CN110827226B - Skeleton point smoothing method and device and electronic equipment - Google Patents

Skeleton point smoothing method and device and electronic equipment Download PDF

Info

Publication number
CN110827226B
CN110827226B CN201911111739.9A CN201911111739A CN110827226B CN 110827226 B CN110827226 B CN 110827226B CN 201911111739 A CN201911111739 A CN 201911111739A CN 110827226 B CN110827226 B CN 110827226B
Authority
CN
China
Prior art keywords
motion parameter
value
parameter
current moment
motion
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201911111739.9A
Other languages
Chinese (zh)
Other versions
CN110827226A (en
Inventor
何立伟
邹佳辰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Megvii Technology Co Ltd
Original Assignee
Beijing Megvii Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Megvii Technology Co Ltd filed Critical Beijing Megvii Technology Co Ltd
Priority to CN201911111739.9A priority Critical patent/CN110827226B/en
Publication of CN110827226A publication Critical patent/CN110827226A/en
Application granted granted Critical
Publication of CN110827226B publication Critical patent/CN110827226B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G06T5/70—
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20172—Image enhancement details
    • G06T2207/20201—Motion blur correction
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30008—Bone

Abstract

The invention provides a method and a device for smoothing bone points and electronic equipment, wherein the method comprises the following steps: acquiring position information of a current skeleton point of a target object and an optimized value of a motion parameter of a previous moment; determining a measurement value of the motion parameter at the current moment and an estimation value of the motion parameter at the current moment based on the position information of the skeleton point and the optimized value of the motion parameter at the previous moment; determining an optimized value of the motion parameter at the current moment according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment; and determining the position information of the bone point after the smoothing processing at the current moment based on the optimized value of the motion parameter at the current moment. The method can well inhibit the vibration of the skeleton points in the human body under the condition that the skeleton points and the human body follow the skeleton points as much as possible, and the smooth effect of the skeleton points is good.

Description

Skeleton point smoothing method and device and electronic equipment
Technical Field
The invention relates to the technical field of computer vision, in particular to a method and a device for smoothing bone points and electronic equipment.
Background
The human skeleton point is important for describing human postures and predicting human behaviors. In the real-time detection engineering of the bone points, when a human body moves rapidly, the bone points output by the real-time detection engineering of the bone points are expected to move along with the rapid movement of the human body; when the human body is still, the bone points output by the real-time bone point detection project are expected to basically stay at the original positions. However, due to the diversity and complexity of the scene, the real-time bone points output by the bone point model in the bone point real-time detection project have obvious jitter in the human body, and in order to solve the problem of jitter of the bone points in the human body and introduce delay as little as possible, a smoothing algorithm (for smoothing the real-time bone points output by the bone point model) is also designed in the bone point real-time detection project.
The existing skeleton point smoothing algorithm usually determines that the smoothed current skeleton point coordinate is a weighted average of the current skeleton point coordinate and the previous skeleton point coordinate or the current skeleton point coordinate according to a relationship between a distance between the current skeleton point coordinate and the previous skeleton point coordinate and a preset threshold. In the weighted average, the weight is preset and cannot be adjusted. If the weight value corresponding to the coordinate value of the skeleton point at the previous moment is too large, the skeleton point cannot follow the human body when the human body moves rapidly, and the delay is serious; if the weight corresponding to the coordinate value of the skeleton point at the previous moment is too small, the effect of suppressing the jitter is not good.
In summary, the conventional bone point smoothing algorithm cannot achieve good jitter suppression while considering delay, that is, the smoothing effect of the conventional bone point smoothing algorithm is poor.
Disclosure of Invention
In view of the above, the present invention provides a method, an apparatus and an electronic device for smoothing bone points, so as to alleviate the technical problem that the smoothing effect of the conventional bone point smoothing algorithm is poor.
In a first aspect, an embodiment of the present invention provides a method for smoothing bone points, including: acquiring position information of a current skeleton point of a target object and an optimized value of a motion parameter of a previous moment; the motion parameters comprise the relative position relation between skeleton points; determining a measurement value of the motion parameter at the current moment and an estimation value of the motion parameter at the current moment based on the position information of the skeleton point and the optimized value of the motion parameter at the previous moment; determining an optimized value of the motion parameter at the current moment according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment; and determining the position information of the skeleton point subjected to smoothing processing at the current moment based on the optimized value of the motion parameter at the current moment.
Further, the step of obtaining the position information of the bone point of the target object at the current moment includes: and detecting the image of the target object at the current moment by adopting a skeleton point detection model to obtain the position information of the skeleton point of the target object at the current moment.
Further, when the skeleton point corresponding to the position information of the skeleton point is a trunk skeleton point, the motion parameter is represented by the position information of the skeleton point; and when the skeleton points corresponding to the position information of the skeleton points are the limb skeleton points or the head skeleton points, the motion parameters are represented by angle parameters, angular acceleration parameters, linear distance parameters and speed parameters.
Further, when the skeleton point corresponding to the position information of the skeleton point is a trunk skeleton point, and when the current time is a first time, the optimized value of the motion parameter at the previous time does not exist, the measured value of the motion parameter at the current time and the estimated value of the motion parameter at the current time are determined based on the position information of the skeleton point.
Further, the step of determining the measured value of the motion parameter at the current time and the estimated value of the motion parameter at the current time based on the position information of the bone point includes: and respectively taking the position information of the skeleton point as a measurement value of the motion parameter at the current moment and an estimation value of the motion parameter at the current moment.
Further, when a skeleton point corresponding to the position information of the skeleton point is a limb skeleton point or a head skeleton point, and when the current time is a first time, the optimized value of the motion parameter at the previous time does not exist, the measured value of the motion parameter at the current time and the estimated value of the motion parameter at the current time are determined based on the position information of the skeleton point.
Further, the step of determining the measured value of the motion parameter at the current time and the estimated value of the motion parameter at the current time based on the position information of the bone point includes: calculating a measurement value of an angle parameter in the measurement values of the motion parameters at the current moment and a measurement value of a linear distance parameter in the measurement values of the motion parameters at the current moment based on the position information of the bone points; taking the measured value of the angle parameter as an estimated value of the angle parameter in the estimated value of the motion parameter at the current moment, and taking the measured value of the linear distance parameter as an estimated value of the linear distance parameter in the estimated value of the motion parameter at the current moment; setting the measured value of the angular acceleration parameter in the measured value of the motion parameter at the current moment and the estimated value of the angular acceleration parameter in the estimated value of the motion parameter at the current moment as a first preset value, and setting the measured value of the velocity parameter in the measured value of the motion parameter at the current moment and the estimated value of the velocity parameter in the estimated value of the motion parameter at the current moment as a second preset value.
Further, when the skeleton point corresponding to the position information of the skeleton point is a trunk skeleton point, the step of determining the measurement value of the motion parameter at the current moment and the estimation value of the motion parameter at the current moment based on the position information of the skeleton point and the optimized value of the motion parameter at the previous moment includes: using the position information of the skeleton point as a measurement value of the motion parameter at the current moment; calculation formula M based on estimated values of motion parameters k ＝MP k-1 + v × t calculating an estimated value of the motion parameter at the current moment; m k An estimated value, MP, representing said motion parameter at the current moment k-1 Represents the optimized value of the motion parameter at the previous moment, v represents the speed of the bone point, and t represents the time difference between the current moment and the previous moment.
Further, when a skeleton point corresponding to the position information of the skeleton point is a limb skeleton point or a head skeleton point, the step of determining the measurement value of the motion parameter at the current moment and the estimation value of the motion parameter at the current moment based on the position information of the skeleton point and the optimized value of the motion parameter at the previous moment includes: determining a measurement value of the motion parameter at the current moment based on the position information of the bone point and the optimized value of the motion parameter at the previous moment; and determining the estimated value of the motion parameter at the current moment based on the optimized value of the motion parameter at the previous moment.
Further, the step of determining the measurement value of the motion parameter at the current time based on the position information of the bone point and the optimized value of the motion parameter at the previous time comprises: calculating the measured value of the angle parameter in the measured value of the motion parameter at the current moment according to the position information of the bone point to be smoothed at the current moment and the position information of the last bone point of the bone point to be smoothed at the current moment; the bone point to be smoothed is any one of the limb bone point or the head bone point; calculating the measurement value of the linear distance parameter in the measurement values of the motion parameters at the current moment according to the position information of the bone point to be smoothed at the current moment and the position information of the last bone point of the bone point to be smoothed at the current moment; calculating the measured value of the angular acceleration parameter in the measured value of the motion parameter at the current moment according to the measured value of the angle parameter and the angle parameter in the optimized value of the motion parameter at the previous moment; and calculating the measured value of the speed parameter in the measured values of the motion parameters at the current moment according to the measured value of the linear distance parameter and the linear distance parameter in the optimized value of the motion parameters at the previous moment.
Further, the step of determining the estimated value of the motion parameter at the current time based on the optimized value of the motion parameter at the previous time comprises: calculating an estimation value of an angle parameter in the estimation value of the motion parameter at the current moment according to the angle parameter in the optimized value of the motion parameter at the previous moment and the angular acceleration parameter in the optimized value of the motion parameter at the previous moment; calculating an estimated value of a linear distance parameter in the estimated value of the motion parameter at the current moment according to the linear distance parameter in the optimized value of the motion parameter at the previous moment and the speed parameter in the optimized value of the motion parameter at the previous moment; determining an estimated value of the angular acceleration parameter in the estimated value of the motion parameter at the current moment according to the angular acceleration parameter in the optimized value of the motion parameter at the previous moment; and determining the estimated value of the speed parameter in the estimated value of the motion parameter at the current moment according to the speed parameter in the optimized value of the motion parameter at the previous moment.
Further, the step of determining the optimized value of the motion parameter at the current moment according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment comprises: determining a first weight and a second weight according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment; the first weight is the weight corresponding to the measured value of the motion parameter at the current moment, and the second weight is the weight corresponding to the estimated value of the motion parameter at the current moment; and performing weighted calculation on the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment based on the first weight and the second weight to obtain an optimized value of the motion parameter at the current moment.
Further, the step of determining the first weight and the second weight according to the measured value of the motion parameter at the current time and the estimated value of the motion parameter at the current time includes: calculating a difference value between the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment; calculating the second weight according to a preset weight calculation function and the difference value; calculating the first weight based on the second weight; when the difference is smaller than a first preset threshold, or when the difference is larger than a second preset threshold, the first weight is positively correlated with the difference, the second weight is negatively correlated with the difference, and the first preset threshold is smaller than the second preset threshold.
Further, when the skeleton point corresponding to the position information of the skeleton point is a trunk skeleton point, the step of determining the position information of the skeleton point after the smoothing processing at the current time based on the optimized value of the motion parameter at the current time includes: and taking the optimized value of the motion parameter at the current moment as the position information of the bone point after the smoothing processing at the current moment.
Further, when a skeleton point corresponding to the position information of the skeleton point is an extremity skeleton point or a head skeleton point, the step of determining the position information of the skeleton point after the smoothing processing at the current time based on the optimized value of the motion parameter at the current time includes: and converting the optimized value of the motion parameter at the current moment into a Cartesian coordinate system to obtain the position information of the bone point smoothed at the current moment.
In a second aspect, an embodiment of the present invention further provides a device for smoothing bone points, including: the acquisition unit is used for acquiring the position information of the current skeleton point of the target object and the optimized value of the motion parameter at the previous moment; the motion parameters comprise the relative position relation between skeleton points; the first determination unit is used for determining a measurement value of the motion parameter at the current moment and an estimation value of the motion parameter at the current moment based on the position information of the bone point and the optimized value of the motion parameter at the previous moment; the optimization unit is used for determining the optimized value of the motion parameter at the current moment according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment; and the second determining unit is used for determining the position information of the bone point subjected to smoothing processing at the current moment based on the optimized value of the motion parameter at the current moment.
In a third aspect, an embodiment of the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the method according to any one of the above first aspects when executing the computer program.
In a fourth aspect, an embodiment of the present invention provides a computer storage medium, on which a computer program is stored, and when the computer program runs on a computer, the computer executes the steps of the method according to any one of the first aspect.
In the embodiment of the invention, the position information of the current skeleton point of the target object and the optimized value of the motion parameter at the previous moment are obtained firstly; then, determining a measurement value of the motion parameter at the current moment and an estimation value of the motion parameter at the current moment based on the position information of the skeleton point and the optimized value of the motion parameter at the previous moment; further, determining an optimized value of the motion parameter at the current moment according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment; and finally, determining the position information of the skeleton point subjected to smoothing processing at the current moment based on the optimized value of the motion parameter at the current moment. As can be seen from the above description, the method of the present invention applies the iterative concept of Kalman filtering, and the motion parameters therein include the relative position relationship between the skeleton points, so that it can be scientifically defined whether the position of a certain skeleton point at the current moment deviates too much, the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment can accurately reflect the intensity of the motion of the target object, therefore, the finally obtained optimized value of the motion parameter at the current moment is more scientific, the position information of the bone point after the smooth processing at the current moment is determined based on the scientific optimized value of the motion parameter at the current moment is more scientific and accurate, so that the smooth effect of the skeleton points is good, and under the condition that the skeleton points and the human body follow each other as much as possible (namely low delay), the method well inhibits the shaking of the bone points in the human body, and solves the technical problem of poor smoothing effect of the conventional bone point smoothing algorithm.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a schematic diagram of an electronic device according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method for smoothing bone points according to an embodiment of the present invention;
fig. 3 is a flowchart of a method for determining a measured value of a motion parameter at a current time and an estimated value of the motion parameter at the current time based on location information of a bone point according to an embodiment of the present invention;
FIG. 4a is a schematic diagram illustrating the locations of bone points of a target object according to an embodiment of the present invention;
FIG. 4b is a simplified diagram of a left arm bone point of a target object according to an embodiment of the present invention;
FIG. 5 is a graph of a second weight represented by a predetermined weight calculation function according to an embodiment of the present invention;
fig. 6 is a flowchart of a method for determining a measurement value of a motion parameter at a current time and an estimation value of the motion parameter at the current time based on position information of a bone point and an optimized value of the motion parameter at a previous time according to an embodiment of the present invention;
fig. 7 is a schematic view of a bone point smoothing device according to an embodiment of the present invention.
Detailed Description
The technical solutions of the present invention will be described clearly and completely with reference to the following embodiments, and it should be understood that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the prior art, a skeleton point smoothing algorithm generally performs weighted average on a current skeleton point coordinate value and a coordinate value of a skeleton point at the previous moment, and sets a threshold value, when the distance between the skeleton points corresponding to the current moment and the next moment is greater than the threshold value, it is determined that a person is moving fast, and then the current skeleton point coordinate value is directly used as the current skeleton point coordinate value after smoothing processing, so that the skeleton point and the human body can be quickly followed; and when the distance between the bone points corresponding to the current moment and the later moment is not more than a threshold value, determining that the human body is still, and further taking the result obtained by the weighted average as the coordinates of the current bone point after smoothing treatment, so that the jitter of the bone points in the human body can be reduced. It is desirable to reduce jitter and to make the resulting bone point follow the human body in the above manner.
However, in the above-described skeleton point smoothing algorithm, the weight of the weighted average is preset, and cannot be adjusted. If the weight value corresponding to the coordinate value of the skeleton point at the previous moment is too large, the skeleton point cannot follow the human body when the human body moves rapidly, and the delay is serious; if the weight corresponding to the coordinate value of the skeleton point at the previous moment is too small, the effect of suppressing the jitter is not good. In addition, since the weighting values are set in advance, the degree to which the jitter within the threshold is limited is also completely equivalent. Based on this, the embodiment of the present invention improves the method for smoothing the bone points, and can ensure that the bone points follow the human body as much as possible (i.e., with low delay), so as to well suppress the vibration of the bone points in the human body.
Example 1:
first, an electronic device 100 for implementing an embodiment of the present invention, which may be used to run the bone point smoothing method of embodiments of the present invention, is described with reference to fig. 1.
As shown in FIG. 1, electronic device 100 includes one or more processors 102, one or more memories 104, an input device 106, an output device 108, and a camera 110, which are interconnected via a bus system 112 and/or other form of connection mechanism (not shown). It should be noted that the components and structure of the electronic device 100 shown in fig. 1 are exemplary only, and not limiting, and the electronic device may have other components and structures as desired.
The processor 102 may be implemented in at least one hardware form of a Digital Signal Processor (DSP), a Field-Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), and an asic (application Specific Integrated circuit), and the processor 102 may be a Central Processing Unit (CPU) or other form of Processing Unit having data Processing capability and/or instruction execution capability, and may control other components in the electronic device 100 to perform desired functions.
The memory 104 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, Random Access Memory (RAM), cache memory (cache), and/or the like. The non-volatile memory may include, for example, Read Only Memory (ROM), hard disk, flash memory, etc. On which one or more computer program instructions may be stored that may be executed by processor 102 to implement client-side functionality (implemented by the processor) and/or other desired functionality in embodiments of the invention described below. Various applications and various data, such as various data used and/or generated by the applications, may also be stored in the computer-readable storage medium.
The input device 106 may be a device used by a user to input instructions and may include one or more of a keyboard, a mouse, a microphone, a touch screen, and the like.
The output device 108 may output various information (e.g., images or sounds) to the outside (e.g., a user), and may include one or more of a display, a speaker, and the like.
The camera 110 is configured to capture an image of the target object in real time, where the image of the target object captured by the camera obtains location information of the smoothed bone points after being processed by the bone point smoothing method, for example, the camera may capture a video desired by a user, and then obtain a video with the smoothed bone points after being processed by the bone point smoothing method, and the camera may further store the captured video in the memory 104 for use by other components.
Exemplarily, the electronic device for implementing the skeletal point smoothing method according to the embodiment of the present invention may be implemented as a smart mobile terminal such as a smartphone, a tablet computer, or the like, and may also be implemented as any other device with computing capability.
Example 2:
in accordance with an embodiment of the present invention, there is provided an embodiment of a method for smoothing bone points, it being noted that the steps illustrated in the flowchart of the drawings may be performed in a computer system such as a set of computer-executable instructions and that, although a logical order is illustrated in the flowchart, in some cases the steps illustrated or described may be performed in an order different than presented herein.
Fig. 2 is a flow chart of a method for smoothing bone points according to an embodiment of the present invention, as shown in fig. 2, the method comprising the steps of:
step S202, obtaining the position information of the current skeleton point of the target object and the optimized value of the motion parameter at the previous moment.
Wherein, the motion parameters comprise the relative position relation between the skeleton points.
In the embodiment of the present invention, the target object may be a human or an animal with a bone, and the target object is not particularly limited in the embodiment of the present invention.
The above-mentioned optimized value of the motion parameter at the previous moment is the motion parameter optimized by the smoothing method of the bone points according to the present invention. In addition, in the embodiment of the present invention, the motion parameters and the motion parameters hereinafter include the relative position relationship between the skeleton points, instead of processing the skeleton points as common discrete points as in the prior art, the motion parameters of the present invention can scientifically reflect the motion condition of the target object, and can accurately define whether the position of a certain skeleton point at the current time deviates too much.
In the target object, the motion amplitudes of different skeleton points are different, for example, for the skeleton point of the trunk, the skeleton point can only move within a small motion amplitude, and for the skeleton point of the limbs, especially the terminal skeleton point of the limbs, the motion amplitude relative to the trunk is large. Therefore, in order to reflect the relative position relationship between the skeleton points more scientifically, in the embodiment of the present invention, the skeleton points are divided into two types, one type is a trunk skeleton point, and the other type is a limb skeleton point or a head skeleton point, and the relative position relationship between the two types of skeleton points is expressed by different motion parameters, which will be described in detail hereinafter, and will not be described again.
And step S204, determining the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment based on the position information of the bone point and the optimized value of the motion parameter at the previous moment.
In the embodiment of the present invention, the measured value of the motion parameter at the current time includes the motion information of the current time of the target object, and the estimated value of the motion parameter at the current time includes the historical motion information of the target object, that is, the measured value of the motion parameter at the current time is the reflection of the motion information of the current time of the target object, and the estimated value of the motion parameter at the current time is the comprehensive reflection of the historical motion information of the target object. The bone point smoothing method not only considers the motion state of the previous moment, but also considers the motion state of a period of time, so that the subsequent determination of the motion intensity of the target object at the current moment according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment is more accurate, and the obtained optimized value of the motion parameter at the current moment is more accurate and scientific.
Step S206, determining the optimized value of the motion parameter at the current moment according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment.
During optimization, the weight corresponding to the measured value of the motion parameter at the current moment and the weight corresponding to the estimated value of the motion parameter at the current moment can be respectively determined according to the difference between the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment, and then the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment are respectively weighted and calculated with the corresponding weights, so that the optimized value of the motion parameter at the current moment is obtained. In essence, the difference between the measured value of the motion parameter at the current time and the estimated value of the motion parameter at the current time (since the measured value of the motion parameter at the current time is a reflection of the motion information of the target object at the current time and the estimated value of the motion parameter at the current time is a comprehensive reflection of the historical motion information of the target object) can reflect the intensity of the motion of the target object at the current time.
The larger the difference value is, the more violent the target object moves, the smaller the weight corresponding to the estimated value of the corresponding motion parameter at the current moment is, and the larger the weight corresponding to the measured value of the motion parameter at the current moment is, so that the proportion of the measured value of the motion parameter at the current moment (the reflection of the motion information of the target object at the current moment) in the optimized value of the motion parameter at the current moment can be increased, and the skeletal point and the human body can follow as much as possible; conversely, the smaller the difference value is, the more gradual the motion of the target object is, the larger the weight corresponding to the estimated value of the motion parameter at the current moment is, and the smaller the weight corresponding to the measured value of the motion parameter at the current moment is, so that the proportion of the estimated value of the motion parameter at the current moment (the comprehensive reflection of the historical motion information of the target object) in the optimized value of the motion parameter at the current moment can be increased, and the jitter of the skeletal point in the human body can be well inhibited.
Therefore, in the method for smoothing the bone points, the weights corresponding to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment are automatically adjusted in real time according to the difference value between the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment, and are not fixed, so that the determined optimized value of the motion parameter at the current moment can be more accurate, the determined position information of the bone points smoothed at the current moment is more accurate, the smoothing effect is good, and the bone points can be well inhibited from shaking in the human body under the condition that the bone points and the human body follow as much as possible (namely, low delay).
And S208, determining the position information of the bone point after the smoothing processing at the current moment based on the optimized value of the motion parameter at the current moment.
For different types of bone points, the corresponding motion parameters are expressed differently, so that the process of determining the position information of the bone point smoothed at the current time based on the optimized value of the motion parameter at the current time is also different, which is described in detail below.
In addition, after the position information of the skeleton point subjected to the smoothing processing at the current moment is obtained, the optimized value of the motion parameter at the current moment is further used as the optimized value of the motion parameter at the previous moment, the step of obtaining the position information of the skeleton point of the target object at the current moment and the optimized value of the motion parameter at the previous moment is returned to be executed, and the skeleton point of the target object at the next moment is continuously subjected to the smoothing processing based on the optimized value of the motion parameter at the current moment obtained by the iteration. That is, for each time, the above steps are performed.
In the embodiment of the invention, the position information of the current skeleton point of the target object and the optimized value of the motion parameter at the previous moment are obtained firstly; then, determining a measurement value of the motion parameter at the current moment and an estimation value of the motion parameter at the current moment based on the position information of the skeleton point and the optimized value of the motion parameter at the previous moment; further, determining an optimized value of the motion parameter at the current moment according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment; and finally, determining the position information of the skeleton point subjected to smoothing processing at the current moment based on the optimized value of the motion parameter at the current moment. As can be seen from the above description, the method of the present invention applies the iterative concept of Kalman filtering, and the motion parameters therein include the relative position relationship between the skeleton points, so that it can be scientifically defined whether the position of a certain skeleton point at the current moment deviates too much, the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment can accurately reflect the intensity of the motion of the target object, therefore, the finally obtained optimized value of the motion parameter at the current moment is more scientific, the position information of the bone point after the smooth processing at the current moment is determined based on the scientific optimized value of the motion parameter at the current moment is more scientific and accurate, so that the smooth effect of the skeleton points is good, and under the condition of ensuring that the skeleton points and the human body follow each other as much as possible (namely low delay), the method well inhibits the shaking of the bone points in the human body, and solves the technical problem of poor smoothing effect of the conventional bone point smoothing algorithm.
The foregoing briefly introduces a method for smoothing bone points according to an embodiment of the present invention, and the following describes the detailed description of the related details.
In this embodiment, an implementation manner of obtaining the position information of the bone point of the target object at the current time is given in step S202, and includes the following steps:
and detecting the image of the target object at the current moment by adopting a skeleton point detection model to obtain the position information of the skeleton point of the target object at the current moment.
Specifically, a skeleton point detection model is adopted to detect the image of the target object at the current moment, and then the position information of the skeleton point of the target object at the current moment is obtained. The skeletal point detection model is a model obtained by pre-training.
As can be seen from the description below step S202, in the embodiment of the present invention, the bone points are divided into two types, and the relative position relationship between the two types of bone points is expressed by different motion parameters, in an alternative implementation:
when the skeleton point corresponding to the position information of the skeleton point is a trunk skeleton point, representing the motion parameter by using the position information of the skeleton point;
when the skeleton point corresponding to the position information of the skeleton point is a limb skeleton point or a head skeleton point, the motion parameter is represented by an angle parameter, an angular acceleration parameter, a linear distance parameter and a speed parameter.
The angle parameter is an included angle between a connecting line of a skeleton point to be smoothed and a previous skeleton point and the vertical direction, the angular acceleration parameter is the change of the angle parameter in unit time, the linear distance parameter is the linear distance between the skeleton point to be smoothed and the previous skeleton point, the speed parameter is the speed of the skeleton point to be smoothed relative to the previous skeleton point, and the skeleton point to be smoothed is any one of a limb skeleton point or a head skeleton point.
That is, when the skeleton point is a trunk skeleton point (in the case of the human body standing forward, the trunk skeleton point can be considered as relatively static), since there is no relative reference point for movement, the movement parameter is directly expressed by the position information of the skeleton point; when the skeleton points are the limb skeleton points or the head skeleton points, the motion of the skeleton points relative to the previous skeleton point shows a certain rule (which can be approximately equal to circular motion around the previous skeleton point, but has no rule for the motion of other indirectly connected skeleton points, so the previous skeleton point is selected), the motion parameters are expressed by an angle parameter, an angular acceleration parameter, a linear distance parameter and a speed parameter, so that the change of the measured value is noise or autonomous motion, and the motion expression is more scientific in nature.
In order to better understand the method for smoothing the skeleton points of the present invention, the method for smoothing the skeleton points of the present invention is described below by taking the skeleton points of the trunk, the skeleton points of the limbs, or the skeleton points of the head as examples.
The first condition is as follows: the condition of the skeletal points of the trunk
In the embodiment of the invention, when the skeleton point corresponding to the position information of the skeleton point is the trunk skeleton point, and when the current moment is the first moment, the optimized value of the motion parameter at the previous moment does not exist, the measurement value of the motion parameter at the current moment and the estimation value of the motion parameter at the current moment are determined based on the position information of the skeleton point. Wherein, the first time is the initial time of the first processing.
In an alternative embodiment, the position information of the bone points is used as a measured value of the motion variable at the current time and as an estimated value of the motion variable at the current time, respectively.
That is, when the first iteration is performed, the obtained position information of the bone point of the target object at the current moment is directly used as the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment.
And then, determining the optimized value of the motion parameter at the current moment according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment. Specifically, a difference between a measured value of the motion parameter at the current moment and an estimated value of the motion parameter at the current moment is calculated; then, calculating a second weight K corresponding to the estimated value of the motion parameter at the current moment according to a preset weight calculation function and the difference value; calculating a first weight W (wherein the sum of the first weight and the second weight is equal to 1) corresponding to the measured value of the motion parameter at the current moment based on a second weight, wherein when the difference is smaller than a first preset threshold, or when the difference is larger than a second preset threshold, the first weight is positively correlated with the difference, the second weight is negatively correlated with the difference, and the first preset threshold is smaller than the second preset threshold; and then, based on the first weight and the second weight, carrying out weighted calculation on the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment to obtain an optimized value of the motion parameter at the current moment. Specifically, the optimized value of the motion parameter at the current moment is as follows: (x) 0,k ,y 0,k ) O ＝K(x 0,k ,y 0,k )+W(x 0,k ,y 0,k ) Wherein (x) 0,k ,y 0,k ) O Represents an optimized value of the motion parameter at the current moment, K represents a second weight, (x) 0,k ,y 0,k ) Representing the current-time motion parameter measurement value/the current-time motion parameter estimation value/the position information of the bone point, W representing the first weight.
The preset weight calculation function may be K ═ f (Δ), where K denotes the second weight and Δ denotes the difference.
Finally, the optimized value (x) of the motion parameter at the current moment is calculated 0,k ,y 0,k ) O As the position information of the bone point after the smoothing processing at the current time.
In the embodiment of the invention, when the skeleton point corresponding to the position information of the skeleton point is the trunk skeleton point, and when the current moment is not the first moment, the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment are determined based on the position information of the skeleton point and the optimized value of the motion parameter at the previous moment.
In an optional embodiment, the position information of the bone point is used as the measured value of the motion parameter at the current moment; and calculating formula M according to the estimated value of the motion parameter k ＝MP k-1 + v × t calculating the estimation value of the motion parameter at the current moment; wherein M is k Estimate, MP, representing a motion parameter at the current time k-1 Represents the optimized value of the motion parameter at the previous moment, v represents the speed of the bone point, and t represents the time difference between the current moment and the previous moment. Such as: estimation value of motion parameter at current time: (x) 0,k ,y 0,k ) N ＝(x 0,k-1 ,y 0,k-1 ) O + v × t, wherein (x) 0,k ,y 0,k ) N Estimated value M representing motion parameter at current time k ，(x 0,k-1 ,y 0,k-1 ) O Optimized value MP representing last moment motion parameter k-1 V denotes the velocity of the bone point, and t denotes the time difference between the current time and the previous time. The velocity of the bone points is explained below: in the second iteration, the speed of the bone point is a preset value 0, in the subsequent iteration process, the speed of the bone point is a quotient of the distance between the position of the bone point subjected to the smoothing processing at the current moment and the position of the bone point subjected to the smoothing processing at the previous moment and the time difference (the time difference between the current moment and the previous moment), and then the calculated speed is used in the next iteration process.
Furthermore, based on the measured value of the motion parameter at the current moment and the current momentThe estimation value of the motion parameter at the previous moment determines the optimized value of the motion parameter at the current moment. Specifically, a difference between a measured value of the motion parameter at the current moment and an estimated value of the motion parameter at the current moment is calculated; then, calculating a second weight K corresponding to the estimated value of the motion parameter at the current moment according to a preset weight calculation function and the difference value; calculating a first weight W (wherein the sum of the first weight and the second weight is equal to 1) corresponding to the measured value of the motion parameter at the current moment based on a second weight, wherein when the difference is smaller than a first preset threshold, or when the difference is larger than a second preset threshold, the first weight is positively correlated with the difference, the second weight is negatively correlated with the difference, and the first preset threshold is smaller than the second preset threshold; and then, based on the first weight and the second weight, carrying out weighted calculation on the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment to obtain an optimized value of the motion parameter at the current moment. Specifically, the optimized value of the motion parameter at the current moment is as follows: (x) 0,k ,y 0,k ) O ＝K(x 0,k ,y 0,k ) N +W(x 0,k ,y 0,k ) Wherein (x) 0,k ,y 0,k ) O Represents an optimized value of the motion parameter at the current moment, K represents a second weight, (x) 0,k ,y 0,k ) N An estimate value representing the motion parameter at the current time, (x) 0,k ,y 0,k ) Position information of a measured value/bone point of the motion parameter at the current time is represented, and W represents a first weight.
The preset weight calculation function may be K ═ f (Δ), where K denotes the second weight and Δ denotes the difference.
Finally, the optimized value (x) of the motion parameter at the current moment is calculated 0,k ,y 0,k ) O As the position information of the bone point after the smoothing processing at the current time.
Case two: in the case of extremity or head bone points
In the embodiment of the invention, when the skeleton point corresponding to the position information of the skeleton point is a limb skeleton point or a head skeleton point, and when the current time is the first time, the optimized value of the motion parameter at the previous time does not exist, the measured value of the motion parameter at the current time and the estimated value of the motion parameter at the current time are determined based on the position information of the skeleton point.
In an alternative embodiment, referring to fig. 3, the determining of the measured value of the motion parameter and the estimated value of the motion parameter at the current time based on the position information of the bone point comprises the following steps:
step S301, calculating the measured value of the angle parameter in the measured values of the motion parameters at the current moment and the measured value of the linear distance parameter in the measured values of the motion parameters at the current moment based on the position information of the bone points.
Specifically, referring to fig. 4a and 4b, the process of smoothing the wrist skeleton point is described by taking the skeleton point at the end of the left hand (i.e., the wrist skeleton point) as an example. It should be noted that, the smoothing process of other limb bone points and head bone points may refer to the smoothing process of wrist bone points, and the embodiments of the present invention are not described in detail.
In FIG. 4b, P 0,k-1 Indicates the wrist bone point, P, at time k-1 (i.e., the last time in the present embodiment) 0,k Wrist skeleton point, P, representing time k (i.e., the current time in an embodiment of the present invention) 1,k-1 Represents the elbow bone point at time k-1, P 1,k Represents the elbow bone point at time k, phi k-1 Represents the angle phi between the connecting line between the wrist skeleton point and the elbow skeleton point (namely the last skeleton point of the wrist skeleton point) at the moment of k-1 and the vertical direction k And an included angle between a connecting line between the wrist skeleton point and the elbow skeleton point at the moment k and the vertical direction is shown.
Correspondingly, the position information of the wrist skeleton point and the elbow skeleton point is obtained by the following steps: wrist skeleton point P at time k-1 0,k-1 Is (x) 0,k-1 ,y 0,k-1 ) Wrist skeleton point P at time k 0,k Is (x) 0,k ,y 0,k ) The elbow bone point at time k-1 is (x) 1,k-1 ,y 1,k-1 ) Elbow bone point P at time k 1,k Is (x) 1,k ,y 1,k )。
After the position information of the skeleton point is obtained, the measured value of the angle parameter in the measured values of the motion parameters at the current moment and the measured value of the linear distance parameter in the measured values of the motion parameters at the current moment are calculated.
Specifically, the method comprises the following steps: calculation formula according to angle parameters
Figure BDA0002271996990000181
Calculating the measured value of the angle parameter in the measured values of the motion parameters at the current moment; wherein phi is k,m Of the measured values representing the motion parameter at the current time (i.e. at time k), the measured value of the angle parameter, x 0,k Abscissa, x, representing wrist skeleton point at time k 1,k Abscissa, y, representing the elbow bone point at time k 0,k Ordinate, y, representing wrist bone point at time k 1,k The ordinate of the elbow bone point at time k is shown.
Calculation of equation d from linear distance k,m ＝|(x 0,k -x 1,k ,y 0,k -y 1,k ) L calculating the measured value of the linear distance parameter from the measured values of the motion parameters at the current moment, wherein d k,m Measurement values of a linear distance parameter, x, of the measurement values of the motion parameter at the current time (i.e. at time k) 0,k Abscissa, x, representing wrist skeleton point at time k 1,k Abscissa, y, representing the elbow bone point at time k 0,k Ordinate, y, representing wrist skeleton point at time k 1,k The ordinate of the elbow bone point at time k is shown.
Step S302, the measured value of the angle parameter is used as the estimated value of the angle parameter in the estimated value of the motion parameter at the current moment, and the measured value of the linear distance parameter is used as the estimated value of the linear distance parameter in the estimated value of the motion parameter at the current moment.
Step S303 sets the measured value of the angular acceleration parameter in the measured value of the motion parameter at the current time and the estimated value of the angular acceleration parameter in the estimated value of the motion parameter at the current time as a first preset value, and sets the measured value of the velocity parameter in the measured value of the motion parameter at the current time and the estimated value of the velocity parameter in the estimated value of the motion parameter at the current time as a second preset value.
In the embodiment of the present invention, the first preset value and the second preset value are the same and are both 0, and the first preset value and the second preset value are not specifically limited in the embodiment of the present invention.
And after the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment are obtained, determining the optimized value of the motion parameter at the current moment according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment. Specifically, a difference between a measured value of the motion parameter at the current moment and an estimated value of the motion parameter at the current moment is calculated; then, calculating a second weight K corresponding to the estimated value of the motion parameter at the current moment according to a preset weight calculation function and the difference value; and calculates a first weight W corresponding to the measured value of the motion parameter at the current time based on the second weight (wherein the sum of the first weight and the second weight is equal to 1). When the difference is smaller than a first preset threshold, or when the difference is larger than a second preset threshold, the first weight is positively correlated with the difference, the second weight is negatively correlated with the difference, and the first preset threshold is smaller than the second preset threshold.
The preset weight calculation function may be that K ═ f (Δ), where K denotes the second weight and Δ denotes the difference.
The graph of the second weight represented by the preset weight calculation function in the embodiment of the invention is shown in FIG. 5, wherein K is 1 、Δ 1 、Δ 2 And Δ 3 Can be adjusted according to the actual situation, delta 1 Representing a first predetermined threshold value, Δ 2 Representing a second predetermined threshold value, Δ 3 Representing a third preset threshold. As can be seen from fig. 5, when the difference Δ is smaller, that is, the motion of the target object is gentler, the second weight corresponding to the estimated value of the motion parameter at the current time is larger (where, when the difference Δ is 0, it indicates that the target object is still, at this time, the second weight corresponding to the estimated value of the motion parameter at the current time is 1, and the weight corresponding to the measured value of the motion parameter at the corresponding current time is 0); and when the difference value delta is larger, namely the target object moves more intensely, the second weight corresponding to the estimated value of the motion parameter at the current moment is smaller, and the weight corresponding to the measured value of the motion parameter at the corresponding current moment is larger. The difference in fig. 5 may represent a difference in an angle parameter, a difference in a linear distance parameter, or a difference in an angular acceleration parameter or a velocityThe difference values of the degree parameters and the preset weight calculation functions corresponding to the parameters may be the same or different, and the embodiment of the present invention does not limit the difference values. In addition, the above and below preset weight calculation functions may also be functions represented by the graphs shown in fig. 5, and of course, the preset weight calculation functions in the embodiment of the present invention may also be functions in other forms, and the embodiment of the present invention is not limited.
And then, based on the first weight and the second weight, carrying out weighted calculation on the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment to obtain an optimized value of the motion parameter at the current moment. Specifically, the optimized value of the motion parameter at the current moment is as follows:
Figure BDA0002271996990000201
wherein phi k Represents an optimized value of an angle parameter among the optimized values of the motion parameters at the present time, K represents a second weight,
Figure BDA0002271996990000202
measured value representing angle parameter/estimated value of angle parameter/angle phi at current moment k,m W denotes a first weight, d k An optimized value of a linear distance parameter in the optimized values representing the motion parameters at the present time, | (x) 0,k -x 1,k ,y 0,k -y 1,k ) | denotes the measured value of the linear distance parameter/the estimated value of the linear distance parameter/the distance d at the present time k,m ，a k An optimized value of an angular acceleration parameter, v, of the optimized values of the motion parameters representing the current moment k And represents an optimized value of the speed parameter in the optimized values of the motion parameters at the current moment.
And finally, converting the optimized value of the motion parameter at the current moment into a Cartesian coordinate system to obtain the position information of the bone point after the smooth processing at the current moment. And further converting the data in the angular coordinate system into a Cartesian coordinate system to obtain the position information of the skeleton point after the current time smoothing processing because the obtained optimized value of the motion parameter at the current time is the data in the angular coordinate system.
In the embodiment of the present invention, when a skeleton point corresponding to the position information of the skeleton point is an extremity skeleton point or a head skeleton point, and when the current time is not the first time, referring to fig. 6, the method determines the measurement value of the motion parameter at the current time and the estimation value of the motion parameter at the current time based on the position information of the skeleton point and the optimal motion parameter, and includes the following steps: determining a measurement value of the motion parameter at the current moment based on the position information of the skeleton point and the optimized value of the motion parameter at the previous moment; and determining the estimation value of the motion parameter at the current moment based on the optimized value of the motion parameter at the previous moment.
Wherein the step of determining the measured value of the motion parameter at the current time based on the position information of the bone point and the optimized value of the motion parameter at the previous time comprises the following processes (a) - (d):
(a) calculating the measured value of the angle parameter in the measured value of the motion parameter at the current moment according to the position information of the bone point to be smoothed at the current moment and the position information of the last bone point of the bone point to be smoothed at the current moment; wherein the skeleton point to be smoothed is any one of a limb skeleton point or a head skeleton point;
(b) calculating the measurement value of the linear distance parameter in the measurement values of the motion parameters at the current moment according to the position information of the bone point to be smoothed at the current moment and the position information of the last bone point of the bone point to be smoothed at the current moment;
(c) calculating the measured value of the angular acceleration parameter in the measured value of the motion parameter at the current moment according to the measured value of the angle parameter and the angle parameter in the optimized value of the motion parameter at the previous moment;
(d) and calculating the measured value of the speed parameter in the measured values of the motion parameters at the current moment according to the measured value of the linear distance parameter and the linear distance parameter in the optimized value of the motion parameters at the previous moment.
In specific implementation, the measurement value of the motion parameter at the current moment can be calculated according to the following measurement value calculation formula:
step S601, calculating formula according to measured value
Figure BDA0002271996990000221
And calculating the measured value of the motion parameter at the current moment.Wherein phi is k,m Measured values of angle parameters in the measured values representing the motion parameters at the present time, d k,m Measurement of a linear distance parameter in the measurement of the motion parameter at the current time, a k,m Measurement value of angular acceleration parameter in measurement value representing motion parameter at present moment, v k,m Measurement value of speed parameter, x, in measurement value representing motion parameter at present moment 0,k Abscissa, y, representing the bone point to be smoothed at the current moment 0,k Ordinate, x, representing the bone point to be smoothed at the current moment 1,k The abscissa, y, of the last bone point representing the bone point to be smoothed at the current moment 1,k The ordinate, phi, of the last bone point representing the bone point to be smoothed at the current moment k-1 Representing the angle parameter in the optimized value of the motion parameter at the previous moment, d k-1 And expressing a linear distance parameter in the optimized value of the motion parameter at the previous moment, wherein the skeleton point to be smoothed is any one of the limb skeleton point or the head skeleton point.
Specifically, for the above wrist skeleton point, the skeleton point to be smoothed in step S601 represents a wrist skeleton point, and the last skeleton point of the skeleton point to be smoothed represents an elbow skeleton point.
The step of determining the estimated value of the motion parameter at the present time based on the optimized value of the motion parameter at the previous time includes the following processes (a) to (D):
(A) calculating an estimation value of an angle parameter in the estimation value of the motion parameter at the current moment according to the angle parameter in the optimized value of the motion parameter at the previous moment and an angular acceleration parameter in the optimized value of the motion parameter at the previous moment;
(B) calculating an estimated value of the linear distance parameter in the estimated value of the motion parameter at the current moment according to the linear distance parameter in the optimized value of the motion parameter at the previous moment and the speed parameter in the optimized value of the motion parameter at the previous moment;
(C) determining an estimated value of the angular acceleration parameter in the estimated value of the motion parameter at the current moment according to the angular acceleration parameter in the optimized value of the motion parameter at the previous moment;
(D) and determining the estimated value of the speed parameter in the estimated value of the motion parameter at the current moment according to the speed parameter in the optimized value of the motion parameter at the previous moment.
In specific implementation, the estimation value of the motion parameter at the current time can be calculated according to the following estimation value calculation formula:
step S602, calculating formula according to the estimated value
Figure BDA0002271996990000231
And calculating an estimated value of the motion parameter at the current moment. Wherein phi is k,p An estimate of an angle parameter, d k,p An estimate of a linear distance parameter in the estimate representing the motion parameter at the current time, a k,p An estimated value of an angular acceleration parameter, v, representing an estimated value of a motion parameter at the present moment k,p An estimated value of a velocity parameter, phi, in the estimated value representing the motion parameter at the present time k-1 Representing the angle parameter in the optimized value of the motion parameter at the previous moment, a k-1 Representing the angular acceleration parameter in the optimized value of the motion parameter at the previous moment, d k-1 Linear distance parameter, v, in the optimized value representing the last moment motion parameter k-1 The velocity parameter in the optimized value of the motion parameter at the previous moment.
And then, determining the optimized value of the motion parameter at the current moment according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment. Specifically, a difference between a measured value of the motion parameter at the current moment and an estimated value of the motion parameter at the current moment is calculated; then, calculating a second weight K corresponding to the estimated value of the motion parameter at the current moment according to a preset weight calculation function and the difference value; calculating a first weight W corresponding to the measured value based on a second weight (wherein the sum of the first weight and the second weight is equal to 1), wherein when the difference is smaller than a first preset threshold, or when the difference is larger than a second preset threshold, the first weight is positively correlated with the difference, the second weight is negatively correlated with the difference, and the first preset threshold is smaller than the second preset threshold; then based on the first weight and the second weight, the measured value of the motion parameter at the current moment and the estimation of the motion parameter at the current moment are estimatedAnd carrying out weighted calculation on the values to obtain the optimized value of the motion parameter at the current moment. Specifically, the optimized value of the motion parameter at the current moment is as follows:
Figure BDA0002271996990000241
φ k angle parameter, K, in the optimized value representing the motion parameter at the current time φ,k A second weight, phi, corresponding to the estimated value of the angle parameter among the estimated values of the motion parameters at the current time k,p An estimate of an angle parameter, W, of the estimates of the motion parameters at the current time φ,k A first weight, phi, corresponding to the measured value of the angle parameter in the measured values of the motion parameter at the current moment k,m Measured values of angle parameters in the measured values representing the motion parameters at the present time, d k Linear distance parameter, K, in the optimized value representing the current moment motion parameter d,k A second weight corresponding to the estimated value of the linear distance parameter among the estimated values of the motion parameters at the current time, d k,p An estimate of a linear distance parameter, W, of the estimates representing the motion parameters at the current moment d,k A first weight corresponding to the measured value of the linear distance parameter in the measured values of the motion parameter at the current moment, d k,m Measurement of a linear distance parameter in the measurement of the motion parameter at the current time, a k Representing the angular acceleration parameter, K, in the optimized value of the motion parameter at the current moment a,k A second weight corresponding to the estimated value of the angular acceleration parameter among the estimated values of the motion parameters at the present time, a k,p An estimated value of an angular acceleration parameter, W, among estimated values representing the motion parameters at the present time a,k A first weight corresponding to the measured value of the angular acceleration parameter in the measured values of the motion parameter at the current moment, a k,m Measurement value of angular acceleration parameter in measurement value representing motion parameter at present moment, v k Representing the speed parameter, K, in the optimized value of the motion parameter at the current moment v,k A second weight, v, corresponding to the estimated value of the velocity parameter among the estimated values of the motion parameters at the present time k,p An estimate of a velocity parameter, W, of an estimate of a motion parameter representing the current moment v,k Measured value medium speed representing motion parameter at current momentFirst weights, v, corresponding to the measured values of the degree parameter k,m A measured value of the speed parameter is indicated among the measured values of the motion parameter at the present time.
And finally, converting the optimized value of the motion parameter at the current moment into a Cartesian coordinate system to obtain the position information of the bone point after the smooth processing at the current moment. Because the obtained optimized value of the motion parameter at the current moment is data in an angular coordinate system, the data in the angular coordinate system is further converted into a cartesian coordinate system, and the position information of the bone point after the current moment is smoothed is obtained, the result is:
Figure BDA0002271996990000251
wherein (x) 0,k ) O Abscissa (x) representing the smoothed bone point at the current time 1,k ) O D, showing the abscissa of the previous skeleton point after smoothing treatment at the current moment (for the wrist skeleton point to be smoothed, the previous skeleton point is the elbow skeleton point, the abscissa of the elbow skeleton point after smoothing treatment at the current moment is known, and the abscissa of the elbow skeleton point after smoothing treatment is calculated according to the information of the previous skeleton point), d k A linear distance parameter, phi, in the optimized value representing the motion parameter at the current moment k An angle parameter in the optimized value representing the motion parameter at the current moment, (y) 0,k ) O Represents the ordinate of the smoothed bone point at the current time, (y) 1,k ) O And (4) the ordinate of the previous skeleton point after smoothing processing at the current moment is shown (for the skeleton point to be smoothed, which is the wrist skeleton point, the ordinate of the elbow skeleton point after smoothing processing at the current moment is known, and the ordinate of the elbow skeleton point after smoothing processing is calculated according to the information of the previous skeleton point).
The method for smoothing the skeleton points can automatically adjust the corresponding weight in real time according to the difference value (reflecting the intensity of the movement of the target object) between the measured value of the movement parameter at the current moment and the estimated value of the movement parameter at the current moment, can ensure that the determined optimized value of the movement parameter is more accurate, further ensure that the position information of the skeleton points after smoothing processing at the current moment is more accurate, has good smoothing effect, and can well inhibit the shake of the skeleton points in the human body under the condition that the skeleton points and the human body follow as much as possible (namely low delay).
Example 3:
the embodiment of the present invention further provides a device for smoothing bone points, which is mainly used for executing the method for smoothing bone points provided by the embodiment of the present invention, and the following describes the device for smoothing bone points provided by the embodiment of the present invention in detail.
Fig. 7 is a schematic view of a bone point smoothing device according to an embodiment of the present invention, as shown in fig. 7, the bone point smoothing device mainly includes: an obtaining unit 10, a first determining unit 20, an optimizing unit 30 and a second determining unit 40, wherein:
the acquisition unit is used for acquiring the position information of the current skeleton point of the target object and the optimized value of the motion parameter at the previous moment; the motion parameters comprise the relative position relation between the skeleton points;
the first determining unit is used for determining a measured value of the motion parameter at the current moment and an estimated value of the motion parameter at the current moment based on the position information of the bone point and the optimized value of the motion parameter at the previous moment;
the optimization unit is used for determining the optimized value of the motion parameter at the current moment according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment;
and the second determining unit is used for determining the position information of the bone point subjected to the smoothing processing at the current moment based on the optimized value of the motion parameter at the current moment.
In the embodiment of the invention, the position information of the current skeleton point of the target object and the optimized value of the motion parameter at the previous moment are obtained firstly; then, determining a measurement value of the motion parameter at the current moment and an estimation value of the motion parameter at the current moment based on the position information of the skeleton point and the optimized value of the motion parameter at the previous moment; further, determining an optimized value of the motion parameter at the current moment according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment; and finally, determining the position information of the skeleton point after the smoothing treatment at the current moment based on the optimized value of the motion parameter at the current moment. As can be seen from the above description, the method of the present invention applies the iterative concept of Kalman filtering, and the motion parameters therein include the relative position relationship between the skeleton points, so that it can be scientifically defined whether the position of a certain skeleton point at the current moment deviates too much, the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment can accurately reflect the intensity of the motion of the target object, therefore, the finally obtained optimized value of the motion parameter at the current moment is more scientific, the position information of the bone point after the smooth processing at the current moment is determined based on the scientific optimized value of the motion parameter at the current moment is more scientific and accurate, so that the smooth effect of the skeleton points is good, and under the condition that the skeleton points and the human body follow each other as much as possible (namely low delay), the method well inhibits the shaking of the bone points in the human body, and solves the technical problem of poor smoothing effect of the conventional bone point smoothing algorithm.
Optionally, the obtaining unit is further configured to: and detecting the image of the target object at the current moment by adopting a skeleton point detection model to obtain the position information of the skeleton point of the target object at the current moment.
Optionally, when the skeleton point corresponding to the position information of the skeleton point is a trunk skeleton point, the motion parameter is represented by the position information of the skeleton point; when the skeleton point corresponding to the position information of the skeleton point is a limb skeleton point or a head skeleton point, the motion parameter is represented by an angle parameter, an angular acceleration parameter, a linear distance parameter and a speed parameter.
Optionally, when the skeleton point corresponding to the position information of the skeleton point is a trunk skeleton point, and when the current time is a first time, the optimized value of the motion parameter at the previous time does not exist, the first determining unit is further configured to: a measured value of the motion parameter at the current time and an estimated value of the motion parameter at the current time are determined based on the position information of the bone points.
Optionally, the first determining unit is further configured to: and respectively taking the position information of the skeleton point as a measured value of the motion parameter at the current moment and an estimated value of the motion parameter at the current moment.
Optionally, when the bone point corresponding to the position information of the bone point is a limb bone point or a head bone point, and when the current time is a first time, the optimized value of the motion parameter at the previous time does not exist, the first determining unit is further configured to: a measured value of the motion parameter at the current time and an estimated value of the motion parameter at the current time are determined based on the position information of the bone points.
Optionally, the first determining unit is further configured to: calculating a measurement value of an angle parameter in the measurement values of the motion parameters at the current moment and a measurement value of a linear distance parameter in the measurement values of the motion parameters at the current moment based on the position information of the skeleton point; taking the measured value of the angle parameter as the estimated value of the angle parameter in the estimated value of the motion parameter at the current moment, and taking the measured value of the linear distance parameter as the estimated value of the linear distance parameter in the estimated value of the motion parameter at the current moment; the measured value of the angular acceleration parameter in the measured value of the motion parameter at the current moment and the estimated value of the angular acceleration parameter in the estimated value of the motion parameter at the current moment are set as a first preset value, and the measured value of the velocity parameter in the measured value of the motion parameter at the current moment and the estimated value of the velocity parameter in the estimated value of the motion parameter at the current moment are set as a second preset value.
Optionally, when the skeleton point corresponding to the position information of the skeleton point is a trunk skeleton point, and when the current time is not the first time, the first determining unit is further configured to: using the position information of the skeleton point as a measurement value of the motion parameter at the current moment; calculation formula M based on estimated values of motion parameters k ＝MP k-1 + v × t calculating the estimation value of the motion parameter at the current moment; m k Estimate, MP, representing a motion parameter at the current time k-1 Represents the optimized value of the motion parameter at the previous moment, v represents the speed of the bone point, and t represents the time difference between the current moment and the previous moment.
Optionally, when the bone point corresponding to the position information of the bone point is a limb bone point or a head bone point, and when the current time is not the first time, the first determining unit is further configured to: determining a measurement value of the motion parameter at the current moment based on the position information of the skeleton point and the optimized value of the motion parameter at the previous moment; and determining the estimated value of the motion parameter at the current moment based on the optimized value of the motion parameter at the previous moment.
Optionally, the first determining unit is further configured to: calculating the measured value of the angle parameter in the measured value of the motion parameter at the current moment according to the position information of the bone point to be smoothed at the current moment and the position information of the last bone point of the bone point to be smoothed at the current moment; the skeleton point to be smoothed is any one of a limb skeleton point or a head skeleton point; calculating the measurement value of the linear distance parameter in the measurement values of the motion parameters at the current moment according to the position information of the bone point to be smoothed at the current moment and the position information of the last bone point of the bone point to be smoothed at the current moment; calculating the measured value of the angular acceleration parameter in the measured value of the motion parameter at the current moment according to the measured value of the angle parameter and the angle parameter in the optimized value of the motion parameter at the previous moment; and calculating the measured value of the speed parameter in the measured values of the motion parameters at the current moment according to the measured value of the linear distance parameter and the linear distance parameter in the optimized value of the motion parameters at the previous moment.
Optionally, the first determining unit is further configured to: calculating an angle parameter estimation value in the current motion parameter estimation value according to an angle parameter in the last time motion parameter optimization value and an angular acceleration parameter in the last time motion parameter optimization value; calculating an estimation value of the linear distance parameter in the estimation value of the motion parameter at the current moment according to the linear distance parameter in the optimization value of the motion parameter at the previous moment and the speed parameter in the optimization value of the motion parameter at the previous moment; determining an estimated value of the angular acceleration parameter in the estimated value of the motion parameter at the current moment according to the angular acceleration parameter in the optimized value of the motion parameter at the previous moment; and determining the estimated value of the speed parameter in the estimated value of the motion parameter at the current moment according to the speed parameter in the optimized value of the motion parameter at the previous moment.
Optionally, the optimization unit is further configured to: determining a first weight and a second weight according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment; the first weight is the weight corresponding to the measured value of the motion parameter at the current moment, and the second weight is the weight corresponding to the estimated value of the motion parameter at the current moment; and performing weighted calculation on the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment based on the first weight and the second weight to obtain an optimized value of the motion parameter at the current moment.
Optionally, the optimization unit is further configured to: calculating a difference value between the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment; calculating a second weight according to a preset weight calculation function and the difference value; calculating a first weight based on the second weight; when the difference is smaller than a first preset threshold, or when the difference is larger than a second preset threshold, the first weight is positively correlated with the difference, the second weight is negatively correlated with the difference, and the first preset threshold is smaller than the second preset threshold.
Optionally, when the skeleton point corresponding to the position information of the skeleton point is a trunk skeleton point, the second determining unit is further configured to: and taking the optimized value of the motion parameter at the current moment as the position information of the bone point after the smoothing processing at the current moment.
Optionally, when the bone point corresponding to the position information of the bone point is an extremity bone point or a head bone point, the second determining unit is further configured to: and converting the optimized value of the motion parameter at the current moment into a Cartesian coordinate system to obtain the position information of the bone point after the smooth processing at the current moment.
The device provided by the embodiment of the present invention has the same implementation principle and the same technical effects as those of the foregoing method embodiments, and for the sake of brief description, reference may be made to corresponding contents in the foregoing method embodiments for the parts of the device embodiments that are not mentioned.
In another implementation of the present invention, there is further provided a computer storage medium having a computer program stored thereon, the computer program, when executed by a computer, performing the steps of the method of any one of the above method embodiments 2.
In addition, in the description of the embodiments of the present invention, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in a specific case to those of ordinary skill in the art.
In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of description and simplicity of description, but do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in software functional units and sold or used as a stand-alone product, may be stored in a non-transitory computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of the present invention or a part thereof which substantially contributes to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present invention, which are used for illustrating the technical solutions of the present invention and not for limiting the same, and the protection scope of the present invention is not limited thereto, although the present invention is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present invention, and they should be construed as being included therein. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (17)

1. A method of smoothing bone points, comprising:
acquiring position information of a current skeleton point of a target object and an optimized value of a motion parameter of a previous moment; the motion parameters comprise the relative position relation between skeleton points;
determining a measurement value of the motion parameter at the current moment and an estimation value of the motion parameter at the current moment based on the position information of the skeleton point and the optimized value of the motion parameter at the previous moment;
determining a first weight and a second weight according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment; the first weight is the weight corresponding to the measured value of the motion parameter at the current moment, and the second weight is the weight corresponding to the estimated value of the motion parameter at the current moment;
performing weighted calculation on the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment based on the first weight and the second weight to obtain an optimized value of the motion parameter at the current moment;
and determining the position information of the skeleton point after the smoothing treatment at the current moment based on the optimized value of the motion parameter at the current moment.
2. The method of claim 1, wherein the step of obtaining the position information of the bone point of the target object at the current time comprises:
and detecting the image of the target object at the current moment by adopting a skeleton point detection model to obtain the position information of the skeleton point of the target object at the current moment.
3. The method of claim 1,
when the skeleton point corresponding to the position information of the skeleton point is a trunk skeleton point, the motion parameter is represented by the position information of the skeleton point;
and when the skeleton point corresponding to the position information of the skeleton point is a limb skeleton point or a head skeleton point, the motion parameter is represented by an angle parameter, an angular acceleration parameter, a linear distance parameter and a speed parameter.
4. The method according to claim 1, wherein when the skeleton point corresponding to the position information of the skeleton point is a trunk skeleton point and when the current time is a first time, the optimized value of the motion parameter at the previous time does not exist, the measured value of the motion parameter at the current time and the estimated value of the motion parameter at the current time are determined based on the position information of the skeleton point.
5. The method of claim 4, wherein the step of determining the measured value of the motion parameter at the current time and the estimated value of the motion parameter at the current time based on the position information of the bone points comprises:
and respectively taking the position information of the skeleton point as a measured value of the motion parameter at the current moment and an estimated value of the motion parameter at the current moment.
6. The method according to claim 1, wherein when the bone point corresponding to the position information of the bone point is a limb bone point or a head bone point, and when the current time is a first time, the optimal value of the motion parameter at the previous time does not exist, the measured value of the motion parameter at the current time and the estimated value of the motion parameter at the current time are determined based on the position information of the bone point.
7. The method of claim 6, wherein the step of determining the measured value of the motion parameter at the current time and the estimated value of the motion parameter at the current time based on the position information of the bone points comprises:
calculating a measurement value of an angle parameter in the measurement values of the motion parameters at the current moment and a measurement value of a linear distance parameter in the measurement values of the motion parameters at the current moment based on the position information of the bone points;
taking the measured value of the angle parameter as an estimated value of the angle parameter in the estimated value of the motion parameter at the current moment, and taking the measured value of the linear distance parameter as an estimated value of the linear distance parameter in the estimated value of the motion parameter at the current moment;
setting the measured value of the angular acceleration parameter in the measured value of the motion parameter at the current moment and the estimated value of the angular acceleration parameter in the estimated value of the motion parameter at the current moment as a first preset value, and setting the measured value of the velocity parameter in the measured value of the motion parameter at the current moment and the estimated value of the velocity parameter in the estimated value of the motion parameter at the current moment as a second preset value.
8. The method according to claim 1, wherein the step of determining the measured value of the motion parameter at the current time and the estimated value of the motion parameter at the current time based on the position information of the skeleton point and the optimized value of the motion parameter at the previous time when the skeleton point corresponding to the position information of the skeleton point is the trunk skeleton point comprises:
using the position information of the skeleton point as a measurement value of the motion parameter at the current moment;
calculation formula M based on estimated values of motion parameters k ＝MP k-1 + v × t calculating an estimated value of the motion parameter at the current moment; m k An estimated value, MP, representing said motion parameter at the current moment k-1 Represents the optimized value of the motion parameter at the previous moment, v represents the speed of the bone point, and t represents the time difference between the current moment and the previous moment.
9. The method according to claim 1, wherein the step of determining the measured value of the motion parameter at the current time and the estimated value of the motion parameter at the current time based on the position information of the bone point and the optimized value of the motion parameter at the previous time when the bone point corresponding to the position information of the bone point is a limb bone point or a head bone point comprises:
determining a measurement value of the motion parameter at the current moment based on the position information of the skeleton point and the optimized value of the motion parameter at the last moment;
and determining the estimated value of the motion parameter at the current moment based on the optimized value of the motion parameter at the previous moment.
10. The method of claim 9, wherein the step of determining the measured value of the motion parameter at the current time based on the position information of the bone point and the optimized value of the motion parameter at the previous time comprises:
calculating the measured value of the angle parameter in the measured value of the motion parameter at the current moment according to the position information of the bone point to be smoothed at the current moment and the position information of the last bone point of the bone point to be smoothed at the current moment; the bone point to be smoothed is any one of the limb bone point or the head bone point;
calculating the measurement value of the linear distance parameter in the measurement values of the motion parameters at the current moment according to the position information of the bone point to be smoothed at the current moment and the position information of the last bone point of the bone point to be smoothed at the current moment;
calculating the measured value of the angular acceleration parameter in the measured value of the motion parameter at the current moment according to the measured value of the angle parameter and the angle parameter in the optimized value of the motion parameter at the previous moment;
and calculating the measured value of the speed parameter in the measured values of the motion parameters at the current moment according to the measured value of the linear distance parameter and the linear distance parameter in the optimized value of the motion parameter at the previous moment.
11. The method of claim 9, wherein the step of determining the estimated value of the motion parameter at the current time based on the optimized value of the motion parameter at the previous time comprises:
calculating an angle parameter estimation value in the current moment motion parameter estimation value according to the angle parameter in the last moment motion parameter optimization value and the angular acceleration parameter in the last moment motion parameter optimization value;
calculating an estimated value of a linear distance parameter in the estimated value of the motion parameter at the current moment according to the linear distance parameter in the optimized value of the motion parameter at the previous moment and the speed parameter in the optimized value of the motion parameter at the previous moment;
determining an estimated value of the angular acceleration parameter in the estimated value of the motion parameter at the current moment according to the angular acceleration parameter in the optimized value of the motion parameter at the previous moment;
and determining the estimated value of the speed parameter in the estimated value of the motion parameter at the current moment according to the speed parameter in the optimized value of the motion parameter at the previous moment.
12. The method of claim 1, wherein the step of determining a first weight and a second weight based on the measured value of the motion parameter at the current time and the estimated value of the motion parameter at the current time comprises:
calculating a difference value between the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment;
calculating the second weight according to a preset weight calculation function and the difference value;
calculating the first weight based on the second weight; when the difference is smaller than a first preset threshold, or when the difference is larger than a second preset threshold, the first weight is positively correlated with the difference, the second weight is negatively correlated with the difference, and the first preset threshold is smaller than the second preset threshold.
13. The method according to claim 1, wherein when the skeleton point corresponding to the position information of the skeleton point is a trunk skeleton point, the step of determining the position information of the skeleton point after the smoothing processing at the current time based on the optimized value of the motion parameter at the current time comprises:
and taking the optimized value of the motion parameter at the current moment as the position information of the bone point after the smoothing processing at the current moment.
14. The method according to claim 1, wherein when the bone point corresponding to the position information of the bone point is a limb bone point or a head bone point, the step of determining the position information of the bone point after the smoothing processing at the current time based on the optimized value of the motion parameter at the current time comprises:
and converting the optimized value of the motion parameter at the current moment into a Cartesian coordinate system to obtain the position information of the bone point smoothed at the current moment.
15. A device for smoothing bone points, comprising:
the acquisition unit is used for acquiring the position information of the current skeleton point of the target object and the optimized value of the motion parameter at the previous moment; the motion parameters comprise the relative position relation between skeleton points;
the first determination unit is used for determining a measurement value of the motion parameter at the current moment and an estimation value of the motion parameter at the current moment based on the position information of the bone point and the optimized value of the motion parameter at the previous moment;
the optimization unit is used for determining a first weight and a second weight according to the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment; the first weight is the weight corresponding to the measured value of the motion parameter at the current moment, and the second weight is the weight corresponding to the estimated value of the motion parameter at the current moment; performing weighted calculation on the measured value of the motion parameter at the current moment and the estimated value of the motion parameter at the current moment based on the first weight and the second weight to obtain an optimized value of the motion parameter at the current moment;
and the second determining unit is used for determining the position information of the bone point subjected to smoothing processing at the current moment based on the optimized value of the motion parameter at the current moment.
16. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any of the preceding claims 1 to 14 when executing the computer program.
17. A computer storage medium, having a computer program stored thereon, which when executed by a computer performs the steps of the method of any of claims 1 to 14.
CN201911111739.9A 2019-11-13 2019-11-13 Skeleton point smoothing method and device and electronic equipment Active CN110827226B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911111739.9A CN110827226B (en) 2019-11-13 2019-11-13 Skeleton point smoothing method and device and electronic equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911111739.9A CN110827226B (en) 2019-11-13 2019-11-13 Skeleton point smoothing method and device and electronic equipment

Publications (2)

Publication Number Publication Date
CN110827226A CN110827226A (en) 2020-02-21
CN110827226B true CN110827226B (en) 2022-09-27

Family

ID=69555061

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911111739.9A Active CN110827226B (en) 2019-11-13 2019-11-13 Skeleton point smoothing method and device and electronic equipment

Country Status (1)

Country Link
CN (1) CN110827226B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112686208B (en) * 2021-01-22 2022-11-08 上海喵眼智能科技有限公司 Motion recognition characteristic parameter algorithm based on machine vision

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104123007A (en) * 2014-07-29 2014-10-29 电子科技大学 Multidimensional weighted 3D recognition method for dynamic gestures
WO2017156741A1 (en) * 2016-03-17 2017-09-21 深圳多哚新技术有限责任公司 Head motion compensation method and associated device
CN108304790A (en) * 2018-01-19 2018-07-20 腾讯科技（深圳）有限公司 Skeleton motion prediction processing method, device and limb motion prediction processing method
CN108536292A (en) * 2018-03-29 2018-09-14 深圳市芯汉感知技术有限公司 A kind of data filtering methods and bone point coordinates accurate positioning method
CN109077731A (en) * 2018-06-11 2018-12-25 上海大学 A kind of human body detecting method and device of view-based access control model

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104123007A (en) * 2014-07-29 2014-10-29 电子科技大学 Multidimensional weighted 3D recognition method for dynamic gestures
WO2017156741A1 (en) * 2016-03-17 2017-09-21 深圳多哚新技术有限责任公司 Head motion compensation method and associated device
CN108304790A (en) * 2018-01-19 2018-07-20 腾讯科技（深圳）有限公司 Skeleton motion prediction processing method, device and limb motion prediction processing method
CN108536292A (en) * 2018-03-29 2018-09-14 深圳市芯汉感知技术有限公司 A kind of data filtering methods and bone point coordinates accurate positioning method
CN109077731A (en) * 2018-06-11 2018-12-25 上海大学 A kind of human body detecting method and device of view-based access control model

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Image noise smoothing using a modified Kalman filter;Jian Pan et al;《Neurocomputing》;20150926;全文 *
结合金字塔模型和随机森林的运动捕获序列语义标注;徐濛等;《华侨大学学报》;20171130;摘要、第1-3节 *

Also Published As

Publication number Publication date
CN110827226A (en) 2020-02-21

Similar Documents

Publication Publication Date Title
JP6871416B2 (en) Methods and devices for determining facial image quality, electronics and computer storage media
JP7151488B2 (en) Moving object detection device, moving object detection method and program
CN107030699A (en) Position and attitude error modification method and device, robot and storage medium
WO2019228196A1 (en) Method for tracking target in panoramic video, and panoramic camera
CN108875526B (en) Method, device and system for line-of-sight detection and computer storage medium
EP3413555B1 (en) Image processing method and device as well as computer program and non-transitory computer-readable medium
JP7067023B2 (en) Information processing device, background update method and background update program
EP3223119A1 (en) Method and device for adjusting object attribute information
CN109640224B (en) Pickup method and device
JP5530399B2 (en) Image processing method, image processing apparatus, and image processing program
JP7129169B2 (en) Image processing device, radiation imaging device, image processing method, and program
US20210021763A1 (en) Photographing control method and controller
CN115147558A (en) Training method of three-dimensional reconstruction model, three-dimensional reconstruction method and device
JP5787456B2 (en) Image processing device
CN110827226B (en) Skeleton point smoothing method and device and electronic equipment
CN110517201B (en) Method and device for smoothing filtering along environment-friendly edge and electronic equipment
JP6100064B2 (en) Electronic apparatus and image processing method
WO2017204120A1 (en) Image processing apparatus, image processing method, and program
AU2010338191B2 (en) Stabilisation method and computer system
CN108876704B (en) Method and device for deforming human face image and computer storage medium
US10621788B1 (en) Reconstructing three-dimensional (3D) human body model based on depth points-to-3D human body model surface distance
CN113542597B (en) Focusing method and electronic device
JP6381092B2 (en) Method and apparatus for estimating the value of an input in the presence of a perturbation factor
CN114063761A (en) Gaze point display method and related device
JP7468515B2 (en) Information processing device, information processing method, and program

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant