of the target occupant, and the initial body shape information includes initial height information. The initial body shape information of the target occupant is predicted based on the three-dimensional feature information of the key points by human factors engineering method, including:
[0117] Through human factor engineering, the height information of the upper body of the target occupant is predicted based on the three-dimensional feature information of key points;
[0118] Predicting height information of the lower body of the target occupant based on the height information of the upper body;
[0119] Based on the upper body height information and the lower body height information, the initial height information of the target occupant is determined.
[0120] The three-dimensional feature information of the key points of the lower body can also be determined based on the three-dimensional feature information of the key points of the upper body, the height information of the upper body of the target occupant can be predicted based on the three-dimensional feature information of the key points of the upper body, and the height information of the lower body of the target occupant can be predicted based on the three-dimensional feature information of the key points of the lower body. The height information of the lower body of the target occupant can also be called the leg length information of the target occupant.
[0121] In this embodiment, the height information of the upper body of the target occupant is predicted based on the three-dimensional feature information of key points through the ergonomics method, and the height information of the lower body of the target occupant is predicted based on the height information of the upper body. The initial height information of the target occupant is determined based on the height information of the upper body and the height information of the lower body. The height information of the lower body that has not been photographed is determined through the ergonomics method, the human feature information of the target occupant is enriched, the accuracy of the height information is further improved, and the accuracy of adjusting the seat based on the height information is further improved.
[0122] In some embodiments, the present embodiment further includes:
[0123] Determining attribute information of the target occupant based on the target image;
[0124] Based on the weight information, the initial body shape information is adjusted to obtain the body shape information of the target occupant, including:
[0125] Based on the weight information and attribute information, the initial body shape information is adjusted to obtain the body shape information of the target occupant.
[0126] Among them, the attribute information of the target occupant may refer to data describing the demographic characteristics of the target occupant, which may be set according to actual conditions. For example, the attribute information of the target occupant may include at least one of the age and gender of the target occupant, which is not limited in this embodiment.
[0127] The method of adjusting the initial body shape information based on the weight information and the attribute information can be set according to the actual situation. For example, by presetting the second mapping strategy, the weight information, the attribute information and the initial body shape information are mapped to obtain the body shape information.
[0128] The type of the preset second mapping strategy may be selected according to actual conditions. For example, the preset second mapping strategy may be a second mapping table or a function, which is not limited in this embodiment.
[0129] In this embodiment, the attribute information of the target occupant is determined based on the target image, and the initial body shape information is adjusted based on the weight information and the attribute information to obtain the body shape information of the target occupant. The body shape information is obtained based on the weight information and the attribute information, thereby further improving the accuracy of the body shape information and further improving the accuracy of adjusting the seat based on the body shape information.
[0130] In some embodiments, determining attribute information of the target occupant based on the target image includes:
[0131] The attribute information of the target occupant is determined based on the target image through the attribute recognition model.
[0132] Among them, the attribute recognition model can be a trained neural network model, and its type can be set according to actual conditions. For example, the attribute recognition model can be a MobieNet or a convolutional neural network model, which is not limited in this embodiment.
[0133] In this embodiment, the attribute information of the target occupant is determined based on the target image through the attribute recognition model, and the trained neural network model is implemented to obtain the attribute information of the target occupant, thereby improving the accuracy of the attribute information.
[0134] It can be understood that before determining the attribute information of the target occupant based on the target image through the attribute recognition model, the attribute recognition model may be trained. The process of training the attribute recognition model may be:
[0135] Acquire a second training sample set, where the second training sample set includes at least one second sample image;
[0136] Determining sample attribute information of the sample occupant based on the second sample image by using the initial attribute recognition model;
[0137] Determine a second loss value based on the sample attribute information and label information corresponding to the second sample image;
[0138] Based on the second loss value, the initial attribute recognition model is trained to obtain an attribute recognition model.
[0139] The second training sample set may be a sample set that has been subjected to sample enhancement processing, and the sample enhancement processing may be at least one of rotation, scaling and translation. The second training sample set and the first training sample set may be the same sample set or may not be the same sample set.
[0140] The second sample images may include at least one of images captured by taking different sample occupants, images captured by taking sample occupants in different environments, and images captured by taking sample occupants at different viewing angles.
[0141] When the attribute information includes gender, the label information of the second sample image may include "0" or "1", "0" represents male, and "1" represents female. When the attribute information includes age, the label information of the second sample image may include "18-25 years old", "26-30 years old", "31-35 years old", "36-40 years old", "41-45 years old", "46-50 years old", "51-55 years old" or "56-60 years old".
[0142] The second loss value may be a value of a loss function, or the second loss value may include a value of the second loss function and a second regularization value. The type of the second loss function may be set according to actual conditions, for example, the loss function may be a mean square error (MSE) or a cross entropy function, which is not limited in this embodiment.
[0143] The method for determining the second regularization value may be set according to actual conditions. For example, the regularization value may be determined by Dropout, L1, or L2, which is not limited in this embodiment.
[0144] When the loss value includes the value of the second loss function and the regularization value, the overfitting phenomenon can be alleviated.
[0145] In some embodiments, during the training of the initial attribute recognition model, a learning rate (LR) decay mechanism may be used to gradually reduce the learning rate as the training progresses, so as to improve the stability and generalization ability of the model.
[0146] In some embodiments, after the attribute recognition model is obtained, the attribute recognition model may be subjected to model quantization processing and / or pruning processing to reduce the storage space and inference space of the model.
[0147] In some embodiments, when the posture information includes sitting posture information, the sitting posture information can be directly determined based on the three-dimensional feature information of the key points, or the body shape information can be first determined based on the three-dimensional feature information of the key points, and then the sitting posture information can be determined based on the body shape information. This embodiment is not limited here.
[0148] Step 300: adjusting the seat in which the target occupant sits based on the body posture information of the target occupant.
[0149] Among them, adjusting the seat on which the target occupant sits may refer to adjusting at least one of the position, angle, protrusion of the lumbar support portion of the seat, and the position of the headrest on the seat on which the target occupant sits.
[0150] In some embodiments, based on the body posture information of the target occupant, the process of adjusting the seat on which the target occupant sits may be:
[0151] Determining target configuration information of a seat for the target occupant based on the body posture information of the target occupant;
[0152] determining adjustment information based on current configuration information and target configuration information of a seat occupied by a target occupant;
[0153] Based on the adjustment information, the seat in which the target occupant is seated is adjusted.
[0154] Among them, the body posture information of the target occupant can be mapped by a preset third mapping strategy to obtain the target configuration information of the seat that the target occupant sits in. The type of the preset third mapping strategy can be set according to actual conditions, for example, the preset third mapping strategy can be a third mapping table or function, which is not limited in this embodiment.
[0155] The third mapping table or function is constructed based on the configuration information of the seat preferred by the user with different body posture information. The current configuration information of the seat refers to at least one of the current position information of the seat, the current angle information, the current protrusion of the lumbar support part of the seat, and the current position of the headrest on the seat. The target configuration information of the seat refers to at least one of the target position information of the seat, the target angle information, the target protrusion of the lumbar support part of the seat, and the target position of the headrest on the seat.
[0156] The adjustment information is used to indicate the adjustment amount of the seat. After the seat is adjusted based on the adjustment information, the target occupant can be placed in a comfortable state, thereby improving the riding comfort of the target occupant and improving the riding experience of the target occupant.
[0157] In some embodiments, the target vehicle may include an RK3568 chip, a camera is connected via the RK3568 chip, and the seat adjustment method provided in the present application is executed via the RK3568 chip.
[0158] According to the following Figure 2 , the seat adjustment method provided in this application is further explained.
[0159] A target image of a target occupant in the target vehicle is acquired through a monocular camera in the target vehicle. The three-dimensional feature information of key points on the target occupant is determined based on the target image through a depth recognition model and a position recognition model in a computing center in the target vehicle. The attribute information of the target occupant is determined based on the target image through an attribute recognition model. The three-dimensional feature information of the key points, the attribute information and the current configuration information of the seat are uploaded to a cloud computing center. The cloud computing center determines adjustment information based on the three-dimensional feature information and the current configuration information of the seat. The adjustment information is returned to the computing center in the target vehicle. The adjustment information is sent to a control center through the computing center in the target vehicle. The control center adjusts the seat occupied by the target occupant based on the adjustment information and displays the adjusted seat on a display device in the target vehicle.
[0160] The specific implementation method and corresponding beneficial effects of this embodiment can be specifically referred to the above-mentioned seat adjustment method embodiment, and this embodiment will not be described in detail here.
[0161] According to the following Figure 3 , the seat adjustment method provided by the present application is further described. In this embodiment, the seat adjustment method includes a model training process, a model reasoning process and a determination process.
[0162] The model training process includes:
[0163] A first sample image and a second sample image are collected through a monocular camera in a vehicle, a depth label is annotated on the first sample image, a two-dimensional position label and an attribute label are annotated on the second sample image, the first sample image is preprocessed and the second sample image is preprocessed, an initial depth recognition model, an initial position recognition model and an initial attribute recognition model are constructed, the initial depth recognition model is trained through the first sample image to obtain a depth recognition model, and the initial position recognition model and the initial attribute recognition model are trained through the second sample image to obtain a position recognition model and an attribute recognition model.
[0164] The inference process of the model includes:
[0165] A target image of a target occupant in the target vehicle is collected through a monocular camera in the target vehicle, and the target image is preprocessed to obtain a processed image. Based on the processed image, depth information of key points of the upper body is obtained through a depth recognition model. Based on the processed image, two-dimensional position feature information of the key points of the upper body is obtained through a position recognition model. Based on the depth feature information and the two-dimensional position feature information, three-dimensional feature information of the key points of the upper body is determined. Based on the target image, attribute information of the target occupant is obtained through an attribute recognition model.
[0166] The determination process includes:
[0167] Based on the three-dimensional feature information of the key points of the upper body, the weight information is determined. Through the ergonomic method, based on the three-dimensional feature information of the key points of the upper body, the lower body information is determined. Based on the upper body key points and the lower body information, the initial body shape information of the target occupant is determined. Based on the weight information and attribute information, the initial body shape information is adjusted to obtain the body shape information. Based on the body shape information, the sitting posture information is determined. Based on the sitting posture information, the seat is adjusted.
[0168] The specific implementation method and corresponding beneficial effects of this embodiment can be specifically referred to the above-mentioned seat adjustment method embodiment, and this embodiment will not be described in detail here.
[0169] According to the following Figure 4 , the seat adjustment method provided by the present application is further described. In this embodiment, the driver is taken as an example for description.
[0170] The driver sits in the driving seat of the target vehicle, the target vehicle is powered on, and the monocular camera in the target vehicle is powered on. The target image of the driver is collected by the monocular camera. The computing center in the target vehicle determines the weight information and attribute information of the driver based on the target image, determines the initial body shape information of the target occupant through the human factors engineering method, determines the body shape information based on the weight information, attribute information, and initial body shape information through the computing center in the target vehicle, determines the sitting posture information based on the body shape information, determines the adjustment information of the driving seat based on the sitting posture information and the current configuration information of the driving seat, and sends the adjustment information to the control center, which adjusts the driving seat to the corresponding position based on the adjustment information.
[0171] Whether a manual adjustment instruction from the driver is received, if the manual adjustment instruction from the driver is received, the driver's seat is adjusted based on the manual adjustment instruction.
[0172] The specific implementation method and corresponding beneficial effects of this embodiment can be specifically referred to the above-mentioned seat adjustment method embodiment, and this embodiment will not be described in detail here.
[0173] As can be seen from the above, in the embodiment of the present application, since the target image of the target occupant in the target vehicle is obtained through the camera in the target vehicle, the cost can be reduced. Based on the target image, the three-dimensional feature information of the key points on the target occupant is determined, and based on the three-dimensional feature information of the key points, the body posture information of the target occupant is predicted; based on the body posture information of the target occupant, the seat for the target occupant is adjusted, thereby improving the accuracy of seat adjustment. Therefore, the present application can improve the accuracy of seat adjustment while reducing costs, thereby further improving the user's riding experience.
[0174] Figure 5is a structural schematic diagram of a seat adjustment device provided in an embodiment of the present application. Figure 5 , the seat adjustment device may include an acquisition module 501, a determination module 502 and an adjustment module 503. Wherein:
[0175] The acquisition module 501 is used to acquire a target image of a target occupant in the target vehicle through a camera in the target vehicle.
[0176] The determination module 502 is used to determine the three-dimensional feature information of the key points on the target occupant based on the target image, and predict the body posture information of the target occupant based on the three-dimensional feature information of the key points.
[0177] The adjustment module 503 is used to adjust the seat on which the target occupant sits based on the body posture information of the target occupant.
[0178] Among them, the acquisition module 501, the determination module 502 and the adjustment module 503 can be used to execute all the steps in the embodiments corresponding to the above-mentioned seat adjustment method respectively. For the specific implementation methods of these modules and more details, please refer to the corresponding method part, which will not be repeated here.
[0179] An embodiment of the present application also provides a computer-readable storage medium, on which instructions are stored. When the instructions are executed by a processor, the processor is configured to execute the above-mentioned seat adjustment method.
[0180] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.
[0181] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0182] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0183] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0184] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0185] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.
[0186] Computer readable media include permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated communication signals and carrier waves.
[0187] In some embodiments, the present application further provides a controller having a computer program stored thereon, and the computer program, when executed by a processor, implements the steps in the embodiments of the present application. The controller may be located in a target vehicle or in a server. When the controller is located in another server, after the controller obtains the posture information or adjustment information, the posture information or adjustment information is sent to the target vehicle, and the target vehicle adjusts the seat based on the posture information or adjustment information.
[0188] In some embodiments, the present application further provides a vehicle, including the controller in the embodiments of the present application. In this embodiment, the vehicle may be a fuel vehicle, a plug-in hybrid vehicle, or a new energy vehicle, etc., and this embodiment does not specifically limit this.
[0189] In some embodiments, the present application also provides a computer program product, including a computer program or instructions, which implement the steps in the embodiments of the present application when the computer program or instructions are executed by a processor.
[0190] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.
[0191] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0192] The embodiments, implementation methods and related technical features of the present application can be combined and replaced with each other without conflict.
[0193] The above are only preferred embodiments of the present application and do not constitute any form of limitation to the present application. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present application without departing from the content of the technical solution of the present application are still within the scope of the technical solution of the present application.
Claims
1. A seat adjustment method, characterized in that: include: Acquiring a target image of a target occupant in the target vehicle through a camera in the target vehicle; Based on the target image, determining three-dimensional feature information of key points on the target occupant, and based on the three-dimensional feature information of the key points, predicting body posture information of the target occupant; Based on the body posture information of the target occupant, the seat in which the target occupant sits is adjusted.
2. The method according to claim 1, characterized in that The body posture information includes body shape information, and the body shape information of the target occupant is predicted based on the three-dimensional feature information of the key points, including: Predicting the volume information of the target occupant based on the three-dimensional feature information of the key points; Based on the volume information, body shape information of the target occupant is predicted.
3. The method according to claim 2, characterized in that The step of predicting the body shape information of the target occupant based on the volume information includes: Based on the volume information, predicting the weight information of the target occupant; Based on the weight information and the three-dimensional feature information of the key points, the body shape information of the target occupant is predicted.
4. The method according to claim 3, characterized in that: The predicting the body shape information of the target occupant based on the weight information and the three-dimensional feature information of the key points includes: Predicting the initial body shape information of the target occupant based on the three-dimensional feature information of the key points by human factor engineering method; Based on the weight information, the initial body shape information is adjusted to obtain the body shape information of the target occupant.
5. The method according to claim 4, characterized in that The key points are key points of the upper body of the target occupant, the initial body shape information includes initial height information, and the predicting of the initial body shape information of the target occupant based on the three-dimensional feature information of the key points by the human factors engineering method includes: Predicting the height information of the upper body of the target occupant based on the three-dimensional feature information of the key points by human factor engineering method; predicting height information of the lower body of the target occupant based on the height information of the upper body; Based on the height information of the upper body and the height information of the lower body, initial height information of the target occupant is determined.
6. The method according to claim 4, characterized in that Also includes: Determining attribute information of the target occupant based on the target image; The adjusting and processing of the initial body shape information based on the weight information to obtain the body shape information of the target occupant includes: Based on the weight information and the attribute information, the initial body shape information is adjusted to obtain the body shape information of the target occupant.
7. The method according to claim 6, characterized in that The determining, based on the target image, the attribute information of the target occupant includes: The attribute information of the target occupant is determined based on the target image by an attribute recognition model.
8. The method according to claim 6, characterized in that The attribute information includes at least one of an age and a gender of the target occupant.
9. The method according to claim 1, characterized in that: The step of determining the three-dimensional feature information of key points on the target occupant based on the target image includes: Determining depth feature information of key points on the target occupant based on the target image using a depth recognition model; Determining two-dimensional position feature information of the key point based on the target image by a position recognition model; Based on the depth feature information and the two-dimensional position feature information, three-dimensional feature information of the key point is determined.
10. The method according to claim 9, characterized in that The depth feature information is used to indicate the absolute depth of the key point.
11. The method according to claim 1, characterized in that: The body posture information includes at least one of body shape information and sitting posture information of the target occupant.
12. The method according to claim 1, characterized in that The step of adjusting the seat on which the target occupant sits based on the body posture information of the target occupant comprises: Determining target configuration information of a seat for the target occupant based on the body posture information of the target occupant; determining adjustment information based on current configuration information of the seat in which the target occupant is seated and the target configuration information; Based on the adjustment information, the seat in which the target occupant sits is adjusted.
13. The method according to any one of claims 1 to 12, characterized in that: The camera is a monocular camera.
14. A seat adjustment device, characterized in that: include: An acquisition module, used for acquiring a target image of a target occupant in the target vehicle through a camera in the target vehicle; a determination module, configured to determine three-dimensional feature information of key points on the target occupant based on the target image, and predict body posture information of the target occupant based on the three-dimensional feature information of the key points; The adjustment module is used to adjust the seat occupied by the target occupant based on the body posture information of the target occupant.
15. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 13 are implemented.
16. A controller having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 13 are implemented.
17. A vehicle, characterized in that: Comprising a controller as claimed in claim 16.
18. A computer program product, characterized in that The method comprises a computer program or an instruction, wherein when the computer program or the instruction is executed by a processor, the steps of the method according to any one of claims 1 to 13 are implemented.