Smart watch and screen-on control method thereof

Abstract

The invention discloses a smart watch and a screen-on control method thereof. The smart watch comprises an inertial sensor, a gesture recognition module and a CSI information detection module. The inertial sensor is used for detecting corresponding motion parameters representing a wrist wearing the smart watch; the gesture recognition module is used for recognizing a first recognition result representing whether the wrist of the user is lifted or not according to the motion parameters; the CSI information detection module is used for acquiring a channel parameter between the smart watch and a head-mounted smart device of an application user when the first identification result meets a preset condition, and further determining a second identification result representing whether the user lifts the wrist to view the smart watch according to the channel parameter, and when the second identification result represents that the user lifts the wrist to check the smart watch, lightening the screen of the smart watch. The screen of the smart watch can be prevented from being ineffectively lightened, and the power consumption of the smart watch is saved.

Description

Technical Field

[0001] This application relates to the field of signal processing technology, specifically to a smartwatch and its screen-on control method. Background Technology

[0002] Smartwatches are becoming increasingly widespread as wearable devices. To conserve power, smartwatch screens cannot remain constantly lit; instead, they illuminate only when the user needs to check the watch. Therefore, raise-to-wake functionality has become a common choice. Raise-to-wake, also known as wrist-raise-to-wake, refers to the watch automatically lighting up when the wrist is naturally raised and the face is turned towards the user, allowing for immediate viewing of the time and relevant notifications without the need for buttons or touch.

[0003] Smart glasses and AI glasses are becoming increasingly widely used, with more and more people wearing both smartwatches and smart glasses or AI glasses. The collaboration between watches and glasses is also becoming increasingly practical.

[0004] Currently, smartwatches typically use accelerometers and gyroscopes to detect wrist raises and changes in orientation. For example, they might determine whether a wrist raise or a user's glance at the watch is a wrist raise based on thresholds such as angle, angular velocity, displacement amplitude, and duration, and then control the smartwatch screen to light up when a wrist raise is detected. However, the inventors' research on smartwatch screen-light control schemes revealed that in application scenarios such as running, brisk walking, large arm swings, and rope skipping, the above-mentioned screen-light control is prone to misjudgment, thus increasing the watch's power consumption. Summary of the Invention

[0005] In view of this, this application provides a smartwatch and its screen-on control method to solve the problem that the screen-on control scheme of the smartwatch is prone to misjudgment and increases the power consumption of the watch.

[0006] This application provides a smartwatch, which includes an inertial sensor, a gesture recognition module, and a CSI information detection module; The inertial sensor is used to detect motion parameters corresponding to the wrist of the person wearing the smartwatch; The gesture recognition module is used to identify a first recognition result that indicates whether the user's wrist has been raised based on the motion parameters; The CSI information detection module is used to obtain the channel parameters of the smartwatch and the corresponding user's head-mounted smart device when the first recognition result meets the preset conditions, further determine the second recognition result indicating whether the user raises their wrist to view the smartwatch based on the channel parameters, and light up the screen of the smartwatch when the second recognition result indicates that the user raises their wrist to view the smartwatch.

[0007] Optionally, the gesture recognition module is further configured to obtain the confidence level corresponding to the user's wrist being raised when the first recognition result indicates that the user's wrist is raised; if the confidence level is greater than or equal to a first threshold, the screen of the smartwatch is directly turned on; if the confidence level is less than a second threshold, it is determined that the user's wrist is not raised.

[0008] Optionally, the smartwatch further includes a first wireless communication module; the head-mounted smart device includes a second wireless communication module; the gesture recognition module is further configured to determine that the first recognition result meets preset conditions if the confidence level is less than a first threshold and greater than or equal to a second threshold, send the first recognition result to the CSI information detection module, and wake up the wireless communication module to establish a wireless communication connection between the first wireless communication module and the second wireless communication module; the CSI information detection module is configured to receive the first recognition result, detect the communication parameters between the first wireless communication module and the second wireless communication module, and obtain the channel parameters.

[0009] Optionally, the gesture recognition module is further configured to acquire the first recognition result at a first moment; the CSI information detection module is configured to acquire the channel parameters of the smartwatch and the corresponding user's head-mounted smart device when the first recognition result meets preset conditions, and further determine the second recognition result representing whether the user raises their wrist to view the smartwatch based on the channel parameters at a second moment.

[0010] Optionally, the gesture recognition module is pre-set with a first AI model; the first AI model is used to output a first recognition result representing whether the user's wrist is raised and the confidence level of the first recognition result based on the motion parameters.

[0011] Optionally, the process of determining the first AI model includes: acquiring multiple first input samples and first labels corresponding to each first input sample, and inputting multiple sets of first input samples and corresponding first labels into a first initial model for training to obtain the first AI model.

[0012] Optionally, the CSI information detection module is preset with a second AI model; the second AI model is used to output a second recognition result representing whether the user raises their wrist to view the smartwatch based on the channel parameters.

[0013] Optionally, the process of determining the second AI model includes: acquiring multiple second input samples and the second label corresponding to each second input sample, and inputting multiple sets of second input samples and the corresponding second labels into a second initial model for training to obtain the second AI model.

[0014] This application also provides a screen-on control method for a smartwatch, which is applied to any of the aforementioned smartwatches and includes the following steps: The detection characterizes the motion parameters corresponding to the wrist wearing the smartwatch; The first recognition result, which indicates whether the user's wrist has been raised, is identified based on the motion parameters. When the first recognition result meets the preset conditions, the channel parameters of the smartwatch and the corresponding user's head-mounted smart device are obtained. Based on the channel parameters, a second recognition result indicating whether the user raises their wrist to view the smartwatch is further determined. When the second recognition result indicates that the user raises their wrist to view the smartwatch, the screen of the smartwatch is turned on.

[0015] Optionally, the step of identifying a first recognition result indicating whether the user's wrist is raised based on the motion parameters includes: when the first recognition result indicates that the user's wrist is raised, obtaining a confidence level corresponding to the user's wrist being raised; if the confidence level is greater than or equal to a first threshold, then directly lighting up the screen of the smartwatch; if the confidence level is less than a second threshold, then determining that the user's wrist is not raised; if the confidence level is less than the first threshold but greater than or equal to the second threshold, then determining that the first recognition result meets preset conditions.

[0016] In the smartwatch and its screen-on control method described in this application, an inertial sensor detects motion parameters corresponding to the wrist of the person wearing the smartwatch. The gesture recognition module identifies a first recognition result indicating whether the user's wrist has been raised based on the motion parameters. The CSI information detection module is used to obtain the channel parameters between the smartwatch and the corresponding user's head-mounted smart device when the first recognition result meets a preset condition, i.e., the user's wrist is suspected of being raised. Based on the channel parameters, a second recognition result indicating whether the user has raised their wrist to view the smartwatch is further determined. When the second recognition result indicates that the user has raised their wrist to view the smartwatch, the screen of the smartwatch is turned on. This is to further confirm through the CSI information detection module that the user has raised their wrist to view the smartwatch before turning on the screen, which can prevent the smartwatch screen from being ineffectively lit, thereby saving the power consumption of the smartwatch. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the structure of a smartwatch according to an embodiment of this application; Figure 2This is a communication diagram between a smartwatch and a head-mounted smart device according to an embodiment of this application; Figure 3 This is a schematic diagram of a smartwatch structure according to another embodiment of this application; Figure 4 This is a schematic flowchart of a smartwatch screen-on control method according to an embodiment of this application. Detailed Implementation

[0019] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. In the absence of conflict, the following embodiments and their technical features can be combined with each other.

[0020] The first aspect of this application provides a smartwatch, as shown in the reference. Figure 1 As shown, the smartwatch includes an inertial sensor 110, a gesture recognition module 120, and a CSI information detection module 130. The gesture recognition module 120 is connected to both the inertial sensor 110 and the CSI information detection module 130 to read the information collected by the inertial sensor 110 and send relevant information to the CSI information detection module 130.

[0021] The inertial sensor 110 is used to detect motion parameters corresponding to the wrist wearing the smartwatch; specifically, these motion parameters include motion parameters generated when the wrist moves while wearing the smartwatch. Optionally, the inertial sensor 110 may include an accelerometer, and the corresponding motion parameters include parameters such as acceleration displacement and / or gravity direction detected by the accelerometer. Optionally, the inertial sensor 110 may include a gyroscope, and the corresponding motion parameters include parameters such as angular velocity, wrist rotation speed and amplitude detected by the gyroscope.

[0022] The gesture recognition module 120 is used to identify a first recognition result indicating whether the user's wrist is raised based on the motion parameters. Specifically, the first recognition result may include whether the user's wrist is raised, whether the user's wrist is not raised, and the confidence level of whether the user's wrist is raised. Specifically, the gesture recognition module 120 can use recognition tools such as AI models to analyze and process the motion parameters to identify the first recognition result corresponding to the motion parameters. The user's wrist being raised in the first recognition result can refer to a specific wrist raising action where the user intends to look at a watch. Not all wrist raising actions are considered to indicate an intention to look at a watch. For example, raising the wrist with the watch face down, or unconscious raising the wrist while walking, will not be considered as the first recognition result of the user's wrist being raised. Whether the first recognition result is that the user's wrist is raised or not raised can be determined by the AI ​​model set in the gesture recognition module 120.

[0023] The CSI information detection module 130 is used to acquire the channel parameters of the smartwatch and the corresponding user's head-mounted smart device when the first recognition result meets preset conditions, and further determine a second recognition result indicating whether the user has raised their wrist to view the smartwatch based on the channel parameters. Optionally, the CSI information detection module 130 can also be used to acquire the channel parameters of the smartwatch and the corresponding user's head-mounted smart device and the motion parameters obtained by the inertial sensor 110 when the first recognition result meets preset conditions, and determine a second recognition result indicating whether the user has raised their wrist to view the smartwatch based on the motion parameters and channel parameters, so as to further improve the reliability of the obtained second recognition result. The CSI information detection module 130 can also turn on the screen of the smartwatch when the second recognition result indicates that the user has raised their wrist to view the smartwatch, so as to further confirm the user's raising of their wrist to view the smartwatch before turning on the screen, which can avoid the smartwatch screen being ineffectively lit, thereby saving the power consumption of the smartwatch.

[0024] Optionally, the preset conditions include the condition that the user's wrist shows a tendency to lift and / or the user's wrist is suspected of being lifted. When the user's wrist shows a tendency to lift and / or the user's wrist is suspected of being lifted, the CSI information detection module 130 acquires the channel parameters of the smartwatch and the corresponding user's head-mounted smart device, and then further identifies whether the user's wrist is lifted based on the channel parameters, which can improve the reliability of the recognition results.

[0025] Optionally, the head-mounted smart device includes at least one of smart glasses, AI glasses, headphones, and true wireless earbuds. The head-mounted smart device may include a wireless communication module such as a Wi-Fi module to enable wireless communication with a smartwatch, as shown in the reference. Figure 2 As shown.

[0026] While the inertial sensor 110 in the aforementioned smartwatch can track the watch's movement and posture, it may struggle to accurately detect the smartwatch's relative position to different parts of the user's body, nor can it easily determine the watch's absolute position. Therefore, in some scenarios, it may falsely detect wrist-raising or watch-looking actions. To further detect wrist-raising or watch-looking actions based on the channel parameters of the wireless communication connection between the smartwatch and the head-mounted smart device, the probability of false detection can be significantly reduced, thereby reducing erroneous screen-on operations and lowering the smartwatch's power consumption.

[0027] In some embodiments, when the gesture recognition module 120 outputs a first recognition result including whether the user's wrist is raised or not, it can also output the confidence level of the first recognition result. The confidence level can include the confidence level of the user's wrist being raised; the higher the confidence level, the more reliable the corresponding result (e.g., the user's wrist is raised). For example, the higher the confidence level of a certain wrist-raising action, the higher the probability that the CSI information detection module 130 will detect the wrist-raising action or the action of looking at a watch, and the lower the probability that the CSI information detection module 130 will detect the user's wrist not being raised. In this case, when the first recognition result indicates that the user's wrist is raised, and the corresponding confidence level is greater than or equal to a first threshold, the gesture recognition module 120 can directly determine that the user's wrist is raised. At this point, the probability of the user raising their wrist to look at a watch is relatively higher, and no further detection is needed, thus avoiding subsequent recognition and simplifying the recognition process.

[0028] Specifically, the gesture recognition module 120 is further configured to obtain the confidence level corresponding to the user's wrist being raised when the first recognition result indicates that the user's wrist is raised. If the confidence level is greater than or equal to a first threshold, it indicates that the gesture recognition module 120 can directly determine that the user's wrist is raised, and the probability of the user raising their wrist to look at the watch is high. In this case, the screen of the smartwatch is directly turned on to avoid the CSI information detection module 130 from further performing the judgment work on whether the user has raised their wrist to look at the watch, which can improve the judgment efficiency, thereby improving the screen lighting efficiency of the smartwatch and saving the power consumption required for the CSI information detection module 130 and other modules to participate in the judgment work. If the confidence level is less than a second threshold, it is determined that the user's wrist is not raised. In this case, there is no need to wake up the wireless communication module 140 of the smartwatch, no need for the CSI information detection module 130 to further perform the judgment work on whether the user has raised their wrist to look at the watch, and no need to turn on the screen of the smartwatch. This allows the wireless communication module 140 and / or the CSI information detection module 130 of the smartwatch to be in a sleep state or a low-power state, which can further save the power consumption of the smartwatch.

[0029] In some examples, reference Figure 3 As shown, the smartwatch also includes a first wireless communication module 140; the head-mounted smart device includes a second wireless communication module 210.

[0030] The preset conditions include the condition that the first recognition result indicates that the user's wrist is raised, and the confidence level is less than a first threshold but greater than or equal to a second threshold. The gesture recognition module 120 is further configured to determine that the first recognition result meets the preset conditions when the first recognition result indicates that the user's wrist is raised, and the confidence level is less than the first threshold but greater than or equal to the second threshold. In this case, the gesture recognition module 120 cannot directly determine that the user's wrist is raised, so it determines that the first recognition result meets the preset conditions, sends the first recognition result to the CSI information detection module 130, and wakes up the wireless communication module 140 to establish a wireless communication connection between the first wireless communication module 140 and the second wireless communication module 210, enabling the first wireless communication module 140 and the second wireless communication module 210 to communicate wirelessly.

[0031] The CSI information detection module 130 is used to receive the first identification result, detect the communication parameters between the first wireless communication module and the second wireless communication module to obtain the channel parameters between the first wireless communication module 140 and the second wireless communication module 210, further determine the second identification result representing whether the user raises their wrist to view the smartwatch based on the channel parameters, and light up the smartwatch screen when the second identification result represents that the user raises their wrist to view the smartwatch, so as to avoid accidentally lighting up the watch screen and ensure the reliability of the lighting operation.

[0032] Optionally, the first wireless communication module 140 may include a Wi-Fi module, and the second wireless communication module 210 may also include a Wi-Fi module. The channel parameters between the first wireless communication module 140 and the second wireless communication module 210 include CSI information (Channel State Information). The CSI information describes the complex frequency response H(f) on each subcarrier of the Wi-Fi OFDM: amplitude and phase, and can typically be given by subcarrier, spatial flow, and antenna dimension.

[0033] Specifically, the first threshold and the second threshold can be determined by multiple tests, for example, the first threshold can be set to 0.98 or 0.99, and the second threshold can be set to 0.87 or 0.90.

[0034] In some examples, the gesture recognition module 120 is also used to acquire the first recognition result at a first time T1; the CSI information detection module 130 is used to acquire the channel parameters of the smartwatch and the corresponding user's head-mounted smart device when the first recognition result meets the preset conditions, that is, when the gesture recognition module 120 detects that the user's wrist is suspected to be raised at the first time T1, and further determine the second recognition result representing whether the user has raised their wrist to check the smartwatch based on the channel parameters at a second time T2, so that the smartwatch can wake up the first wireless communication module 140 during the interval between the first time T1 and the second time T2, so that the first wireless communication module 140 can communicate wirelessly with the second wireless communication module 210, thereby enabling the CSI information detection module 130 to acquire more stable and reliable channel parameters at the second time T2, which can shorten the screen lighting delay while improving the reliability of the final screen lighting operation.

[0035] Specifically, the gesture recognition module 120 recognizes a hand-raising motion 1 with relatively low confidence at the first time T1, and the CSI information detection module 130 obtains channel parameters such as CSI information at the second time T2. There is often a time interval (e.g., tens of milliseconds or 10-20 milliseconds) between the first time T1 and the second time T2. This time interval can include the Wi-Fi connection establishment time between the smartwatch and the head-mounted smart device, the Wi-Fi frame transmission time between the smartwatch and the head-mounted smart device, or the time to wake up the first wireless communication module 140 of the smartwatch. The Wi-Fi frame transmission time can include multiple Wi-Fi frame transmission times, thus providing multiple frames of CSI information and other channel parameters as input, which improves the reliability of the channel parameters used. By performing preliminary gesture recognition at the first time T1 to obtain channel parameters such as CSI information, and determining whether to perform a screen-on operation based on the gesture recognition result and CSI information at the second time T2, the latency of the screen-on operation can be reduced.

[0036] Optionally, while acquiring signal parameters at the second time T2, the CSI information detection module 130 can also acquire motion parameters detected by the inertial sensor 110 up to the second time T2, so as to further detect whether the user raises his wrist to look at the watch based on the signal parameters during the period between the first time T1 and the second time T2 and the motion parameters up to the second time T2, thereby further improving the reliability of the results.

[0037] In some embodiments, the gesture recognition module 120 is preset with a first AI model; the first AI model is used to output a first recognition result representing whether the user's wrist is raised and the confidence level of the first recognition result based on the motion parameters.

[0038] Specifically, the process of determining the first AI model includes: acquiring multiple first input samples and the first labels corresponding to each first input sample; inputting multiple sets of first input samples and their corresponding first labels into a first initial model for training to obtain the first AI model. The first initial model includes a neural network model, which can be a 2D CNN / ResNet / ConvNeXt model or a Transformer model.

[0039] Optionally, the first input sample includes motion parameters detected by the inertial sensor 110 when the user wears the smart device in multiple specific application scenarios; the first label includes annotation information when the inertial sensor 110 detects the corresponding motion parameters. This annotation information may include labels such as wrist raised and wrist not raised, and may also include confidence levels for wrist raised and / or wrist not raised, as well as other target values ​​used to determine whether the user's wrist has been raised. Optionally, further, when the first label corresponds to the action of the user's wrist not being raised, the action can be combined with the user's simultaneous judgment that they did not raise their wrist to look at the watch, to ensure the reliability of the determined first label. Optionally, the sources of the first input samples cover the following scenarios: running, brisk walking, large arm swings, rope skipping, swinging a racket / bat, throwing actions, clapping, waving, greeting; turning the steering wheel, turning on the lights, shifting gears while driving; riding, skateboarding, electric bike handlebars; typing, using a mouse, writing, turning pages; looking at the screen / taking a selfie with a mobile phone, making and receiving phone calls; drinking water / eating by raising a cup or chopsticks; washing face, brushing teeth, rinsing mouth, combing hair, hair dryer; cooking, stir-frying, wiping windows, mopping, sweeping; lifting dumbbells (bice-curl / press), bench press, push-ups, rowing machine; tennis, badminton, golf swings; shaking hands, saluting, raising hands to signal; turning over at night, stretching, raising arms to adjust the blanket, etc., so that the scenarios represented by the first input samples and the corresponding first labels are more comprehensive, and the trained first AI model can more accurately output the first recognition result and confidence level corresponding to the motion parameters.

[0040] In some embodiments, the CSI information detection module is pre-set with a second AI model.

[0041] Optionally, the second AI model can output a second recognition result representing whether the user raises their wrist to view the smartwatch based on the channel parameters. Accordingly, the process of determining the second AI model includes: acquiring multiple second input samples and corresponding second labels for each second input sample; inputting multiple sets of second input samples and corresponding second labels into a second initial model for training to obtain the second AI model. Optionally, the second input samples include the channel parameters detected by the CSI information detection module 130 when the user wears the head-mounted smart device in multiple specific application scenarios; the second labels include the annotation information when the CSI information detection module 130 detects the channel parameters. This annotation information can include explicit annotations such as yes or no, for example, yes could indicate that the user raised their wrist to view the smartwatch, and no could indicate that the user did not raise their wrist to view the smartwatch, etc.

[0042] Optionally, the second AI model can output a second recognition result representing whether the user raised their wrist to check the smartwatch based on the motion parameters detected by the inertial sensor 110 and the channel parameters detected by the CSI information detection module 130. Accordingly, the process of determining the second AI model includes: acquiring multiple third input samples and corresponding third labels for each third input sample; inputting multiple sets of third input samples and corresponding third labels into the second initial model for training to obtain the second AI model. Optionally, the third input samples include the channel parameters detected by the CSI information detection module 130 and the motion parameters detected by the inertial sensor 110 when the user wears the head-mounted smart device in multiple specific application scenarios; the third labels include the annotation information of the smartwatch when detecting each set of channel parameters and motion parameters. This annotation information can include explicit annotations such as yes or no, for example, yes can indicate that the user raised their wrist to check the smartwatch, and no can indicate that the user did not raise their wrist to check the smartwatch, etc.

[0043] The second initial model includes a neural network model, which can be a 2D CNN / ResNet / ConvNeXt model or a Transformer model.

[0044] Optionally, the second input sample includes channel parameters detected by the CSI information detection module 130 when the user wears the smartwatch in multiple specific application scenarios. The second label includes whether the user raises their wrist to view the smartwatch's annotation information when the CSI information detection module 130 detects the channel parameters. The third input sample includes motion parameters detected by the inertial sensor 110 and channel parameters detected by the CSI information detection module 130 when the user wears the smartwatch in multiple specific application scenarios. The third label includes whether the user raises their wrist to view the smartwatch's annotation information when the smartwatch acquires the motion parameters and channel parameters.

[0045] Optionally, the sources of the second and third input samples cover various life scenarios, sports scenarios, leisure scenarios, and work scenarios of different users. For example, they need to cover: running, brisk walking, large arm swings, rope skipping, swinging a racket / bat, throwing actions, clapping, waving, and greeting; turning the steering wheel, turning the lights, and shifting gears while driving; riding, skateboarding, and electric bike handlebars; typing, using a mouse, writing, and turning pages; looking at the screen / taking a selfie with a mobile phone, and making and receiving phone calls; drinking water / eating by raising a cup and chopsticks; washing face, brushing teeth, rinsing mouth, combing hair, and using a hair dryer; cooking, stir-frying, and wiping windows. Mopping and sweeping; lifting dumbbells (bicep curls / presses), bench presses, push-ups, rowing machines; tennis, badminton, and golf swings; shaking hands, saluting, and raising hands to signal; turning over at night, stretching, and raising arms to adjust the blankets, etc., to make the second input sample and the corresponding second label represent more comprehensive scenarios, so that the trained second AI model can more accurately output the wrist movements corresponding to the channel parameters; to make the third input sample and the corresponding third label represent more comprehensive scenarios, so that the trained second AI model can more accurately output the wrist movements corresponding to the motion parameters and channel parameters.

[0046] In the above-mentioned smartwatch, considering that although the inertial sensor 110 can track the movement and posture of the watch, it may be difficult to accurately detect the relative position of the smartwatch with respect to various parts of the user's body, and it is also difficult to obtain the absolute position of the watch itself, the inertial sensor 110 detects the motion parameters corresponding to the wrist wearing the smartwatch. The gesture recognition module 120 identifies a first recognition result indicating whether the user's wrist is raised based on the motion parameters. The CSI information detection module 130 is used to obtain the channel parameters between the smartwatch and the corresponding user's head-mounted smart device when the first recognition result meets the preset conditions, that is, when the user's wrist is suspected to be raised. Based on the channel parameters, a second recognition result indicating whether the user has raised their wrist to look at the smartwatch is further determined. When the second recognition result indicates that the user has raised their wrist to look at the smartwatch, the screen of the smartwatch is turned on. This is to further confirm through the CSI information detection module 130 that the user has raised their wrist to look at the smartwatch before turning on the screen, which can avoid the smartwatch screen being ineffectively lit, thereby saving the power consumption of the smartwatch. It is evident that the smartwatch provided in this application can further detect wrist movements such as raising the wrist or looking at a watch based on the channel parameters of the wireless communication connection between the smartwatch and the head-mounted smart device. This can greatly reduce the probability of false detection, thereby reducing erroneous screen-on operations on the smartwatch and reducing the power consumption of the smartwatch.

[0047] A second aspect of this application provides a method for controlling the screen brightness of a smartwatch, which is applied to the smartwatch described in any of the above embodiments. (See reference...) Figure 4 As shown, the above-mentioned smartwatch screen-on control method includes steps S310 to S330.

[0048] S310, detect and characterize the motion parameters corresponding to the wrist wearing the smartwatch.

[0049] S320, based on the motion parameters, identify a first recognition result indicating whether the user's wrist has been raised.

[0050] S330, when the first recognition result meets the preset conditions, obtain the channel parameters of the smartwatch and the corresponding user's head-mounted smart device, further determine the second recognition result indicating whether the user raises their wrist to view the smartwatch based on the channel parameters, and when the second recognition result indicates that the user raises their wrist to view the smartwatch, turn on the screen of the smartwatch.

[0051] In some embodiments, identifying a first recognition result characterizing whether the user's wrist has been raised based on the motion parameters includes: When the first recognition result indicates that the user's wrist is raised, the confidence level corresponding to the user's wrist being raised is obtained. If the confidence level is greater than or equal to the first threshold, the screen of the smartwatch is directly turned on. If the confidence level is less than the second threshold, it is determined that the user's wrist has not been raised. If the confidence level is less than the first threshold and greater than or equal to the second threshold, then the first identification result is determined to meet the preset conditions.

[0052] The confidence level mentioned above can include the confidence level of the user raising their wrist. The higher the confidence level, the more reliable the corresponding result (such as the user raising their wrist). The first threshold can be set to 0.98 or 0.99, and the second threshold can be set to 0.87 or 0.90, etc.

[0053] The above-described smartwatch screen-on control method, when applied to any of the smartwatches described in the above embodiments, has all the beneficial effects of the smartwatches described in the aforementioned embodiments, and will not be repeated here.

[0054] Although this application has been shown and described with respect to one or more implementations, equivalent variations and modifications will occur to those skilled in the art based on a reading and understanding of this specification and the accompanying drawings. This application includes all such modifications and variations and is limited only by the scope of the appended claims. In particular, with respect to the various functions performed by the aforementioned components, the terminology used to describe such components is intended to correspond to any component (unless otherwise indicated) that performs the specified function of said component (e.g., is functionally equivalent to it), even if structurally not equivalent to the disclosed structure performing the functions in the exemplary implementations of this specification shown herein.

[0055] That is, the above description is only an embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural changes made using the content of this application’s specification and drawings, such as the combination of technical features between different embodiments, or direct or indirect application in other related technical fields, are similarly included within the patent protection scope of this application.

[0056] Furthermore, it should be understood that in the description of this application, the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Additionally, for structural elements with the same or similar characteristics, this application may use the same or different reference numerals for identification. Moreover, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0057] In this application, the term "exemplary" is used to mean "serving as an example, illustration, or description." Any embodiment described as "exemplary" in this application is not necessarily to be construed as more preferred or advantageous than other embodiments. This application has been provided above to enable any person skilled in the art to implement and use it. Various details have been set forth in the above description for purposes of explanation. It should be understood that those skilled in the art will recognize that this application can be implemented without using these specific details. In other embodiments, well-known structures and processes will not be described in detail to avoid obscuring the description of this application with unnecessary detail. Therefore, this application is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

Claims

1. A smartwatch, characterized in that, The smartwatch includes an inertial sensor, a gesture recognition module, and a CSI information detection module; The inertial sensor is used to detect motion parameters corresponding to the wrist of the person wearing the smartwatch; The gesture recognition module is used to identify a first recognition result that indicates whether the user's wrist has been raised based on the motion parameters; The CSI information detection module is used to obtain the channel parameters of the smartwatch and the corresponding user's head-mounted smart device when the first recognition result meets the preset conditions, further determine the second recognition result indicating whether the user raises their wrist to view the smartwatch based on the channel parameters, and light up the screen of the smartwatch when the second recognition result indicates that the user raises their wrist to view the smartwatch.

2. The smartwatch according to claim 1, characterized in that, The gesture recognition module is further configured to obtain the confidence level corresponding to the user's wrist being raised when the first recognition result indicates that the user's wrist is raised; if the confidence level is greater than or equal to a first threshold, the screen of the smartwatch is directly turned on; if the confidence level is less than a second threshold, it is determined that the user's wrist is not raised.

3. The smartwatch according to claim 2, characterized in that, The smartwatch also includes a first wireless communication module; the head-mounted smart device includes a second wireless communication module. The gesture recognition module is further configured to determine that the first recognition result meets the preset conditions if the confidence level is less than the first threshold and greater than or equal to the second threshold, send the first recognition result to the CSI information detection module, and wake up the wireless communication module to establish a wireless communication connection between the first wireless communication module and the second wireless communication module. The CSI information detection module is used to receive the first identification result and detect the communication parameters between the first wireless communication module and the second wireless communication module to obtain the channel parameters.

4. The smartwatch according to claim 2, characterized in that, The gesture recognition module is also used to acquire the first recognition result at a first moment; The CSI information detection module is used to obtain the channel parameters of the smartwatch and the corresponding user's head-mounted smart device when the first recognition result meets the preset conditions, and further determine the second recognition result representing whether the user raises their wrist to view the smartwatch based on the channel parameters at the second moment.

5. The smartwatch according to claim 1, characterized in that, The gesture recognition module is pre-set with a first AI model; the first AI model is used to output a first recognition result representing whether the user's wrist is raised and the confidence level of the first recognition result based on the motion parameters.

6. The smartwatch according to claim 5, characterized in that, The process of determining the first AI model includes: Multiple first input samples and their corresponding first labels are obtained. Multiple sets of first input samples and their corresponding first labels are input into a first initial model for training to obtain the first AI model.

7. The smartwatch according to claim 1, characterized in that, The CSI information detection module is pre-set with a second AI model; the second AI model is used to output a second recognition result representing whether the user raises their wrist to view the smartwatch based on the channel parameters.

8. The smartwatch according to claim 7, characterized in that, The process of determining the second AI model includes: Multiple second input samples and their corresponding second labels are obtained. Multiple sets of second input samples and their corresponding second labels are input into a second initial model for training to obtain the second AI model.

9. A method for controlling the screen brightness of a smartwatch, characterized in that, The screen-on control method for the smartwatch, applied to the smartwatch according to any one of claims 1 to 8, includes the following steps: The detection characterizes the motion parameters corresponding to the wrist wearing the smartwatch; The first recognition result, which indicates whether the user's wrist has been raised, is identified based on the motion parameters. When the first recognition result meets the preset conditions, the channel parameters of the smartwatch and the corresponding user's head-mounted smart device are obtained. Based on the channel parameters, a second recognition result indicating whether the user raises their wrist to view the smartwatch is further determined. When the second recognition result indicates that the user raises their wrist to view the smartwatch, the screen of the smartwatch is turned on.

10. The screen-on control method for a smartwatch according to claim 9, characterized in that, The first recognition result, which identifies whether the user's wrist has been raised based on the motion parameters, includes: When the first recognition result indicates that the user's wrist is raised, the confidence level corresponding to the user's wrist being raised is obtained. If the confidence level is greater than or equal to the first threshold, the screen of the smartwatch is directly turned on. If the confidence level is less than the second threshold, it is determined that the user's wrist has not been raised. If the confidence level is less than the first threshold and greater than or equal to the second threshold, then the first identification result is determined to meet the preset conditions.

Images