Object movement detection apparatus, method and non-transitory computer readable storage medium

Abstract

An object movement detection apparatus, method, and non-transitory computer readable storage medium are provided. The apparatus calculates a detected reference pose corresponding to the target object and the reference object when the target object is in a detecting state. In response to a difference between the detected reference pose and an initial reference pose being greater than a threshold, the apparatus determines that the target object is held.

Description

Technical Field

[0001] This disclosure relates to an object motion detection device, method, and non-transitory computer-readable storage medium thereof, and particularly to an object motion detection device, method, and non-transitory computer-readable storage medium based on relative attitude. Prior Technology

[0002] In existing technologies, due to limitations in hardware specifications, when only three degrees of freedom (3DoF) data of an object can be obtained, additional sensors need to be installed on the object to determine whether the object has been picked up and used by the user.

[0003] Furthermore, if subjected to external interference or used in unintended situations, the additional sensors may produce false judgments.

[0004] In view of this, providing reliable object motion detection technology that does not require additional sensors is a goal that the industry urgently needs to strive for. Summary of the Invention

[0005] To address the aforementioned problems, this disclosure proposes a dynamic object detection method applicable to an electronic device. The method includes the following steps: calculating a detection relative posture between a detection target posture of a target object and a detection reference posture of a reference object, wherein the detection target posture and the detection reference posture are measured when the target object is in a detection state; and determining that the target object has been picked up and setting the target object to a picked-up state in response to a first difference between the detection relative posture and an initial relative posture being greater than a first threshold.

[0006] This disclosure also provides an object motion detection device, including a communication interface and a processor. The communication interface is used to communicate with a target object and a reference object. The processor is electrically connected to the communication interface and is used to perform the following operations: calculating a detection relative attitude between a detection target attitude of the target object and a detection reference attitude of the reference object, wherein the detection target attitude and the detection reference attitude are measured when the target object is in a detection state; and determining that the target object has been picked up and setting the target object to a picked-up state in response to a first difference between the detection relative attitude and an initial relative attitude being greater than a first threshold.

[0007] This disclosure also provides a non-transitory computer-readable storage medium having at least one set of instructions stored thereon. When a processor executes the instructions, the instructions execute an object dynamic detection method. The object dynamic detection method includes the following steps: calculating a detection relative attitude between a detection target attitude of a target object and a detection reference attitude of a reference object, wherein the detection target attitude and the detection reference attitude are measured when the target object is in a detection state; and determining that the target object has been picked up and setting the target object to a picked-up state in response to a first difference between the detection relative attitude and an initial relative attitude being greater than a first threshold.

[0008] It should be understood that the foregoing general description and the following specific description are merely exemplary and explanatory, and are intended to provide further explanation of the claimed disclosure. Simple Explanation of the Diagram

[0009] To make the above and other objects, features, advantages and embodiments of this disclosure more apparent and understandable, the accompanying drawings are described below: Figure 1 is a schematic diagram of the object dynamic detection device in the first embodiment of this disclosure; Figure 2A shows a usage diagram of a user wearing the target object and a reference object in some embodiments of this disclosure; Figure 2B is a diagram illustrating the user's handling of the target object in one of the embodiments disclosed herein. Figure 3A shows a usage diagram of a user wearing the target object and the reference object in another embodiment of this disclosure; Figure 3B is a diagram illustrating the user's action of picking up a target object in another embodiment of this disclosure; Figure 4 is a schematic diagram illustrating how the object motion detection device determines whether a target object has been taken in some embodiments of this disclosure; Figure 5 is a schematic diagram showing the object dynamic detection device setting the target object to its initial state in one of the embodiments disclosed herein; Figure 6 is a schematic diagram of the object dynamic detection device in some embodiments of this disclosure determining whether the target object has been put back after it has been taken; Figure 7 is a schematic diagram of the inertial signal when the target object falls in some embodiments of this disclosure; Figure 8 is a schematic diagram of the object dynamic detection device in one of the embodiments disclosed herein setting the target object to a falling state; Figure 9 is a schematic diagram illustrating how the object motion detection device determines whether a target object has been picked up or dropped in one of the embodiments disclosed herein; Figure 10 is a schematic diagram illustrating how the object dynamic detection device in some embodiments of this disclosure determines whether the target object has been returned after it has fallen. Figure 11 is a schematic diagram illustrating how the object motion detection device in some embodiments of this disclosure determines whether the target object has been picked up after it has fallen; and Figure 12 is a flowchart of the object dynamic detection method in the second embodiment of this disclosure. Implementation

[0010] To make the description of this disclosure more detailed and complete, reference can be made to the accompanying drawings and the various embodiments described below, in which the same numbers represent the same or similar elements.

[0011] Please refer to Figure 1, which is a schematic diagram of the object motion detection device 1 in the first embodiment of this disclosure. The object motion detection device 1 includes a processor 12 and a communication interface 14, wherein the processor 12 is electrically connected to the communication interface 14. The object motion detection device 1 is used to determine whether a target object has been picked up by a user. In some embodiments, the object motion detection device 1 may be installed in a head-mounted display (HMD) or other device to track target objects.

[0012] In some embodiments, processor 12 may include a central processing unit (CPU), a graphics processing unit (GPU), a multiprocessor, a distributed processing system, an application-specific integrated circuit (ASIC), and / or a suitable arithmetic unit.

[0013] The communication interface 14 is used to communicate between a target object and a reference object, wherein the target object is the target to be detected by the object dynamic detection device 1, and the reference object is a reference object for relative attitude.

[0014] In some embodiments, the communication interface 14 may include Ethernet, Bluetooth, Wi-Fi and / or other wired or wireless data transceiver interfaces, and transmit and receive data to the target object and the reference object through the corresponding communication protocol.

[0015] In some embodiments, the target object and the reference object each include sensors for tracking attitude, such as image capturing elements for image tracking and / or inertial measurement units (IMUs) for providing inertial data.

[0016] The target object and the reference object provide attitude data to the object motion detection device 1 via the communication interface 14. For example, the target object and the reference object continuously provide inertial data measured by the inertial measurement unit to the communication interface 14 via streaming. Correspondingly, the object motion detection device 1 can calculate the three-degree-of-freedom attitude of the target object and the reference object based on the inertial data.

[0017] In some embodiments, the attitude of the reference object can also be measured by other components. For example, the reference object is a non-electronic component located near the target object, and after the reference object is observed by an external camera, an image of the reference object or the attitude of the reference object calculated based on the image is transmitted to the object motion detection device 1. Correspondingly, the object motion detection device 1 can calculate the attitude of the reference object based on the image or directly obtain the attitude.

[0018] For the usage scenarios of the object motion detection device 1, please refer to Figures 2A and 2B. Figure 2A shows the scenario where the user U wears the target object T and the reference object R, while Figure 2B shows the scenario where the user U picks up the target object T.

[0019] In the embodiments illustrated in Figures 2A and 2B, the target object T is a pistol, and the reference object R is a holster worn on the user U's thigh, with the pistol placed inside the holster (as shown in Figure 2A). When the user U does not remove the pistol, the target object T and the reference object R will move and rotate synchronously because they are fixed together (as shown in Figure 2A), thus maintaining similar relative postures. Conversely, when the user U removes the pistol from the holster, the target object T will undergo different posture changes compared to the reference object R due to the user U's operation (as shown in Figure 2B).

[0020] For another use case of the object motion detection device 1, please refer to Figures 3A and 3B. Figure 3A shows the scenario where the user U wears the target object T and the reference object R, while Figure 3B shows the scenario where the user U picks up the target object T.

[0021] Similarly, in the embodiments illustrated in Figures 3A and 3B, the target object T is a pistol, and the reference object R is a belt worn on the user U's waist, with the pistol placed in a holster on the belt (as shown in Figure 3A). When the user U does not take the pistol, the target object T and the reference object R will move and rotate synchronously because they are fixed together (as shown in Figure 3A), thus maintaining similar relative postures. Conversely, when the user U takes the pistol out of the holster, the target object T will undergo different posture changes compared to the reference object R due to the user U's operation (as shown in Figure 3B).

[0022] In this embodiment, the sensor of the reference object R is located at the position of the belt buckle in front of the user U's abdomen (i.e., the position indicated by label R in Figure 3A). As can be seen from Figure 3A, when the pistol is placed in its holster, the posture measured by the sensor in the target object T and the sensor in the reference object R will change synchronously with the user U's movements. Conversely, when the user U picks up the pistol, the posture measured by the sensor in the target object T will change differently from the posture measured by the sensor in the reference object R.

[0023] Based on the posture change characteristics between the target object T and the reference object R in the aforementioned embodiments, the object dynamic detection device 1 can determine whether the target object T has been taken by (user U or other person) based on the relative posture between the target object T and the reference object R.

[0024] For example, if the relative posture between the target object T and the reference object R changes only slightly, the object dynamic detection device 1 determines that the target object T has not been picked up. Conversely, if the relative posture changes significantly, the object dynamic detection device 1 determines that the target object T is being picked up.

[0025] Specifically, the processor 12 performs the following operations: calculating a detection relative pose between a detection target pose of the target object and a detection reference pose of the reference object, wherein the detection target pose and the detection reference pose are measured when the target object is in a detection state; and in response to a first difference between the detection relative pose and an initial relative pose being greater than a first threshold, determining that the target object has been picked up and setting the target object to a picked-up state.

[0026] In this way, by comparing the relative posture of the target object T when it was placed in its original position with the relative posture during detection, the object dynamic detection device 1 can detect whether the target object T has been taken.

[0027] In some embodiments, in order to obtain an initial relative posture, the object dynamic detection device 1 calculates the initial relative posture of the target object T and the reference object R as a reference for subsequent judgment when the target object T has not yet been picked up (e.g., the situation illustrated in Figures 2A and / or 3A). Further, after calculating the initial relative posture, the object dynamic detection device 1 sets the target object T to a detection state to begin detection.

[0028] Specifically, the processor 12 performs the following operations: calculating the initial relative pose between an initial target pose of the target object and an initial reference pose of the reference object, wherein the initial target pose and the initial reference pose are measured when the target object is in an initial state; and after calculating the initial relative pose, setting the target object to a detection state.

[0029] After calculating the initial relative attitude, the object dynamic detection device 1 begins to detect whether the target object T has been picked up, continuously calculates the relative attitude between the target object T and the reference object R, and judges the difference between the calculated relative attitude and the initial relative attitude. If the difference is too large, the object dynamic detection device 1 determines that the target object T has been picked up.

[0030] To illustrate the above operation in more detail, please refer to Figure 4, which is a schematic diagram of the object dynamic detection device 1 determining whether the target object has been taken in some embodiments of this disclosure.

[0031] First, the object dynamic detection device 1 performs operation P101 to calculate the initial relative posture of the target object T and the reference object R. At this time, the target object T is placed in the initial position, that is, the state before it has been picked up by the user U.

[0032] Next, the object motion detection device 1 performs operation P103, setting the target object T to the detection state. At this time, the object motion detection device 1 begins to determine whether the target object T has been taken.

[0033] Next, the object dynamic detection device 1 performs operation P105 to calculate the relative orientation of the target object T and the reference object R.

[0034] Correspondingly, the object dynamic detection device 1 further executes operation P107 to determine whether the difference between the detected relative posture and the initial relative posture is too large. When the difference is greater than a first threshold, the object dynamic detection device 1 executes operation P109 to determine that the target object T has been picked up and sets the target object T to a picked-up state. Conversely, when the difference is not greater than the first threshold, the object dynamic detection device 1 returns to operation P105 and continues to calculate and detect the relative posture.

[0035] Based on the above embodiments, the object motion detection device 1 can determine whether the target object T has been picked up. In practical applications, the object motion detection device 1 can be applied to interactive devices, such as game objects; or it can be applied to objects that can be manipulated by the user, such as guns or sports equipment.

[0036] It should be noted that the relative attitudes (e.g., initial relative attitude and detected relative attitude) between the target object T and the reference object R can be represented by quaternions and / or matrices, indicating the position and / or orientation relationship between them. In some embodiments, the object dynamic detection device 1 calculates the relative attitude based on the three-degree-of-freedom attitudes of the target object T and the reference object R respectively.

[0037] On the other hand, the object dynamic detection device 1 can subtract two relative postures to obtain the distance and / or angle difference between them, and use this difference as the difference between the two relative postures.

[0038] In some embodiments, before setting the target object T to the initial state, the object dynamic detection device 1 first determines whether the relative attitude of the target object T and the reference object R remains stable and is sufficient to calculate a reliable initial relative attitude.

[0039] Specifically, the processor 12 is further configured to perform the following operations: calculate a plurality of first relative poses between a plurality of first target poses of the target object and a plurality of first reference poses of the reference object, wherein the first target poses and the first reference poses are measured in a time interval; and set the target object to the initial state in response to a second difference between the first relative poses being less than a second threshold.

[0040] For details on how the object dynamic detection device 1 sets the target object T to its initial state, please refer to Figure 5.

[0041] First, in operation P201, the object dynamic detection device 1 begins to track the poses of the target object T and the reference object R, and continuously calculates the relative poses of the two (e.g., calculates a set of relative poses every 10 milliseconds).

[0042] Next, in operation P203, the object dynamics detection device 1 determines whether the relative attitudes have stabilized. For example, the object dynamics detection device 1 calculates the relative attitude changes between multiple time points. When the relative attitude difference within a certain time interval (e.g., 0.5 seconds) is less than a second threshold (e.g., an angle difference of less than 5 degrees), the object dynamics detection device 1 determines that the relative attitude between the target object T and the reference object R has reached a stable state. Accordingly, the object dynamics detection device 1 executes operation P205, setting the target object T to its initial state to calculate the initial relative attitude. Conversely, if the relative attitudes have not reached a stable state, the object dynamics detection device 1 returns to operation P201 to continue calculating the relative attitude.

[0043] According to the above embodiment, the object motion detection device 1 can determine whether to set the target object T to the initial state. After setting the target object T to the initial state, as shown in Figure 5, the object motion detection device 1 enters node A and continues to execute the operation P101 shown in Figure 4.

[0044] In some embodiments, after determining that the target object T has been taken, the object dynamic detection device 1 begins to detect whether the target object T is continuously taken or returns to the placed state (e.g., the situation illustrated in Figures 2A and / or 3A).

[0045] Specifically, the processor 12 is further configured to perform the following operations: calculate a post-taken target pose of the target object and a post-taken reference pose of the reference object, wherein the post-taken target pose and the post-taken reference pose are measured when the target object is in the taken state; and set the target object to the detection state in response to a third difference between the post-taken relative pose and the initial relative pose being less than a third threshold.

[0046] For details on the operation of the object dynamic detection device 1 in determining whether the target object T has been returned after it has been taken, please refer to Figure 6.

[0047] First, in operation P301, after the target object T is picked up, the object dynamic detection device 1 continuously tracks the posture of the target object T and the reference object R, and continuously calculates the relative posture of the two objects in the picking up process.

[0048] Next, in operation P303, the object motion detection device 1 determines whether the relative posture of the object being picked up is similar to the initial relative posture. For example, the object motion detection device 1 calculates the difference between the relative posture of the object being picked up and the initial relative posture. When the difference is less than a third threshold, it indicates that the user U may have already put the target object T back in its original position (e.g., putting a gun back in its holster). Therefore, the object motion detection device 1 determines that the target object T has stopped being picked up by the user U and further executes operation P305. Conversely, when the difference is not less than the third threshold, it indicates that the user U may still be picking up the target object T. Therefore, the object motion detection device 1 returns to operation P301 to continue calculating the relative posture of the object being picked up.

[0049] Correspondingly, in operation P305, the object motion detection device 1 returns the target object T to the detection state. Further, as shown in Figure 6, the object motion detection device 1 enters node B and continues to execute operation P105 as illustrated in Figure 4.

[0050] In some embodiments, if the target object T is accidentally dropped, the object dynamic detection device 1 may mistakenly determine that the target object T has been taken. Please further refer to Figure 7, which is a schematic diagram of the inertial signal when the target object T falls in some embodiments of this disclosure. As shown in the figure, the inertial data DP of the target object T changes significantly during the fall. In this embodiment, the inertial data includes angular velocity (i.e., right vertical axis GYRO) and acceleration (i.e., left vertical axis ACC) data on the x, y, and z axes, respectively. Accordingly, the object dynamic detection device 1 can determine whether the target object T has fallen based on the inertial data DP of the target object T.

[0051] For details on how to determine whether the target object T has fallen, please refer to Figure 8.

[0052] First, during operation of P501, the object dynamic detection device 1 continuously acquires the inertial data of the target object T, for example, the communication interface 14 receives the inertial data measured by the inertial measurement unit from the target object T.

[0053] Next, during the operation of P503, the object dynamic detection device 1 determines whether the target object T has fallen based on inertial data.

[0054] The object dynamic detection device 1 can determine whether the target object T has fallen in a variety of different ways. For example, since the inertial data will change significantly when the target object T falls, the object dynamic detection device 1 determines that the target object T has fallen when the value of the inertial data (e.g., the absolute value of angular velocity and / or acceleration on one or more axes) is greater than a certain value.

[0055] Specifically, the processor 12 is further configured to perform the following operations: acquire inertial data measured by an inertial measurement element of the target object; and, in response to the inertial data being greater than an inertial threshold, set the target object to a falling state.

[0056] In another example, since the inertial data of the target object T will have specific characteristics when it falls, the object dynamic detection device 1 can also determine whether the target object T has fallen through a machine learning model.

[0057] Specifically, the processor 12 is further configured to perform the following operations: acquire inertial data measured by an inertial measurement element of the target object; and, based on the inertial data, determine whether to set the target object to a falling state using a detection model.

[0058] It should be noted that the detection model can be generated by using multiple inertial data points measured when the target object T or other objects fall as training data to train the machine learning model.

[0059] Please return to Figure 8. After the object motion detection device 1 determines that the target object T has fallen in operation P503, it further executes operation P505 to set the target object T to a falling state. Conversely, if the object motion detection device 1 determines that the target object T has not fallen, it returns to operation P501 to continue determining whether the target object T has fallen.

[0060] The object motion detection device 1 can determine whether the target object T has fallen at multiple different time points. In some embodiments, since the relative posture of the target object T when it falls will differ from its initial relative posture, when it is determined that the target object T may be picked up (e.g., the operation P107 shown in Figure 4 determines "yes"), the object motion detection device 1 will first determine whether the target object T has fallen, and then decide whether to set the target object T to a picked-up state or a fallen state.

[0061] Please refer to Figure 9, and in conjunction with the two embodiments in Figures 4 and 8, after executing operations P105 and P107, if the object dynamic detection device 1 determines that the difference between the detected relative posture and the initial relative posture is greater than a threshold, then it executes operation P503. Further, when the object dynamic detection device 1 determines that the target object T meets the characteristics of falling, it executes operation P505 to set the target object T to a falling state. Conversely, when the object dynamic detection device 1 determines that the target object T does not meet the characteristics of falling, it executes operation P109 to determine that the target object T has been picked up and sets it to a picked-up state.

[0062] Specifically, the processor 12's operation of determining that the target object has been taken further includes: in response to the first difference being greater than the first threshold, determining whether the target object is in a falling state based on an inertial data of the target object; and in response to the target object not being in the falling state, determining that the target object has been taken and setting the target object to the taken state.

[0063] It should be noted that the details of each operation in Figure 9 are similar to those in the aforementioned embodiments, so the similarities will not be repeated. In some examples, the object motion detection device 1 will first execute the operation P501 shown in Figure 8 before executing operation P503 to obtain the required inertial data.

[0064] In some embodiments, after the target object T is set to a falling state, the object dynamic detection device 1 further detects whether the target object T is returned to its original position.

[0065] Specifically, the processor 12 is further configured to perform the following operations: calculate a relative falling posture between a falling target posture of the target object and a falling reference posture of the reference object, wherein the falling target posture and the falling reference posture are measured when the target object is in a falling state; and set the target object to the detection state in response to a fourth difference between the relative falling posture and the initial relative posture being less than a fourth threshold.

[0066] For specific operational details, please refer to Figure 10. Following the embodiments illustrated in Figures 8 and / or 9, after the operation of the object dynamic detection device 1 enters node D, the object dynamic detection device 1 can also perform operation P601 to calculate the relative posture of the target object T after it falls (i.e., the relative posture of the fall).

[0067] Furthermore, the object dynamic detection device 1 performs operation P603 to determine whether the falling relative posture is similar to the initial relative posture. Operation P603 can be implemented based on operation P303 shown in Figure 6.

[0068] When the falling relative posture is similar to the initial relative posture, the object dynamic detection device 1 determines that the target object T has been returned to its original position and accordingly sets the target object T to the detection state in operation P605. Afterwards, the object dynamic detection device 1 can also enter node B to return to the operation shown in Figure 4 or 9. Conversely, when the falling relative posture is significantly different from the initial relative posture, the object dynamic detection device 1 determines that the target object T has not been returned to its original position and returns to operation P601 to continue detection.

[0069] In some embodiments, after the target object T is set to a falling state, the object dynamic detection device 1 further detects whether the target object T is picked up by the user U.

[0070] Specifically, the processor 12 is further configured to perform the following operations: in response to setting the target object to a falling state, acquiring inertial data measured by an inertial measurement element of the target object; based on the inertial data, determining whether the target object has been moved; and in response to determining that the target object has been moved, setting the target object to the picking state.

[0071] For specific operational details, please refer to Figure 11. Following the embodiments illustrated in Figures 8 and / or 9, after the object motion detection device 1 enters node D, the object motion detection device 1 can also perform operation P701 to obtain the inertial data of the target object T.

[0072] Furthermore, the object dynamic detection device 1 performs operation P703 to determine whether the target object T has been moved based on the numerical value of the inertial data (e.g., after the target object T falls, there is still a change in acceleration and / or angular velocity).

[0073] When it is determined that the target object T has been moved, it means that the target object T may have been picked up by the user U. Therefore, the object motion detection device 1 determines that the target object T has been picked up in operation P705 and sets it to a picked-up state. Afterwards, the object motion detection device 1 can also enter node C to return to the operation shown in Figure 6. Conversely, when it is determined that the target object T has not been moved, the object motion detection device 1 returns to operation P701 to continue detection.

[0074] Through the above embodiments, the object motion detection device 1 can determine whether the target object T is in an initial state, a picked-up state, or a dropped state. Accordingly, the object motion detection device 1 can also perform different control operations on the target object T according to different states.

[0075] In some embodiments, after setting the target object T to a picking state, the object motion detection device 1 further determines the pose of the target object T based on the hand pose of the user U, or determines the hand pose of the user U based on the pose of the target object T. Correspondingly, the object motion detection device 1 may further render a virtual object corresponding to the target object T and / or an image of the user U's hand.

[0076] Specifically, the processor 12 further performs the following operations: in response to the target object being set to the picking state, it determines an object pose of the target object based on a part pose of a user.

[0077] For example, when the object motion detection device 1 can only obtain the three-degree-of-freedom data of the target object T, but can obtain the six-degree-of-freedom pose of the user U in three-dimensional space through methods such as image tracking, when the target object T is picked up, the object motion detection device 1 can directly track the user U's hand that picks up the target object T, and use the pose of the hand as the six-degree-of-freedom pose of the target object T. Accordingly, without the need to set up additional sensors, the object motion detection device 1 can also obtain more detailed pose information of the target object T.

[0078] In some embodiments, after the target object T is set to a pick-up state, the object dynamic detection device 1 unlocks at least one function of the target object T.

[0079] Specifically, the processor 12 further performs the following operation: unlocking a function of the target object in response to setting the target object to the pick-up state.

[0080] For example, when user U picks up target object T, it means that user U is using target object T, and the object motion detection device 1 unlocks certain functions. For example, it may activate auxiliary components on the gun or activate sensors on target object T.

[0081] Conversely, in some cases, when it is determined that the target object T has been dropped or placed back in its original position, the object motion detection device 1 can lock some functions of the target object T to prevent accidental touch, unauthorized use, or wasted energy. For example, activating the gun safety, disabling components and / or sensors.

[0082] In summary, the object motion detection device 1 disclosed herein can determine the state of the target object, including whether it has been picked up, dropped, or placed back in its original position, based on the relative posture of the target object and the reference object. The technical means employed by the object motion detection device 1 does not require additional sensors on the target object, thus eliminating the possibility of sensor interference leading to misjudgments. Furthermore, the object motion detection device 1 can adopt different control methods based on different states of the target object.

[0083] Please refer to Figure 12, which is a flowchart of the object motion detection method 800 in the second embodiment of this disclosure. The object motion detection method 800 includes steps S801 and S803. The object motion detection method 800 is used to determine whether a target object has been picked up. The object motion detection method 800 can be executed by an electronic device (e.g., the object motion detection device 1 in the first embodiment).

[0084] First, in step S801, the electronic device calculates a relative detection attitude between a detection target attitude of the target object and a detection reference attitude of the reference object, wherein the detection target attitude and the detection reference attitude are measured when the target object is in the detection state.

[0085] Finally, in step S803, in response to a first difference between the detected relative posture and the initial relative posture being greater than a first threshold, the electronic device determines that the target object has been picked up and sets the target object to a picked-up state.

[0086] In some embodiments, the object dynamic detection method 800 further includes the electronic device calculating an initial relative attitude between an initial target attitude of a target object and an initial reference attitude of a reference object, wherein the initial target attitude and the initial reference attitude are measured when the target object is in an initial state; and after the electronic device calculates the initial relative attitude, it sets the target object to a detection state.

[0087] In some embodiments, the object dynamic detection method 800 further includes the electronic device calculating a plurality of first relative postures between a plurality of first target postures of the target object and a plurality of first reference postures of the reference object, wherein the first target postures and the first reference postures are measured in a time interval; and in response to a second difference between the first relative postures being less than a second threshold, the electronic device sets the target object to the initial state.

[0088] In some embodiments, the object dynamic detection method 800 further includes the electronic device calculating a post-removal target posture of the target object and a post-removal reference posture of the reference object, wherein the post-removal target posture and the post-removal reference posture are measured when the target object is in the removal state; and in response to a third difference between the post-removal relative posture and the initial relative posture being less than a third threshold, the electronic device sets the target object to the detection state.

[0089] In some embodiments, the step of determining that the target object has been picked up further includes, in response to the first difference being greater than the first threshold, the electronic device determining whether the target object is in a falling state based on an inertial data of the target object; and in response to the target object not being in the falling state, the electronic device determining that the target object has been picked up and setting the target object to the picked-up state.

[0090] In some embodiments, the object dynamic detection method 800 further includes the electronic device acquiring inertial data measured by an inertial measurement element of the target object; and in response to the inertial data being greater than an inertial threshold, the electronic device setting the target object to a falling state.

[0091] In some embodiments, the object dynamic detection method 800 further includes the electronic device acquiring inertial data measured by an inertial measurement element of the target object; and the electronic device determining, based on the inertial data, whether to set the target object to a falling state using a detection model.

[0092] In some embodiments, the object dynamic detection method 800 further includes the electronic device calculating a falling target posture of the target object and a falling reference posture of the reference object, wherein the falling target posture and the falling reference posture are measured when the target object is in a falling state; and in response to a fourth difference between the falling relative posture and the initial relative posture being less than a fourth threshold, the electronic device sets the target object to the detection state.

[0093] In some embodiments, the object dynamic detection method 800 further includes, in response to setting the target object to a falling state, the electronic device acquires inertial data measured by an inertial measurement element of the target object; the electronic device determines whether the target object has been moved based on the inertial data; and in response to determining that the target object has been moved, the electronic device sets the target object to the picking state.

[0094] In some embodiments, the object dynamic detection method 800 further includes, in response to the target object being set to the picking state, the electronic device determining an object pose of the target object based on a part pose of a user.

[0095] In some embodiments, the object dynamic detection method 800 further includes a function that the electronic device unlocks the target object in response to setting the target object to the picked-up state.

[0096] In summary, the object dynamic detection method 800 disclosed herein can determine the state of the target object, including whether it has been picked up, dropped, or placed in its initial position, based on the relative posture of the target object and the reference object. The technical means employed by the object dynamic detection method 800 does not require additional sensors on the target object, thus eliminating the possibility of sensor interference leading to misjudgments. Furthermore, the object dynamic detection method 800 can adopt different control measures based on different states of the target object.

[0097] The object motion detection method described in the second embodiment can be implemented by a computer program having a plurality of instructions. Each computer program can be a file that can be transmitted over a network, or it can be stored in a non-transitory computer-readable storage medium. For each computer program, after the instructions contained therein are loaded into an electronic device (e.g., object motion detection device 1), the computer program executes the object motion detection method described in the second embodiment. The non-transitory computer-readable storage medium can be an electronic product, such as: a read-only memory (ROM), a flash memory, a floppy disk, a hard disk, a compact disk (CD), a USB flash drive, a database accessible via a network, or any other storage medium known to those skilled in the art to which this disclosure pertains and having the same function.

[0098] Although several embodiments have been described in detail above as examples, the object dynamic detection device and method disclosed herein can also be implemented in other systems, hardware, software, storage media, or combinations thereof. Therefore, the scope of protection of this disclosure should not be limited to the specific implementations described in the embodiments disclosed herein, but should be determined by the claims outlined in the appended patent claims.

[0099] It will be apparent to those skilled in the art to which this disclosure pertains that various modifications and variations can be made to the structure of this disclosure without departing from its scope or spirit. In view of the foregoing, the scope of protection of this disclosure also covers modifications and variations made within the scope of the appended patent applications.

[0100] 1: Object motion detection device 12: Processor 14: Communication Interface U: User T: Target object R: Reference object P101, P103, P105, P107, P109, P201, P203, P205, P301, P303, P305, P501, P503, P505, P601, P603, P605, P701, P703, P705: Operation A, B, C, D: Nodes DP: Inertial Data 800: Object Motion Detection Method S801, S803: Steps

[0101] Domestic storage information (please note in order of storage institution, date, and number) none Overseas storage information (please note in the order of storage country, institution, date, and number) none

Claims

1. A method for detecting object dynamics, applicable to an electronic device, comprising the steps of: calculating a detection relative posture between a detection target posture of a target object and a detection reference posture of a reference object, wherein the detection target posture and the detection reference posture are measured when the target object is in a detection state; and determining that the target object has been picked up and setting the target object to a picked-up state in response to a first difference between the detection relative posture and an initial relative posture being greater than a first threshold; wherein the object dynamics detection method further comprises: calculating the initial relative posture between an initial target posture of the target object and an initial reference posture of the reference object, wherein the initial target posture and the initial reference posture are measured when the target object is in an initial state; and setting the target object to the detection state after calculating the initial relative posture; wherein the object dynamics detection method further comprises: Calculate a plurality of first relative poses between a plurality of first target poses of the target object and a plurality of first reference poses of the reference object, wherein the first target poses and the first reference poses are measured in a time interval; and set the target object to the initial state in response to a second difference between the first relative poses being less than a second threshold.

2. The object dynamic detection method as described in claim 1 further includes: calculating a post-grab target posture of the target object and a post-grab reference posture of the reference object, wherein the post-grab target posture and the post-grab reference posture are measured when the target object is in the grab state; and setting the target object to the detection state in response to a third difference between the post-grab relative posture and the initial relative posture being less than a third threshold.

3. The object dynamic detection method as described in claim 1, wherein the step of determining that the target object has been picked up further comprises: in response to the first difference being greater than the first threshold, determining whether the target object is in a falling state based on an inertial data of the target object; and in response to the target object not being in the falling state, determining that the target object has been picked up and setting the target object to the picked-up state.

4. The object dynamic detection method as described in claim 1 further includes: obtaining inertial data measured by an inertial measurement element of the target object; and setting the target object to a falling state in response to the inertial data being greater than an inertial threshold.

5. The object dynamic detection method as described in claim 1 further includes: obtaining inertial data measured by an inertial measurement element of the target object; and determining, based on the inertial data, whether to set the target object to a falling state using a detection model.

6. The object dynamic detection method as described in claim 1 further includes: calculating a falling target posture of the target object and a falling reference posture of the reference object, wherein the falling target posture and the falling reference posture are measured when the target object is in a falling state; and setting the target object to the detection state in response to a fourth difference between the falling relative posture and the initial relative posture being less than a fourth threshold.

7. The object dynamic detection method as described in claim 1 further includes: in response to setting the target object to a falling state, acquiring inertial data measured by an inertial measurement element of the target object; determining, based on the inertial data, whether the target object has been moved; and in response to determining that the target object has been moved, setting the target object to the picking state.

8. The object dynamic detection method as described in claim 1 further includes: in response to the target object being set to the picking state, determining an object pose of the target object based on a part pose of a user.

9. An object dynamic detection device, comprising: a communication interface for communicatively connecting a target object and a reference object; and a processor electrically connected to the communication interface, configured to perform the following operations: calculating a detection relative attitude between a detection target attitude of the target object and a detection reference attitude of the reference object, wherein the detection target attitude and the detection reference attitude are measured when the target object is in a detection state; and determining that the target object has been picked up and setting the target object to a picked-up state in response to a first difference between the detection relative attitude and an initial relative attitude being greater than a first threshold; wherein the object dynamic detection device further comprises: calculating the initial relative attitude between an initial target attitude of the target object and an initial reference attitude of the reference object, wherein the initial target attitude and the initial reference attitude are measured when the target object is in an initial state; and setting the target object to the detection state after calculating the initial relative attitude; wherein the object dynamic detection device further comprises: Calculate a plurality of first relative poses between a plurality of first target poses of the target object and a plurality of first reference poses of the reference object, wherein the first target poses and the first reference poses are measured in a time interval; and set the target object to the initial state in response to a second difference between the first relative poses being less than a second threshold.

10. The object motion detection apparatus as claimed in claim 9, wherein the processor is further configured to perform the following operations: calculating a post-removal target pose of the target object and a post-removal reference pose of the reference object, wherein the post-removal target pose and the post-removal reference pose are measured when the target object is in the removal state; and setting the target object to the detection state in response to a third difference between the post-removal relative pose and the initial relative pose being less than a third threshold.

11. The object motion detection device as claimed in claim 9, wherein the operation of determining that the target object has been picked up further comprises: in response to the first difference being greater than the first threshold, determining whether the target object is in a falling state based on an inertial data of the target object; and in response to the target object not being in the falling state, determining that the target object has been picked up and setting the target object to the picked-up state.

12. The object motion detection device as claimed in claim 9, wherein the processor is further configured to perform the following operations: acquiring inertial data measured by an inertial measurement element of the target object; and setting the target object to a falling state in response to the inertial data being greater than an inertial threshold.

13. The object motion detection apparatus as claimed in claim 9, wherein the processor is further configured to perform the following operations: calculate a relative falling posture between a target falling posture of the target object and a reference falling posture of the reference object, wherein the target falling posture and the reference falling posture are measured when the target object is in a falling state; and set the target object to the detection state in response to a fourth difference between the relative falling posture and the initial relative posture being less than a fourth threshold.

14. The object motion detection device as claimed in claim 9, wherein the processor is further configured to perform the following operations: in response to setting the target object to a falling state, acquiring inertial data measured by an inertial measurement element of the target object; determining, based on the inertial data, whether the target object has been moved; and in response to determining that the target object has been moved, setting the target object to the picking state.

15. The object motion detection device as claimed in claim 9, wherein the processor is further configured to perform the following operations: in response to the target object being set to the picking state, determining an object pose of the target object based on a part pose of a user.

16. A non-transitory computer-readable storage medium having at least one instruction stored thereon, wherein when a processor executes the at least one instruction, the at least one instruction executes an object dynamic detection method, the object dynamic detection method comprising the steps of: calculating a detection relative attitude between a detection target attitude of a target object and a detection reference attitude of a reference object, wherein the detection target attitude and the detection reference attitude are measured when the target object is in a detection state; and determining that the target object has been picked up and setting the target object to a picked-up state in response to a first difference between the detection relative attitude and an initial relative attitude being greater than a first threshold; wherein the object dynamic detection method further comprises: calculating the initial relative attitude between an initial target attitude of the target object and an initial reference attitude of the reference object, wherein the initial target attitude and the initial reference attitude are measured when the target object is in an initial state; and setting the target object to the detection state after calculating the initial relative attitude; The object dynamic detection method further includes: calculating a plurality of first relative poses between a plurality of first target poses of the target object and a plurality of first reference poses of the reference object, wherein the first target poses and the first reference poses are measured in a time interval; and setting the target object to the initial state in response to a second difference between the first relative poses being lower than a second threshold.

Images