Seat adjusting method and related device

Abstract

The invention discloses a seat adjusting method and a related device, and relates to the technical field of seat adjustment, and the method comprises the following steps: obtaining a head image and a leg image of a driver; determining eyebrow or eye height according to the head image, and obtaining leg state information according to the leg image; based on a pre-constructed seat position prediction model, on the basis of eyebrow or eye height and leg state information, a seat position matched with a driver is predicted, and the seat position prediction model is constructed based on sample data of a plurality of personnel samples with different heights; the sample data of any person sample comprises a seat position matched with the person sample and a head image and a leg image when the person sample sits on a seat at the matched seat position; and the seat is adjusted to the seat position matched with the driver. According to the seat adjusting method, the seat can be automatically adjusted to the seat position matched with the driver, manual operation of a user is not needed, the adjusting convenience is greatly improved, and the user experience is good.

Description

Technical Field

[0001] The present application relates to the technical field of seat adjustment, and in particular to a seat adjustment method and related devices. Background Art

[0002] In order to provide the driver with a reasonable field of vision and operating space, the driver needs to adjust the vehicle's seat to a suitable position before driving the vehicle.

[0003] Most of the current seat adjustment solutions are manual adjustment solutions, that is, the driver manually adjusts the car seat to a suitable position. However, manual adjustment of the car seat is less convenient and has a poor user experience. Summary of the invention

[0004] In view of this, the present application provides a seat adjustment method and related devices to solve the problems of poor convenience and poor user experience of existing seat adjustment solutions. The technical solution is as follows:

[0005] A first aspect of the present application provides a seat adjustment method, comprising:

[0006] Acquire a head image and a leg image of a driver sitting on a seat, wherein the head image includes a facial area;

[0007] Determining the height of the driver's eyebrows or eyes according to the driver's head image, and acquiring the driver's leg status information according to the driver's leg image;

[0008] Based on a pre-built seat position prediction model, a seat position suitable for the driver is predicted based on the height of the driver's eyebrows or eyes and the driver's leg state information, wherein the seat position prediction model is built based on sample data of several samples of people with different heights, and the sample data of any sample of the person includes a seat position suitable for the sample of the person, and a head image and a leg image of the sample of the person when the sample of the person sits in a seat at the suitable seat position;

[0009] The seat is adjusted to a seat position suitable for the driver.

[0010] In a possible implementation, determining the height of the driver's eyebrow center or eye level according to the driver's head image includes:

[0011] Based on a two-dimensional facial key point detection algorithm, facial key point detection is performed on the driver's head image to obtain the coordinates of the driver's eyebrows or eyes on the image plane;

[0012] Performing depth estimation on the driver's head image based on a depth estimation algorithm to obtain the depth of the driver's eyebrows or eyes in a camera coordinate system;

[0013] Determine the coordinates of the driver's eyebrow center or eye in the camera coordinate system according to the camera intrinsic parameters, the coordinates of the driver's eyebrow center or eye in the image plane, and the depth of the driver's eyebrow center or eye in the camera coordinate system;

[0014] The height of the driver's eyebrow center or eye is obtained according to the coordinates of the driver's eyebrow center or eye in the camera coordinate system.

[0015] In a possible implementation manner, acquiring the driver's leg status information according to the driver's leg image includes:

[0016] Segmenting a leg region from the driver's leg image based on a pre-trained leg segmentation model, wherein the leg segmentation model is trained using a training leg image with the leg region marked;

[0017] The driver's leg status information is determined according to the segmented leg regions.

[0018] In a possible implementation, the seat position prediction model based on the pre-constructed seat position is used as a basis to predict the seat position suitable for the driver based on the height of the driver's eyebrows or eyes and the driver's leg status information, including:

[0019] Based on a pre-built seat position prediction model, the seat height suitable for the driver is predicted based on the eyebrow center or eye height of the driver;

[0020] Based on the seat position prediction model and taking the driver's leg status information as a basis, a front and rear position of a seat suitable for the driver is predicted.

[0021] In a possible implementation, the seat adjustment method further includes: determining the depth of the driver's brow center or eyes according to the driver's head image;

[0022] The method of predicting a front and rear seat position suitable for the driver based on the seat position prediction model and taking the driver's leg state information as a basis includes:

[0023] Based on the seat position prediction model, the front and rear position of the seat suitable for the driver is predicted based on the driver's leg status information and combined with the driver's eyebrow center or eye depth.

[0024] In a possible implementation, the process of constructing the seat position prediction model includes:

[0025] Obtain sample data of several samples of people of different heights;

[0026] For any sample data of any person sample: determine the eyebrow center or eye height according to the head image in the sample data, and determine the leg status information according to the leg image in the sample data, use the eyebrow center or eye height and the leg status information determined according to the image data in the sample data as the first training sample, and use the seat position in the sample data as the sample label of the first training sample, to obtain a first training sample labeled with the seat position;

[0027] After obtaining a plurality of first training samples with seat positions marked according to the sample data of the plurality of samples of persons of different heights, the seat position prediction model is trained by using the plurality of first training samples with seat positions marked.

[0028] In a possible implementation, the process of constructing the seat position prediction model includes:

[0029] Obtain sample data of several samples of people of different heights;

[0030] For any sample data of any person sample: taking the head image and the leg image in the sample data as the second training sample, taking the seat position in the sample data as the sample label of the second training sample, and obtaining a second training sample labeled with the seat position;

[0031] After obtaining a plurality of second training samples with seat positions marked according to the sample data of the plurality of samples of people of different heights, the seat position prediction model is trained by using the plurality of second training samples with seat positions marked;

[0032] The step of determining the height of the driver's eyebrow center or eye area according to the driver's head image, and acquiring the driver's leg status information according to the driver's leg image, includes:

[0033] Based on the trained seat position prediction model, the height of the driver's eyebrow or eye is determined based on the driver's head image, and the driver's leg status information is obtained based on the driver's leg image.

[0034] In a possible implementation, the leg status information includes: a leg curve, an angle between the thigh and calf, and a length of the thigh and calf exceeding the seat.

[0035] In a possible implementation, the sample data of any personnel sample also includes key human indicators of the personnel sample;

[0036] The key anthropometric indicators of a human sample include some or all of the following indicators: height, upper body height, thigh length, and calf length.

[0037] A second aspect of the present application provides a seat adjustment device, comprising: an image acquisition module, a parameter acquisition module, a seat position determination module, and a seat adjustment module;

[0038] The image acquisition module is used to acquire a head image and a leg image of the driver sitting on the seat, wherein the head image includes a facial area;

[0039] The parameter acquisition module is used to determine the height of the driver's eyebrows or eyes according to the driver's head image, and to acquire the driver's leg status information according to the driver's leg image;

[0040] The seat position determination module is used to predict a seat position suitable for the driver based on a pre-constructed seat position prediction model, based on the driver's eyebrow center or eye height and the driver's leg state information, wherein the seat position prediction model is constructed based on sample data of a number of samples of people with different heights, and the sample data of any sample of the person includes a seat position suitable for the sample of the person, and a head image and a leg image of the sample of the person when the sample of the person sits in a seat at the suitable seat position;

[0041] The seat adjustment module is used to adjust the seat to a seat position suitable for the driver.

[0042] A third aspect of the present application provides an electronic device, comprising at least one processor and a memory connected to the processor, wherein:

[0043] The memory is used to store computer programs;

[0044] The processor is used to execute the computer program so that the electronic device can implement the steps of any one of the seat adjustment methods described above.

[0045] A fourth aspect of the present application provides a computer storage medium, which carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement the steps of any one of the seat adjustment methods described above.

[0046] A fifth aspect of the present application provides a computer program product, comprising computer-readable instructions, which, when executed on an electronic device, enables the electronic device to implement the steps of any one of the seat adjustment methods described above.

[0047] By means of the above technical solution, the seat adjustment method provided by the present application first obtains the head image and leg image of the driver sitting on the seat, then determines the driver's eyebrow or eye height based on the driver's head image, and obtains the driver's leg status information based on the driver's leg image, and then predicts the seat position suitable for the driver based on the driver's eyebrow or eye height and the driver's leg status information based on the pre-built seat position prediction model, and finally adjusts the seat to the seat position suitable for the driver. The seat adjustment method provided by the present application can automatically determine the seat position suitable for the driver based on the driver's head image and leg image, and can automatically adjust the seat to the seat position suitable for the driver without the driver having to manually adjust the seat, which greatly improves the convenience of adjustment and brings a better experience to the user. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0049] Figure 1 A flowchart of a seat adjustment method provided in an embodiment of the present application;

[0050] Figure 2 A flowchart of a specific example of a seat adjustment method provided in an embodiment of the present application;

[0051] Figure 3 A flowchart of another specific example of the seat adjustment method provided in an embodiment of the present application;

[0052] Figure 4 This is a schematic diagram of the structure of a seat adjustment device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0054] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0055] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0056] The seat adjustment method provided in the present application can be applied to a control device, which can be a host device of a vehicle, a seat controller specifically used to control a seat, or other control devices capable of controlling a seat.

[0057] The control device may include a processor and a memory, and may also include other functional modules.

[0058] The processor is the control center of the control device, which uses various interfaces and lines to connect various parts of the entire control device, and performs various functions of the control device by running or executing instructions stored in the memory and calling data stored in the memory. Optionally, the processor may include one or more processing units.

[0059] The memory may include volatile memory, such as random access memory (RAM). The memory may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD) or solid state drive (SSD).

[0060] The memory may be used to store software codes related to the seat adjustment method, and the processor may execute the software codes in the memory and may also schedule other units to implement corresponding functions.

[0061] In some embodiments, the processor and memory can be implemented on a single chip, and in some embodiments, they can also be implemented on separate chips. The processor can also be used to generate corresponding operation control signals, send them to corresponding components of the computing and processing device, read and process data in the software, especially read and process data and programs in the memory, so that each functional module therein performs corresponding functions, thereby controlling the corresponding components to act according to the requirements of the instructions.

[0062] Most of the current seat adjustment solutions are manual adjustment solutions, that is, the driver manually adjusts the car seat to a suitable position. However, manual adjustment of the car seat is less convenient and has a poor user experience.

[0063] In view of the fact that the current seat adjustment solutions are less convenient and provide a poor user experience, the inventor of this case conducted research. The initial idea was to pre-store the driver's facial information and a seat position that matches the driver, and to associate the driver's facial information with the corresponding seat position information. When the driver sits on the seat, the driver's facial image is collected to determine whether there is pre-stored facial information that matches the facial information in the currently collected facial image. If so, the seat is adjusted to the seat position associated with the facial information that matches the facial information in the currently collected facial image.

[0064] The inventor of this case studied the above idea and found that although the above idea does not require the driver to manually adjust the seat, it requires the pre-storage of the driver's facial information and the corresponding seat position, and it cannot cover drivers who do not have pre-stored facial information and corresponding seat positions.

[0065] In response to the problems existing in the above-mentioned ideas, the inventor of this case continued to conduct research and came up with a seat adjustment solution based on human height, that is, obtaining the driver's head position information and seat position information, determining the driver's height based on the driver's head position information and seat position information, and adjusting the seat position based on the driver's height.

[0066] The inventor of this case studied the above-mentioned height-based seat adjustment scheme and found that although the height-based seat adjustment scheme does not need to pre-store the driver's facial information and the corresponding seat position, the scheme cannot accurately determine the driver's height, and there are differences in the upper and lower body proportions of different drivers. As a result, the adjustment effect of this scheme is not good.

[0067] In view of the defects of the above-mentioned height-based seat adjustment scheme, the inventor of this case conducted further research and finally proposed a seat adjustment method through continuous research. This seat adjustment method does not require manual adjustment by the user, nor does it require pre-storage of the driver's facial information and the corresponding seat position, and can ensure the adjustment effect.

[0068] Next, the seat adjustment method provided by the present application is introduced through the following embodiments.

[0069] See also Figure 1 , shows a schematic flow chart of a seat adjustment method provided in an embodiment of the present application, the method may include:

[0070] Step S101: Acquire a head image and a leg image of a driver sitting on a seat.

[0071] Among them, the head image contains the facial area.

[0072] The driver's head image may be acquired based on the first camera. As for the driver's leg image, in one possible implementation, the driver's leg image may be acquired based on the second camera, and in another possible implementation, the driver's leg image may be acquired based on an infrared sensor.

[0073] A first camera capable of capturing an image of the driver's head and a second camera or infrared sensor capable of capturing an image of the driver's legs may be arranged on the vehicle (for example, the second camera or infrared sensor may be arranged next to the footwell to capture an image of the entire leg). Then, when the seat adjustment triggering conditions are met (for example, the driver places his foot on the brake pedal, the driver presses a button that triggers seat adjustment, etc.), the driver's head image is acquired based on the first camera, and the driver's leg image is acquired based on the second camera or infrared sensor.

[0074] Step S102: determining the driver's eyebrow center or eye height according to the driver's head image, and acquiring the driver's leg status information according to the driver's leg image.

[0075] Among them, the driver's leg status information may include but is not limited to the leg curve (the leg curve may indicate whether the leg is straight or bent), the calf-femoral angle (the angle between the thigh and the calf), the length of the calf-femoral length exceeding the seat, etc.

[0076] Step S103: Based on the pre-constructed seat position prediction model, the seat position suitable for the driver is predicted based on the driver's eyebrow center or eye height and the driver's leg status information.

[0077] It should be noted that the seat position adapted to the driver refers to a position that is suitable for the driver and makes the driver feel comfortable.

[0078] The seat position prediction model in this embodiment has the ability to predict a seat position suitable for the driver based on the driver's eyebrow center or eye height and the driver's leg status information.

[0079] The seat position prediction model in this embodiment is constructed based on sample data of several personnel samples of different heights. The sample data of any personnel sample includes the seat position suitable for the personnel sample, and the head image (including the facial area) and leg image of the personnel sample when sitting in the seat at the suitable seat position.

[0080] Based on a pre-constructed seat position prediction model, the process of predicting a seat position suitable for the driver based on the driver's eyebrow center or eye height and the driver's leg condition information may include: based on the pre-constructed seat position prediction model, based on the driver's eyebrow center or eye height, predicting a seat height suitable for the driver; based on the pre-constructed seat position prediction model, based on the driver's leg condition information, predicting a seat fore-and-aft position suitable for the driver.

[0081] That is, this embodiment predicts the seat height suitable for the driver according to the driver's eyebrow center or eye height, and predicts the seat front and rear position suitable for the driver according to the driver's leg status information, thereby obtaining a seat position suitable for the driver.

[0082] Step S104: Adjust the seat to a seat position suitable for the driver.

[0083] Specifically, the seat can be automatically adjusted to a seat position suitable for the driver by driving the seat adjustment motor.

[0084] When adjusting the seat to a seat position suitable for the driver, the height of the seat can be adjusted to a seat height suitable for the driver first, and then the front and rear position of the seat can be adjusted to a seat front and rear position suitable for the driver.

[0085] Of course, the present embodiment is not limited thereto. For example, the front and rear position of the seat may be first adjusted to a front and rear position of the seat that matches the driver, and then the height of the seat may be adjusted to a seat height that matches the driver.

[0086] The seat adjustment method provided in the embodiment of the present application first obtains the head image and leg image of the driver sitting on the seat, then determines the driver's eyebrow or eye height based on the driver's head image, and obtains the driver's leg status information based on the driver's leg image, and then predicts the seat position that is suitable for the driver based on the driver's eyebrow or eye height and the driver's leg status information based on the pre-built seat position prediction model, and finally adjusts the seat to the seat position that is suitable for the driver. The seat adjustment method provided in the embodiment of the present application can automatically determine the seat position that is suitable for the driver based on the driver's head image and leg image, and can automatically adjust the seat to the seat position that is suitable for the driver without the driver having to manually adjust the seat, which greatly improves the convenience of adjustment and brings a better experience to the user.

[0087] Based on the seat adjustment method provided in the above embodiment, a specific example of the seat adjustment method is given below. Figure 2 , showing a schematic diagram of the process of this specific example, which may include:

[0088] Step S201: Acquire a head image and a leg image of a driver sitting on a seat.

[0089] The driver's head image can be obtained based on the first camera, and the driver's leg image can be obtained based on the second camera or the infrared sensor. In order to obtain more accurate leg status information later, the present embodiment preferably obtains the driver's leg image based on the infrared sensor.

[0090] Step S202: determining the driver's eyebrow center or eye height according to the driver's head image, and obtaining the driver's leg status information according to the driver's leg image.

[0091] In one possible implementation, the process of determining the height of the driver's brow center or eye based on the driver's head image may include: performing facial key point detection on the driver's head image based on a two-dimensional facial key point detection algorithm to obtain the coordinates (x, y) of the driver's brow center or eye on the image plane; performing depth estimation on the driver's head image based on a depth estimation algorithm to obtain the depth z of the driver's brow center or eye in the camera coordinate system; determining the coordinates (three-dimensional coordinates) of the driver's brow center or eye in the camera coordinate system based on the camera intrinsic parameters, the coordinates (x, y) of the driver's brow center or eye on the image plane, and the depth z of the driver's brow center or eye in the camera coordinate system; and obtaining the height of the driver's brow center or eye based on the coordinates of the driver's brow center or eye in the camera coordinate system.

[0092] In a possible implementation, when the depth of the driver's head image is estimated based on the depth estimation algorithm to obtain the depth z of the driver's eyebrow center or eye in the camera coordinate system, the depth of the driver's head image can be estimated based on a global depth estimation algorithm (such as the AdaBins algorithm) to obtain a rough global eyebrow center or eye depth, and a local depth estimation algorithm (such as the LocalBins algorithm) is used to perform depth estimation on the driver's head image to obtain a fine local eyebrow center or eye depth, and then the global eyebrow center or eye depth is fused with the local eyebrow center or eye depth (for example, the global eyebrow center or eye depth can be fused with the local eyebrow center or eye depth by a weighted summation fusion method), and the fused eyebrow center or eye depth is used as the final eyebrow center or eye depth.

[0093] In one possible implementation, the process of obtaining the driver's leg status information based on the driver's leg image may include: segmenting the leg area from the driver's leg image based on a pre-trained leg segmentation model; and determining the driver's leg status information based on the segmented leg area.

[0094] The leg segmentation model is trained using training leg images with leg areas marked. Optionally, the leg segmentation model may be a model based on a UNet network. In order to reduce the amount of computation, the leg segmentation model may be a model based on a lightweight UNet network.

[0095] The driver's leg status information may include but is not limited to the leg curve, the angle between the thigh and calf, the length of the thigh and calf beyond the seat, etc.

[0096] Step S203: Based on the pre-trained seat position prediction model, the seat position suitable for the driver is predicted based on the driver's eyebrow center or eye height and the driver's leg status information.

[0097] The seat position prediction model in this embodiment is trained using the training data in the first training data set. The first training data set includes multiple training data. Each training data in the first training data set is a first training sample marked with the seat position. The first training sample includes the eyebrow or eye height and leg status information.

[0098] The process of constructing the first training data set may include:

[0099] Step a1: Obtain sample data of several samples of people of different heights.

[0100] The sample data of any person sample includes a seat position that matches the person sample, and a head image and a leg image of the person sample when the person sample sits on a seat at the matched seat position.

[0101] Optionally, the sample data of any person sample may also include key human indicators of the person sample. The key human indicators of a person sample include part or all of the following indicators: height, upper body height, thigh length, and calf length.

[0102] In this embodiment, the sample data of any personnel sample may be one piece or multiple pieces.

[0103] Considering that there may be more than one seat position that matches a personnel sample, and that the seat positions that match different personnel samples of the same height may be different, the present application can obtain multiple sample data for any personnel sample of any height, and the seat positions in the multiple sample data are different. Exemplarily, for a personnel sample with a height of 175 cm, the seat position pos1 that matches the personnel sample and the head image and leg image of the personnel sample when sitting on the seat at pos1 are obtained, the seat position pos2 that matches the personnel sample and the head image and leg image of the personnel sample when sitting on the seat at pos2 are obtained, ..., the seat position posN that matches the personnel sample and the head image and leg image of the personnel sample when sitting on the seat at posN are obtained, so as to obtain multiple sample data of a personnel sample with a height of 175 cm.

[0104] Step a2, for any sample data of any personnel sample: determine the height of the center of eyebrows or eyes according to the head image in the sample data, determine the leg status information according to the leg image in the sample data, and use the height of the center of eyebrows or eyes and the leg status information determined according to the image data in the sample data as the first training sample (if the sample data contains key indicators of the human body, the height of the center of eyebrows or eyes, the leg status information and the key indicators of the human body determined according to the image data in the sample data are used as the first training sample), and use the seat position in the sample data as the sample label of the first training sample to obtain a first training sample marked with the seat position.

[0105] Step a3: after obtaining a plurality of first training samples with marked seat positions based on sample data of samples of persons of different heights, a first training data set is formed by the plurality of first training samples with marked seat positions.

[0106] After obtaining the first training data set, the seat position prediction model may be trained using the first training data set. Specifically, the training process of the seat position prediction model may include:

[0107] Step b1: Obtain a first training sample from a first training data set.

[0108] Step b2: input the first training sample into the seat position prediction model to obtain a seat position prediction result corresponding to the first training sample output by the seat position prediction model.

[0109] After the first training sample is input into the seat position prediction model, the seat position prediction model predicts an adapted seat position according to the first training sample and outputs a seat position prediction result.

[0110] Step b3: determining the seat position prediction loss according to the seat position prediction result corresponding to the first training sample and the seat position marked by the first training sample.

[0111] The difference between the seat position prediction result corresponding to the first training sample and the seat position marked by the first training sample can be calculated as the seat position prediction loss.

[0112] Step b4: update the parameters of the seat position prediction model according to the seat position prediction loss.

[0113] In this embodiment, the seat position prediction model is trained by taking the seat position predicted by the seat position prediction model according to the first training sample as a training target to be close to the seat position marked by the first training sample.

[0114] The seat position prediction model is iteratively trained for multiple times according to the process of the above steps b1 to b4 until the training end condition is met, wherein the training end condition may be, but is not limited to, model convergence, reaching a set number of training iterations, etc. After the training is completed, a seat position prediction model can be obtained that can predict a seat position suitable for the driver based on the driver's eyebrow center or eye height and leg state information.

[0115] In this embodiment, based on the pre-trained seat position prediction model, the process of predicting the seat position that matches the driver based on the driver's eyebrow or eye height and the driver's leg state information may include: based on the pre-trained seat position prediction model, based on the driver's eyebrow or eye height, predicting the seat height that matches the driver; based on the pre-trained seat position prediction model, based on the driver's leg state information, predicting the front and rear position of the seat that matches the driver. That is, the driver's eyebrow or eye height and the driver's leg state information are input into the seat position prediction model, and the seat position prediction model predicts the seat height that matches the driver based on the driver's eyebrow or eye height, and predicts the front and rear position of the seat that matches the driver based on the driver's leg state information.

[0116] From the above content, it can be known that the present embodiment predicts a seat position suitable for the driver according to the height of the driver's brow or eye and the driver's leg information. In order to improve the prediction effect, in another possible implementation, the seat position suitable for the driver can be predicted according to the height of the driver's brow or eye and the driver's leg information, combined with the driver's brow or eye depth.

[0117] Specifically, based on the pre-trained seat position prediction model, the seat height information suitable for the driver is predicted based on the driver's eyebrow or eye height, and based on the pre-trained seat position prediction model, the front and rear position of the seat suitable for the driver is predicted based on the driver's leg state information and the driver's eyebrow or eye depth. That is, the driver's eyebrow or eye height, the driver's eyebrow or eye depth, and the driver's leg state information are input into the seat position prediction model, and the seat position prediction model predicts the seat height suitable for the driver based on the driver's eyebrow or eye height, and predicts the front and rear position of the seat suitable for the driver based on the driver's leg state information and the driver's eyebrow or eye depth. The introduction of the driver's eyebrow or eye depth can improve the accuracy of the seat position prediction model in predicting the front and rear position of the seat.

[0118] In addition, it should be noted that if the driver's eyebrow or eye depth is introduced to assist in prediction, the eyebrow or eye depth can also be introduced during the model training stage, that is, the first training sample mentioned above includes the eyebrow or eye depth in addition to the eyebrow or eye height and leg status information.

[0119] Step S204: Adjust the seat to a seat position suitable for the driver.

[0120] The height of the seat can be adjusted to a seat height suitable for the driver first, and then the front and rear position of the seat can be adjusted to a seat front and rear position suitable for the driver.

[0121] The seat adjustment method provided in the embodiment of the present application first obtains the head image and leg image of the driver sitting on the seat, then determines the driver's eyebrow or eye height based on the driver's head image, and obtains the driver's leg status information based on the driver's leg image, then uses a first training sample based on a number of seat positions marked (the first training sample includes the eyebrow or eye height and leg status information) to train to obtain a seat position prediction model, and predicts the seat position that is suitable for the driver based on the driver's eyebrow or eye height and the driver's leg status information, and finally adjusts the seat to the seat position that is suitable for the driver. The seat adjustment method provided in the embodiment of the present application can use the seat position prediction model obtained by training based on a number of first training samples marked with seat positions to predict the seat position that is suitable for the driver, and then adjust the seat to the seat position that is suitable for the driver. The seat adjustment method provided in the embodiment of the present application can automatically adjust the seat to the seat position that is suitable for the driver without the driver having to manually adjust it, which greatly improves the convenience of adjustment and brings a better experience to the user.

[0122] Based on the seat adjustment method provided in the above embodiment, another specific example of the seat adjustment method is given below. Figure 3 , showing a schematic diagram of the process of this specific example, which may include:

[0123] Step S301: Acquire a head image and a leg image of a driver sitting on a seat.

[0124] The driver's head image can be obtained based on the first camera, and the driver's leg image can be obtained based on the second camera or the infrared sensor. In order to obtain more accurate leg status information later, the present embodiment preferably obtains the driver's leg image based on the infrared sensor.

[0125] Step S302: Based on the pre-trained seat position prediction model, the driver's eyebrow center or eye height is determined based on the driver's head image, and the driver's leg status information is obtained based on the driver's leg image.

[0126] Step S303: Based on the pre-trained seat position prediction model, a seat position suitable for the driver is predicted based on the driver's eyebrow center or eye height and the driver's leg status information.

[0127] The seat position prediction model in this embodiment is trained using the training data in the constructed second training data set, wherein the second training data set includes multiple training data, each training data in the second training data set is a second training sample marked with the seat position, and the second training sample includes a head image and a leg image.

[0128] The process of constructing the second training data set may include:

[0129] Step c1: Obtain sample data of several samples of people of different heights.

[0130] The sample data of any person sample includes a seat position that matches the person sample, and a head image and a leg image of the person sample when the person sample sits on a seat at the matched seat position.

[0131] Optionally, the sample data of any person sample may also include key human indicators of the person sample. The key human indicators of a person sample include part or all of the following indicators: height, upper body height, thigh length, and calf length.

[0132] In this embodiment, the sample data of any personnel sample may be one piece or multiple pieces.

[0133] Step c2, for any sample data of any person sample: use the head image and leg image in the sample data as the second training sample (if the sample data includes key human body indicators, the head image and leg image and key human body indicators in the sample data can be used as the second training sample), and use the seat position in the sample data as the sample label of the second training sample to obtain a second training sample marked with the seat position.

[0134] Step c3: after obtaining a plurality of second training samples with marked seat positions based on the sample data of a plurality of samples of people of different heights, a second training data set is formed by the plurality of second training samples with marked seat positions.

[0135] After obtaining the second training data set, the seat position prediction model can be trained using the second training data set. Specifically, the training process of the seat position prediction model may include:

[0136] Step d1: Obtain a second training sample from a second training data set.

[0137] Step d2: input the second training sample into the seat position prediction model to obtain a seat position prediction result corresponding to the second training sample output by the seat position prediction model.

[0138] After the second training sample is input into the seat position prediction model, the seat position prediction model first determines the eyebrow or eye height based on the head image in the second training sample, and determines the leg status information based on the leg image in the second training sample, and then predicts the appropriate seat position based on the determined eyebrow or eye height and leg status information, and outputs the seat position prediction result.

[0139] Optionally, the seat position prediction model can also determine the eyebrow center or eye depth based on the head image in the second training sample, and then, when predicting the suitable seat position, the determined eyebrow center or eye depth is combined for prediction.

[0140] It should be noted that if the second training sample includes key indicators of the human body, the seat position prediction model can predict the seat position in combination with the key indicators of the human body.

[0141] Step d3: Determine the seat position prediction loss according to the seat position prediction result corresponding to the second training sample and the seat position marked by the second training sample.

[0142] The difference between the seat position prediction result corresponding to the second training sample and the seat position marked by the second training sample can be calculated as the seat position prediction loss.

[0143] Step d4: update the parameters of the seat position prediction model according to the seat position prediction loss.

[0144] In this embodiment, the seat position prediction model is trained by taking the seat position predicted by the seat position prediction model according to the second training sample as a training target to be close to the seat position marked by the second training sample.

[0145] The seat position prediction model is iteratively trained multiple times according to the process of the above steps d1 to d4 until the training end condition is met, wherein the training end condition may be, but is not limited to, model convergence, reaching a set number of training iterations, etc. After the training is completed, a seat position prediction model can be obtained that can predict a seat position suitable for the driver based on the driver's head image and leg image.

[0146] In this embodiment, the process of predicting a seat position suitable for the driver based on a trained seat position prediction model may include: inputting the driver's head image and leg image into the trained seat position prediction model, the trained seat position prediction model determines the driver's brow center or eye height according to the driver's head image (optionally, the driver's brow center or eye depth can also be determined), and determines the driver's leg status information according to the driver's leg image, and then predicts a seat height suitable for the driver according to the driver's brow center or eye height, predicts a seat fore-and-aft position suitable for the driver according to the driver's leg status information (optionally, combined with the driver's brow center or eye depth), and outputs a seat position prediction result including a seat height suitable for the driver and a seat fore-and-aft position.

[0147] Step S304: Adjust the seat to a seat position suitable for the driver.

[0148] The height of the seat can be adjusted to a seat height suitable for the driver first, and then the front and rear position of the seat can be adjusted to a seat front and rear position suitable for the driver.

[0149] The seat adjustment method provided in the embodiment of the present application first obtains the head image and leg image of the driver sitting on the seat, and then uses the seat position prediction model trained based on a number of second training samples marked with seat positions (the second training samples include head images and leg images) to predict the seat position that matches the driver based on the driver's head image and leg image, and finally adjusts the seat to the seat position that matches the driver. The seat adjustment method provided in the embodiment of the present application can use the seat position prediction model trained based on a number of second training samples marked with seat positions to predict the seat position that matches the driver, and then adjust the seat to the seat position that matches the driver. The seat adjustment method provided in the embodiment of the present application can automatically adjust the seat to a seat position that matches the driver without the need for manual adjustment by the driver, and has good adjustment convenience and user experience.

[0150] The seat adjustment method provided in the embodiment of the present application is introduced above, and the device corresponding to the above-mentioned seat adjustment method will be introduced below.

[0151] See also Figure 4 , Figure 4 This is a structural schematic diagram of a seat adjustment device provided in an embodiment of the present application. The seat adjustment device may include: an image acquisition module 401, a parameter acquisition module 402, a seat position determination module 403 and a seat adjustment module 404.

[0152] The image acquisition module 401 is used to acquire the head image and leg image of the driver sitting on the seat.

[0153] Wherein, the head image includes a facial area.

[0154] The parameter acquisition module 402 is used to determine the height of the driver's eyebrows or eyes according to the driver's head image, and to acquire the driver's leg status information according to the driver's leg image.

[0155] The seat position determination module 403 is used to predict a seat position suitable for the driver based on a pre-built seat position prediction model and based on the driver's eyebrow or eye height and the driver's leg status information.

[0156] Among them, the seat position prediction model is constructed based on sample data of several personnel samples of different heights. The sample data of any personnel sample includes the seat position adapted to the personnel sample, and the head image and leg image of the personnel sample when sitting in the seat at the adapted seat position.

[0157] The seat adjustment module 404 is used to adjust the seat to a seat position suitable for the driver.

[0158] In a possible implementation, when the parameter acquisition module 402 determines the height of the driver's eyebrow center or eye area according to the driver's head image, it is specifically used to:

[0159] Based on a two-dimensional facial key point detection algorithm, facial key point detection is performed on the driver's head image to obtain the coordinates of the driver's eyebrows or eyes on the image plane;

[0160] Performing depth estimation on the driver's head image based on a depth estimation algorithm to obtain the depth of the driver's eyebrows or eyes in a camera coordinate system;

[0161] Determine the coordinates of the driver's eyebrow center or eye in the camera coordinate system according to the camera intrinsic parameters, the coordinates of the driver's eyebrow center or eye in the image plane, and the depth of the driver's eyebrow center or eye in the camera coordinate system;

[0162] The height of the driver's eyebrow center or eye is obtained according to the coordinates of the driver's eyebrow center or eye in the camera coordinate system.

[0163] In a possible implementation, when the parameter acquisition module 402 acquires the driver's leg status information according to the driver's leg image, it is specifically configured to include:

[0164] Segmenting a leg region from the driver's leg image based on a pre-trained leg segmentation model, wherein the leg segmentation model is trained using a training leg image with the leg region marked;

[0165] The driver's leg status information is determined according to the segmented leg regions.

[0166] In a possible implementation, the seat position determination module 403 is specifically used to predict the seat position suitable for the driver based on the pre-built seat position prediction model and the height of the driver's eyebrows or eyes and the driver's leg state information:

[0167] Based on a pre-built seat position prediction model, the seat height suitable for the driver is predicted based on the eyebrow center or eye height of the driver;

[0168] Based on the seat position prediction model and taking the driver's leg status information as a basis, a front and rear position of a seat suitable for the driver is predicted.

[0169] In a possible implementation, the parameter acquisition module 402 is further configured to determine the depth of the driver's brow center or eyes according to the driver's head image.

[0170] The seat position determination module 403 is specifically used to predict the front and rear position of the seat adapted to the driver based on the seat position prediction model and the leg state information of the driver:

[0171] Based on the seat position prediction model, the front and rear position of the seat suitable for the driver is predicted based on the driver's leg status information and combined with the driver's eyebrow center or eye depth.

[0172] In a possible implementation, the seat adjustment device provided in the embodiment of the present application may further include: a first model building module. The first model building module is used to:

[0173] Obtain sample data of several samples of people of different heights;

[0174] For any sample data of any person sample: determine the eyebrow center or eye height according to the head image in the sample data, and determine the leg status information according to the leg image in the sample data, use the eyebrow center or eye height and the leg status information determined according to the image data in the sample data as the first training sample, and use the seat position in the sample data as the sample label of the first training sample, to obtain a first training sample labeled with the seat position;

[0175] After obtaining a plurality of first training samples with seat positions marked according to the sample data of the plurality of samples of persons of different heights, the seat position prediction model is trained by using the plurality of first training samples with seat positions marked.

[0176] In a possible implementation, the seat adjustment device provided in the embodiment of the present application may further include: a second model building module. The second model building module is used to:

[0177] Obtain sample data of several samples of people of different heights;

[0178] For any sample data of any person sample: taking the head image and the leg image in the sample data as the second training sample, taking the seat position in the sample data as the sample label of the second training sample, and obtaining a second training sample labeled with the seat position;

[0179] After obtaining a plurality of second training samples with seat positions marked according to the sample data of the plurality of samples of persons of different heights, the seat position prediction model is trained by using the plurality of second training samples with seat positions marked.

[0180] The step of determining the height of the driver's eyebrow center or eye area according to the driver's head image, and acquiring the driver's leg status information according to the driver's leg image, includes:

[0181] Based on the trained seat position prediction model, the height of the driver's eyebrow or eye is determined based on the driver's head image, and the driver's leg status information is obtained based on the driver's leg image.

[0182] In a possible implementation, the leg status information acquired by the parameter acquisition module 402 may include, but is not limited to: leg curve, thigh-feet angle, thigh-feet length exceeding the seat, etc.

[0183] The seat adjustment device provided in the embodiment of the present application first obtains the head image and leg image of the driver sitting on the seat, then determines the driver's eyebrow or eye height based on the driver's head image, and obtains the driver's leg status information based on the driver's leg image, and then predicts the seat position suitable for the driver based on the driver's eyebrow or eye height and the driver's leg status information based on the pre-built seat position prediction model, and finally adjusts the seat to the seat position suitable for the driver. The seat adjustment device provided in the embodiment of the present application can automatically determine the seat position suitable for the driver based on the driver's head image and leg image, and can automatically adjust the seat to the seat position suitable for the driver without the driver having to manually adjust the seat, which greatly improves the convenience of adjustment and brings a better experience to the user.

[0184] An embodiment of the present application also provides an electronic device, which may include: at least one processor, at least one communication interface, at least one memory and at least one communication bus.

[0185] In the embodiment of the present application, the number of the processor, the communication interface, the memory, and the communication bus is at least one, and the processor, the communication interface, and the memory communicate with each other through the communication bus;

[0186] The processor may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application;

[0187] The memory may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory;

[0188] The memory stores a program, and the processor can call the program stored in the memory, and the program is used to implement the steps of the seat adjustment method provided in the above embodiment.

[0189] An embodiment of the present application also provides a computer storage medium, which carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement the steps of the seat adjustment method provided in the above embodiment.

[0190] An embodiment of the present application also provides a computer program product, including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements the steps of the seat adjustment method provided in the above embodiment.

[0191] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0192] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0193] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0194] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

Claims

1. A seat adjustment method, characterized in that: include: Acquire a head image and a leg image of a driver sitting on a seat, wherein the head image includes a facial area; Determining the height of the driver's eyebrows or eyes according to the driver's head image, and acquiring the driver's leg status information according to the driver's leg image; Based on a pre-built seat position prediction model, a seat position suitable for the driver is predicted based on the height of the driver's eyebrows or eyes and the driver's leg state information, wherein the seat position prediction model is built based on sample data of several samples of people with different heights, and the sample data of any sample of the person includes a seat position suitable for the sample of the person, and a head image and a leg image of the sample of the person when the sample of the person sits on a seat at the suitable seat position; The seat is adjusted to a seat position suitable for the driver.

2. The seat adjustment method according to claim 1, characterized in that: The step of determining the height of the driver's brow center or eye area according to the driver's head image includes: Based on a two-dimensional facial key point detection algorithm, facial key point detection is performed on the driver's head image to obtain the coordinates of the driver's eyebrows or eyes on the image plane; Performing depth estimation on the driver's head image based on a depth estimation algorithm to obtain the depth of the driver's eyebrows or eyes in a camera coordinate system; Determine the coordinates of the driver's eyebrow center or eye in the camera coordinate system according to the camera intrinsic parameters, the coordinates of the driver's eyebrow center or eye in the image plane, and the depth of the driver's eyebrow center or eye in the camera coordinate system; The height of the driver's eyebrow center or eye is obtained according to the coordinates of the driver's eyebrow center or eye in the camera coordinate system.

3. The seat adjustment method according to claim 1, characterized in that: The acquiring the driver's leg status information according to the driver's leg image includes: Segmenting a leg region from the driver's leg image based on a pre-trained leg segmentation model, wherein the leg segmentation model is trained using a training leg image with the leg region marked; The driver's leg status information is determined according to the segmented leg regions.

4. The seat adjustment method according to claim 1, characterized in that: The method of predicting a seat position suitable for the driver based on the pre-built seat position prediction model and taking the eyebrow center or eye height of the driver and the leg state information of the driver as a basis includes: Based on a pre-built seat position prediction model, the seat height suitable for the driver is predicted based on the eyebrow center or eye height of the driver; Based on the seat position prediction model and taking the driver's leg status information as a basis, a front and rear position of a seat suitable for the driver is predicted.

5. The seat adjustment method according to claim 4, characterized in that: Also includes: Determining the driver's eyebrow center or eye depth according to the driver's head image; The method of predicting a front and rear seat position suitable for the driver based on the seat position prediction model and taking the driver's leg state information as a basis includes: Based on the seat position prediction model, the front and rear position of the seat suitable for the driver is predicted based on the driver's leg status information and combined with the driver's eyebrow center or eye depth.

6. The seat adjustment method according to claim 1, characterized in that: The process of constructing the seat position prediction model includes: Obtain sample data of several samples of people of different heights; For any sample data of any person sample: determine the eyebrow center or eye height according to the head image in the sample data, and determine the leg status information according to the leg image in the sample data, use the eyebrow center or eye height and the leg status information determined according to the image data in the sample data as the first training sample, and use the seat position in the sample data as the sample label of the first training sample, to obtain a first training sample labeled with the seat position; After obtaining a plurality of first training samples with seat positions marked according to the sample data of the plurality of samples of persons of different heights, the plurality of first training samples with seat positions marked are used to train a seat position prediction model.

7. The seat adjustment method according to claim 1, characterized in that: The process of constructing the seat position prediction model includes: Obtain sample data of several samples of people of different heights; For any sample data of any person sample: taking the head image and the leg image in the sample data as the second training sample, taking the seat position in the sample data as the sample label of the second training sample, and obtaining a second training sample labeled with the seat position; After obtaining a plurality of second training samples with seat positions marked according to the sample data of the plurality of samples of people of different heights, the seat position prediction model is trained by using the plurality of second training samples with seat positions marked; The step of determining the height of the driver's eyebrow center or eye area according to the driver's head image, and acquiring the driver's leg status information according to the driver's leg image, includes: Based on the trained seat position prediction model, the height of the driver's eyebrow or eye is determined based on the driver's head image, and the driver's leg status information is obtained based on the driver's leg image.

8. The seat adjustment method according to claim 1, characterized in that: The leg status information includes: leg curve, leg-feet angle, and leg-feet length exceeding the seat.

9. The seat adjustment method according to claim 1, characterized in that: The sample data of any personnel sample also includes the key human indicators of the personnel sample; The key anthropometric indicators of a human sample include some or all of the following indicators: height, upper body height, thigh length, and calf length.

10. A seat adjustment device, characterized in that: include: An image acquisition module, a parameter acquisition module, a seat position determination module and a seat adjustment module; The image acquisition module is used to acquire a head image and a leg image of the driver sitting on the seat, wherein the head image includes a facial area; The parameter acquisition module is used to determine the height of the driver's eyebrows or eyes according to the driver's head image, and to acquire the driver's leg status information according to the driver's leg image; The seat position determination module is used to predict a seat position suitable for the driver based on a pre-constructed seat position prediction model, based on the driver's eyebrow center or eye height and the driver's leg state information, wherein the seat position prediction model is constructed based on sample data of a number of samples of people with different heights, and the sample data of any sample of the person includes a seat position suitable for the sample of the person, and a head image and a leg image of the sample of the person when the sample of the person sits in a seat at the suitable seat position; The seat adjustment module is used to adjust the seat to a seat position suitable for the driver.

11. An electronic device, characterized in that: The method comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the electronic device can implement the steps of the seat adjustment method as described in any one of claims 1 to 9.

12. A computer storage medium, characterized in that: The storage medium carries one or more computer programs, and when the one or more computer programs are executed by an electronic device, the electronic device can implement the steps of the seat adjustment method as described in any one of claims 1 to 9.

13. A computer program product, characterized in that The method comprises computer-readable instructions, and when the computer-readable instructions are executed on an electronic device, the electronic device implements the steps of the seat adjustment method according to any one of claims 1 to 9.

Images