Spine motion capture method, apparatus, electronic device, and storage medium

Abstract

The present application provides a kind of vertebral motion capture method, device, electronic equipment and storage medium, by being divided into multiple bone segments with continuous bone segment chain, optical marker point is set on the starting end bone segment of continuous bone segment chain and the shared parent bone segment, the posture information of starting end bone segment and shared parent bone segment under world coordinate system is acquired based on three-dimensional coordinate data collected by optical marker point, and the initial posture information of at least one shared child bone segment between starting end bone segment and the shared parent bone segment is calculated;Optimization problem objective function is constructed to cooperatively optimize the posture information of each bone segment in continuous bone segment chain, and the optimized posture information of all bone segments in continuous bone segment chain is obtained, the present application can reduce specific bone segment paste point, so as to quickly construct motion capture system, by cooperative optimization mechanism, only need to set optical marker in the start and end of spinal column chain, key point, high-precision motion posture of each bone segment of whole spinal column can be robustly and efficiently reconstructed.

Description

Technical Field

[0001] This invention relates to the field of computer vision technology, and in particular to a method, apparatus, electronic device, and storage medium for capturing spinal motion. Background Technology

[0002] With the rapid development of virtual reality, film and television special effects, sports science analysis, and medical rehabilitation, the demand for high-precision, real-time motion capture of humans or animals is increasing. Optical motion capture technology, due to its high precision and non-contact characteristics, has become the mainstream motion capture solution. In the process of skeletal motion reconstruction based on optical markers, reflective markers are usually attached to key anatomical locations of various joints and bone segments of the subject. The three-dimensional spatial coordinates of these markers are captured by multiple high-speed infrared cameras, and then the six degrees of freedom (i.e., three-dimensional spatial position and three-dimensional orientation) information of each bone segment is calculated based on a preset human skeletal model and kinematic principles, thereby reconstructing the complete skeletal motion. However, for complex structures such as the spine, which consists of multiple vertebrae and has continuous, multi-segmental characteristics, attaching and tracking a set of markers independently for each bone segment would lead to a surge in the number of markers. This not only consumes a lot of manpower and time in the preparation work before data acquisition (marking and calibration), but also makes the entire system construction process extremely cumbersome. Furthermore, a large number of markers will generate massive amounts of three-dimensional coordinate data. When calculating the degrees of freedom for each bone segment, the system requires complex matrix operations and iterative optimization. The increase in the number of marker points directly leads to an exponential increase in computational complexity, severely impacting data processing speed and making it difficult to meet the application requirements of real-time motion capture and feedback. Dense marker points are prone to data loss or confusion during movement due to body occlusion, marker adhesion, or blurring caused by rapid motion. Especially during extreme movements such as large-angle bending and twisting of the spine, traditional methods relying on a large number of independent marker points are prone to overall reconstruction failure or significant cumulative errors due to the failure of some data, resulting in poor system stability. Summary of the Invention

[0003] This invention provides a spinal motion capture method, device, electronic device, and storage medium to address the shortcomings of existing optical motion capture methods when processing multi-segment continuous structures such as the spine. These methods require independent pasting and calculation of marker points for each bone segment, resulting in complex system construction, low computational efficiency, and overall reconstruction failure or large cumulative errors when motion occlusion or large-angle bending occurs.

[0004] This invention provides a method for capturing spinal motion, comprising: The continuous bone segment chain is divided into multiple bone segments, and from the multiple bone segments, a starting bone segment, a shared parent bone segment, and a shared child bone segment located between the starting bone segment and the shared parent bone segment are determined; Optical markers are set on the starting bone segment and the shared parent bone segment of the continuous bone segment chain. The attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system is obtained based on the three-dimensional coordinate data collected by the optical markers. Based on the orientation information of the starting bone segment and the shared parent bone segment in the world coordinate system, calculate the initial orientation information of at least one shared child bone segment located between the starting bone segment and the shared parent bone segment. Using the initial pose information of the shared sub-bone segment as the initial value, an optimization problem objective function is constructed to collaboratively optimize the pose information of each bone segment in the continuous bone segment chain, thereby obtaining the optimized pose information of all bone segments in the continuous bone segment chain.

[0005] According to the spinal motion capture method provided by the present invention, the step of determining the initial bone segment, the shared parent bone segment, and the shared child bone segment located between the initial bone segment and the shared parent bone segment from the plurality of bone segments includes: The bone segment with six degrees of freedom in the continuous bone segment chain, which serves as the kinematic root node of the entire chain structure, is defined as the starting bone segment. The bone segment in the continuous bone segment chain that has independent rotational degrees of freedom and is used to provide a rotational reference for downstream bone segments is defined as the shared parent bone segment; A bone segment located between the starting bone segment and the shared parent bone segment, whose rotational degrees of freedom are configured to be inherited from the shared parent bone segment, is defined as a shared child bone segment.

[0006] According to the spinal motion capture method provided by the present invention, the step of calculating the initial posture information of at least one shared child bone segment located between the initial bone segment and the shared parent bone segment based on the posture information of the initial bone segment and the shared parent bone segment in the world coordinate system includes: Spherical linear interpolation is performed on the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system to generate the rotational attitude of the at least one shared child bone segment, which serves as the initial attitude information of the at least one shared child bone segment. The interpolation ratio coefficient of the spherical linear interpolation is dynamically adjusted according to the overall bending angle between the starting bone segment and the shared parent bone segment.

[0007] According to the spinal motion capture method provided by the present invention, the interpolation ratio coefficient of the spherical linear interpolation is dynamically adjusted based on the overall bending angle between the initial bone segment and the shared parent bone segment, including: When the overall bending angle is within the first angle range, the interpolation ratio coefficient is calculated according to the first function; When the overall bending angle is within the second angle range, the interpolation ratio coefficient is calculated according to the second function, which includes additional terms related to the angle. When the overall bending angle is within the third angle range, the interpolation ratio coefficient is calculated according to the third function; Wherein, the upper limit of the first angle interval is less than the lower limit of the second angle interval, and the upper limit of the second angle interval is less than the lower limit of the third angle interval.

[0008] According to the spinal motion capture method provided by the present invention, the objective function of the optimization problem includes: The position error term includes the difference between the observed position based on the optical markers on the shared parent bone segment and the position calculated based on the bone segment pose in the optimization. The length constraint error term includes the difference between the total model length based on the optimized continuous bone segment chain and the preset total anatomical length.

[0009] According to the spinal motion capture method provided by the present invention, the step of collaboratively optimizing the posture information of each bone segment in the continuous bone segment chain includes: The Levenberg-Marquardt algorithm is used to iteratively solve the objective function of the optimization problem in order to optimize the pose information of each bone segment in the continuous bone segment chain, including the shared parent bone segment and the shared child bone segment; The iterative solution process terminates when at least one of the following convergence conditions is met: The positional error change of the shared parent bone segment is less than a first preset threshold. The number of iterations has reached the preset maximum number of iterations; The position error calculated in the first iteration is less than the second preset threshold.

[0010] According to the spinal motion capture method provided by the present invention, the iterative solution process further includes: Dynamically adjusted weighting coefficients are applied to the position error term and the length constraint error term; In the initial stage of iteration, the weight of the position error term is greater than the weight of the length constraint error term; as the iteration progresses, the weight of the length constraint error term is gradually increased.

[0011] The present invention also provides a spinal motion capture device, comprising: A definition module is used to divide a continuous bone segment chain into multiple bone segments, and to determine the starting bone segment, the shared parent bone segment, and the shared child bone segment located between the starting bone segment and the shared parent bone segment from the multiple bone segments; The acquisition module is used to set optical markers on the starting bone segment and the shared parent bone segment of the continuous bone segment chain, and to acquire the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system based on the three-dimensional coordinate data collected by the optical markers. The calculation module is used to calculate the initial attitude information of at least one shared child bone segment located between the starting bone segment and the shared parent bone segment based on the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system. The optimization module is used to construct an optimization problem objective function based on the initial posture information of the shared sub-bone segment as the initial value, and to collaboratively optimize the posture information of each bone segment in the continuous bone segment chain to obtain the optimized posture information of all bone segments in the continuous bone segment chain.

[0012] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the spinal motion capture method as described in any of the preceding claims.

[0013] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the spinal motion capture method described in any of the preceding claims.

[0014] The spinal motion capture method, apparatus, electronic device, and storage medium provided by this invention divide a continuous bone segment chain into multiple bone segments, and determine a starting bone segment, a shared parent bone segment, and a shared child bone segment located between the starting bone segment and the shared parent bone segment from the multiple bone segments; optical markers are set on the starting bone segment and the shared parent bone segment of the continuous bone segment chain, and the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system is obtained based on the three-dimensional coordinate data collected by the optical markers; based on the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system, the initial attitude information of at least one shared child bone segment located between the starting bone segment and the shared parent bone segment is calculated; and so on. The initial pose information of the shared sub-segment is used as the initial value. An optimization problem objective function is constructed to collaboratively optimize the pose information of each segment in the continuous bone segment chain, thereby obtaining the optimized pose information of all segments in the continuous bone segment chain. This invention can reduce the number of specific bone segment attachment points by setting some special bone segments as shared parent bone segments and some special bone segments as shared child bone segments, thus enabling rapid construction of motion capture systems. Through the collaborative optimization mechanism, only optical markers need to be set at the start and end points and key points of the spinal chain to robustly and efficiently reconstruct the high-precision motion pose of each bone segment of the entire spine. This fundamentally solves the technical problems of traditional methods, such as cumbersome system construction, large computational load, and easy failure under motion occlusion caused by excessively dense marker points. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0016] Figure 1 This is one of the flowcharts of the spinal motion capture method provided in the embodiments of the present invention; Figure 2 This is a schematic diagram of the human skeleton model provided in an embodiment of the present invention; Figure 3 This is the second flowchart of the spinal motion capture method provided in the embodiments of the present invention; Figure 4 This is one of the schematic diagrams illustrating the changes in bone position when the human body bends, provided in an embodiment of the present invention; Figure 5 This is the second schematic diagram of the changes in the position of bones when the human body bends, provided in the embodiments of the present invention; Figure 6 This is a functional structure diagram of the spinal motion capture device provided in an embodiment of the present invention; Figure 7 This is a functional structure diagram of the electronic device provided in the embodiments of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0018] Figure 1 A flowchart of the spinal motion capture method provided in the embodiments of the present invention is shown below. Figure 1 As shown, the spinal motion capture method provided in this embodiment of the invention includes: Step 101: Divide the continuous bone segment chain into multiple bone segments, and determine the starting bone segment, the shared parent bone segment, and the shared child bone segment located between the starting bone segment and the shared parent bone segment from the multiple bone segments; Step 102: Set optical markers on the starting bone segment and the shared parent bone segment of the continuous bone segment chain, and obtain the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system based on the three-dimensional coordinate data collected by the optical markers. Step 103: Based on the orientation information of the starting bone segment and the shared parent bone segment in the world coordinate system, calculate the initial orientation information of at least one shared child bone segment located between the starting bone segment and the shared parent bone segment; Step 104: Using the initial posture information of the shared sub-bone segment as the initial value, construct an optimization problem objective function to collaboratively optimize the posture information of each bone segment in the continuous bone segment chain, and obtain the optimized posture information of all bone segments in the continuous bone segment chain.

[0019] Traditional skeletal motion reconstruction processes based on optical markers typically require attaching reflective markers to key anatomical locations at various joints and bone segments of the subject. This not only consumes significant manpower and time in the preparatory work before data acquisition (marking and calibration), but also makes the entire system construction process extremely cumbersome. Furthermore, calculating the degrees of freedom for each bone segment requires complex matrix operations and iterative optimization. Dense markers are prone to data loss or confusion during movement due to body occlusion, marker adhesion, or blurring caused by rapid movement. Especially during extreme movements such as large-angle bending and twisting of the spine, traditional methods relying on a large number of independent markers are prone to overall reconstruction failure or large cumulative errors due to the failure of some data, resulting in poor system stability.

[0020] The spinal motion capture method provided in this embodiment of the invention divides a continuous bone segment chain into multiple segments, and determines a starting bone segment, a shared parent bone segment, and a shared child bone segment located between the starting bone segment and the shared parent bone segment from the multiple bone segments; optical markers are set on the starting bone segment and the shared parent bone segment of the continuous bone segment chain, and the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system is obtained based on the three-dimensional coordinate data collected by the optical markers; based on the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system, the initial attitude information of at least one shared child bone segment located between the starting bone segment and the shared parent bone segment is calculated; and the initial attitude information of the shared child bone segment is calculated based on the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system. The initial posture information is used as the initial value. An optimization problem objective function is constructed to collaboratively optimize the posture information of each bone segment in the continuous bone segment chain, thereby obtaining the optimized posture information of all bone segments in the continuous bone segment chain. In this embodiment of the invention, some special bone segments can be set as shared parent bone segments and some special bone segments can be set as shared child bone segments, reducing the number of specific bone segment attachment points. This allows for rapid construction of a motion capture system. Through the collaborative optimization mechanism, only optical markers need to be set at the start and end points and key points of the spinal chain to robustly and efficiently reconstruct the high-precision motion posture of each bone segment of the entire spine. This fundamentally solves the technical problems of traditional methods, such as cumbersome system construction, large computational load, and easy failure under motion occlusion caused by excessively dense marker points.

[0021] Based on any of the above embodiments, the step of determining the initial bone segment, the shared parent bone segment, and the shared child bone segment located between the initial bone segment and the shared parent bone segment from the plurality of bone segments includes: The bone segment with six degrees of freedom in the continuous bone segment chain, which serves as the kinematic root node of the entire chain structure, is defined as the starting bone segment. The bone segment in the continuous bone segment chain that has independent rotational degrees of freedom and is used to provide a rotational reference for downstream bone segments is defined as the shared parent bone segment; A bone segment located between the starting bone segment and the shared parent bone segment, whose rotational degrees of freedom are configured to be inherited from the shared parent bone segment, is defined as a shared child bone segment.

[0022] Since the human spine is composed of multiple bone segments, and according to the principles of human kinesiology, the movement between these segments is continuous. Therefore, the spine can be divided into four segments: Spin 0, Spin 1, Spin 2, and Spin 3. Figure 2 As shown, we define pelvis (parent bone segment, 6 degrees of freedom), spin3 (shared parent bone segment, 3 degrees of freedom), and spin0-2 (shared child bone segments, inheriting rotational degrees of freedom). Six degrees of freedom represent displacement in the x, y, and z directions and rotation about the x, y, and z directions respectively. Three degrees of freedom represent rotation about the x, y, and z directions respectively. Spin3 is defined as the shared parent bone segment, and spin0, spin1, and spin2 are defined as shared child bone segments with three degrees of freedom each. The shared child bone segments spin0, spin1, and spin2 have the same rotational information as the parent bone segment spin3.

[0023] Based on any of the above embodiments, calculating the initial pose information of at least one shared child bone segment located between the initial bone segment and the shared parent bone segment based on the pose information of the initial bone segment and the shared parent bone segment in the world coordinate system includes: Spherical linear interpolation (TSlerp) is performed on the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system to generate the rotational attitude of the at least one shared sub-bone segment, which serves as the initial attitude information of the at least one shared sub-bone segment. The interpolation ratio coefficient of the spherical linear interpolation is dynamically adjusted according to the overall bending angle between the starting bone segment and the shared parent bone segment.

[0024] In this embodiment of the invention, the interpolation ratio coefficient of the spherical linear interpolation is dynamically adjusted according to the overall bending angle between the starting bone segment and the shared parent bone segment, including: When the overall bending angle is within the first angle range, the interpolation ratio coefficient is calculated according to the first function; When the overall bending angle is within the second angle range, the interpolation ratio coefficient is calculated according to the second function, which includes additional terms related to the angle. When the overall bending angle is within the third angle range, the interpolation ratio coefficient is calculated according to the third function; Wherein, the upper limit of the first angle interval is less than the lower limit of the second angle interval, and the upper limit of the second angle interval is less than the lower limit of the third angle interval.

[0025] Based on any of the above embodiments, the objective function of the optimization problem includes: The position error term includes the difference between the observed position based on the optical markers on the shared parent bone segment and the position calculated based on the bone segment pose in the optimization. The length constraint error term includes the difference between the total model length based on the optimized continuous bone segment chain and the preset total anatomical length.

[0026] Based on any of the above embodiments, the collaborative optimization of the pose information of each bone segment in the continuous bone segment chain includes: The Levenberg-Marquardt algorithm is used to iteratively solve the objective function of the optimization problem in order to optimize the pose information of each bone segment in the continuous bone segment chain, including the shared parent bone segment and the shared child bone segment; The iterative solution process terminates when at least one of the following convergence conditions is met: The positional error change of the shared parent bone segment is less than a first preset threshold. The number of iterations has reached the preset maximum number of iterations; The position error calculated in the first iteration is less than the second preset threshold.

[0027] In this embodiment of the invention, the first preset threshold is, for example, 1 mm, the maximum number of iterations is, for example, 5 times, and the second preset threshold is, for example, 3 mm.

[0028] In this embodiment of the invention, the iterative solution process further includes: Dynamically adjusted weighting coefficients are applied to the position error term and the length constraint error term; In the initial stage of iteration, the weight of the position error term is greater than the weight of the length constraint error term; as the iteration progresses, the weight of the length constraint error term is gradually increased.

[0029] This invention takes a game character spinal curvature capture implementation as an example to illustrate the following: Figure 2The degrees of freedom information of the shared sub-skeletal segments spin0, spin1, and spin2 are used for calculation.

[0030] like Figure 3 As shown, the spinal motion capture method provided in this embodiment of the invention specifically includes: (1) Place reflective markers on the bone segments defined by Pelvis and Spin3 on the body surface and obtain 3D coordinates through the calibrated motion capture system; (2) Based on the preset human skeleton model, calculate the local coordinates of the above reflective markers on the human skeleton, as well as the skeleton length; (3) Obtain the 3D coordinate data of the reflective markers on the bone segment through the motion capture system, as well as the known local coordinates of the reflective markers on the human skeleton. Solve the rotation information of the skeleton by iterating the nearest point and denote the bone position of spine3 as P0.

[0031] (4) Perform spherical linear interpolation on the rotation of pelvis and spine3.

[0032] Let θ be the angle between the vector from the origin O to the second end E of the target sub-bone segment and the vector from the origin O to the parent bone segment. Degrees of freedom Where O represents the origin of the target coordinate system, A represents the degree of freedom of the parent bone segment (i.e., the position of the first end of the target object), and E represents the degree of freedom of the target child bone segment (i.e., the position of the second end of the target object). (Vector) This indicates the position (i.e., degree of freedom) of the interpolation on the arc AE. This represents the vector from the origin O to the second end E of the target sub-bone segment. Let t represent the vector from the origin O to the parent bone segment, and t represent the scaling factor, which has a value greater than or equal to 0 and less than or equal to 1.

[0033] In addition, regarding the angle of bone segment curvature The scaling factor can be adjusted to more flexibly adapt to different bending angle scenarios.

[0034] in The sensitivity coefficients for different intervals need to be calibrated experimentally (e.g. =0.1, =0.3, =0.5); Additionally, add an angle-dependent term (e.g., This prevents abrupt changes in step size. For example, when θ=30°, t=0.2.

[0035] (5) Calculate the rotational information of each segment of the spine relative to its parent segment based on the bone lengths, and recalculate the position of the bone segment spine3 under the condition of satisfying the shared degrees of freedom. At this time and do not coincide exactly, especially when the human body is bent, as shown in Figure 4 , 5 . The line segment AE is the length of the entire spine. At this time, spin0, spin1, spin2 calculated by the above interpolation, and the line segment AB + BC + CD + DE is the true spine length. At this time, there is + BC + CD + DE < AE. In order to make and coincide as much as possible and reflect the true position of the spine spine3, use the Levenberg-Marquardt algorithm to iteratively calculate and optimize the rotational information of each spine to make and coincide as much as possible.

[0036] The specific steps of using the Levenberg-Marquardt algorithm to iteratively calculate and optimize the rotational information of each spine are as follows: (1) Use the initial pose information of the at least one shared sub-bone segment and the first pose information of the shared parent bone segment as the initial values of the optimization variables; (2) In the current iteration, according to the current values of the optimization variables, calculate the position error between the model position P1 of the shared parent bone segment and the observed position P0 determined based on its optical marker points, and the length error between the current total model length of the continuous bone segment chain and the preset anatomical total length; (3) Based on the position error and the length error, construct the increment equation for the current iteration, and solve to obtain the increment of the optimization variables; (4) Use the increment to update the optimization variables to obtain the new pose information of each bone segment; (5) Determine whether the preset convergence condition is satisfied; if satisfied, output the current pose information of each bone segment as the optimization result; if not satisfied, return to step (2) for the next iteration.

[0037] Define the position error ( ) and the bone segment length constraint as the objective function In addition, dynamically adjust the weights. In the initial iteration stage (the first 2 times), focus on the position error ( = 0.1); In subsequent iterations, gradually strengthen the length constraint ( →1.0). The convergence condition is defined as the rate of change of position error < 1%, i.e. To prevent overoptimization, the maximum number of iterations is defined as ≤ 5. Furthermore, when the initial iteration error... Directly outputting results shortens the overall optimization time.

[0038] In this embodiment of the invention, spinal motion capture includes, but is not limited to, the calculation of the degrees of freedom of the human spine, the calculation of the degrees of freedom of finger joints and the spine of quadrupedal animals.

[0039] The spinal motion capture method provided by this invention, through a "shared degree of freedom" skeletal model, only requires setting optical markers at the key bone segments at the beginning and end of the kinetic chain (the starting end and the shared parent bone segment). Dynamic spherical interpolation and a collaborative optimization algorithm can robustly and efficiently reconstruct the six-degree-of-freedom pose information of all bone segments, solving the problem caused by excessively dense markers. This saves significant manpower and time costs associated with marker placement and calibration, improving usability and scalability. By reducing the number of markers to be calculated and employing an intelligent convergence strategy, high-precision real-time motion capture is achieved, meeting the needs of interactive applications. Even in extreme motion scenarios such as large-angle bending, twisting, or marker occlusion, stable and biomechanically sound motion data can still be output through model constraints.

[0040] The spinal motion capture method provided by the present invention is described below. The spinal motion capture method described below can be referred to in correspondence with the spinal motion capture method described above.

[0041] Figure 6 This is a functional structural diagram of the spinal motion capture device provided in an embodiment of the present invention, as shown below. Figure 6 As shown, the spinal motion capture device provided in this embodiment of the invention includes: The definition module 601 is used to divide a continuous bone segment chain into multiple bone segments, and to determine the starting bone segment, the shared parent bone segment, and the shared child bone segment located between the starting bone segment and the shared parent bone segment from the multiple bone segments. The acquisition module 602 is used to set optical markers on the starting bone segment and the shared parent bone segment of the continuous bone segment chain, and to acquire the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system based on the three-dimensional coordinate data collected by the optical markers. The calculation module 603 is used to calculate the initial posture information of at least one shared child bone segment located between the starting bone segment and the shared parent bone segment based on the posture information of the starting bone segment and the shared parent bone segment in the world coordinate system. The optimization module 604 is used to construct an optimization problem objective function with the initial posture information of the shared sub-bone segment as the initial value, and to perform collaborative optimization on the posture information of each bone segment in the continuous bone segment chain, so as to obtain the optimized posture information of all bone segments in the continuous bone segment chain.

[0042] The spinal motion capture device provided in this embodiment of the invention divides a continuous bone segment chain into multiple segments, and determines a starting bone segment, a shared parent bone segment, and a shared child bone segment located between the starting bone segment and the shared parent bone segment from the multiple bone segments; optical markers are set on the starting bone segment and the shared parent bone segment of the continuous bone segment chain, and the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system is obtained based on the three-dimensional coordinate data collected by the optical markers; based on the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system, the initial attitude information of at least one shared child bone segment located between the starting bone segment and the shared parent bone segment is calculated; and the initial attitude information of the shared child bone segment is calculated based on the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system. The initial posture information is used as the initial value. An optimization problem objective function is constructed to collaboratively optimize the posture information of each bone segment in the continuous bone segment chain, thereby obtaining the optimized posture information of all bone segments in the continuous bone segment chain. In this embodiment of the invention, some special bone segments can be set as shared parent bone segments and some special bone segments can be set as shared child bone segments, reducing the number of specific bone segment attachment points. This allows for rapid construction of a motion capture system. Through the collaborative optimization mechanism, only optical markers need to be set at the start and end points and key points of the spinal chain to robustly and efficiently reconstruct the high-precision motion posture of each bone segment of the entire spine. This fundamentally solves the technical problems of traditional methods, such as cumbersome system construction, large computational load, and easy failure under motion occlusion caused by excessively dense marker points.

[0043] Figure 7 An example is a schematic diagram of the physical structure of a communication device, such as... Figure 7As shown, the communication device may include: a processor 710, a communication interface 720, a memory 730, and a communication bus 740, wherein the processor 710, the communication interface 720, and the memory 730 communicate with each other through the communication bus 740. The memory 730 includes a computer program, an operating system, and acquired data. The processor 710 can call logical instructions in the memory 730 to execute a spinal motion capture method. The method includes: dividing a continuous bone segment chain into multiple bone segments; determining a starting bone segment, a shared parent bone segment, and a shared child bone segment located between the starting bone segment and the shared parent bone segment from the multiple bone segments; setting optical markers on the starting bone segment and the shared parent bone segment of the continuous bone segment chain; acquiring the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system based on the three-dimensional coordinate data collected by the optical markers; calculating the initial attitude information of at least one shared child bone segment located between the starting bone segment and the shared parent bone segment based on the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system; and constructing an optimization problem objective function using the initial attitude information of the shared child bone segment as the initial value to collaboratively optimize the attitude information of each bone segment in the continuous bone segment chain, thereby obtaining the optimized attitude information of all bone segments in the continuous bone segment chain.

[0044] Furthermore, the logical instructions in the aforementioned memory 730 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to related technologies, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0045] On the other hand, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the spinal motion capture method provided by the above methods. The method includes: dividing a continuous bone segment chain into multiple bone segments; determining a starting bone segment, a shared parent bone segment, and a shared child bone segment located between the starting bone segment and the shared parent bone segment from the multiple bone segments; setting optical markers on the starting bone segment and the shared parent bone segment of the continuous bone segment chain; acquiring the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system based on the three-dimensional coordinate data collected by the optical markers; calculating the initial attitude information of at least one shared child bone segment located between the starting bone segment and the shared parent bone segment based on the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system; and constructing an optimization problem objective function using the initial attitude information of the shared child bone segment as the initial value to collaboratively optimize the attitude information of each bone segment in the continuous bone segment chain, thereby obtaining optimized attitude information of all bone segments in the continuous bone segment chain.

[0046] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0047] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the related technology, can be embodied in the form of software products. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0048] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for capturing spinal motion, characterized in that, include: The continuous bone segment chain is divided into multiple bone segments, and from the multiple bone segments, a starting bone segment, a shared parent bone segment, and a shared child bone segment located between the starting bone segment and the shared parent bone segment are determined; Optical markers are set on the starting bone segment and the shared parent bone segment of the continuous bone segment chain. The attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system is obtained based on the three-dimensional coordinate data collected by the optical markers. Based on the orientation information of the starting bone segment and the shared parent bone segment in the world coordinate system, calculate the initial orientation information of at least one shared child bone segment located between the starting bone segment and the shared parent bone segment. Using the initial pose information of the shared sub-bone segment as the initial value, an optimization problem objective function is constructed to collaboratively optimize the pose information of each bone segment in the continuous bone segment chain, thereby obtaining the optimized pose information of all bone segments in the continuous bone segment chain.

2. The spinal motion capture method according to claim 1, characterized in that, The determination of the initial bone segment, the shared parent bone segment, and the shared child bone segment located between the initial bone segment and the shared parent bone segment from the plurality of bone segments includes: The bone segment with six degrees of freedom in the continuous bone segment chain, which serves as the kinematic root node of the entire chain structure, is defined as the starting bone segment. The bone segment in the continuous bone segment chain that has independent rotational degrees of freedom and is used to provide a rotational reference for downstream bone segments is defined as the shared parent bone segment; A bone segment located between the starting bone segment and the shared parent bone segment, whose rotational degrees of freedom are configured to be inherited from the shared parent bone segment, is defined as a shared child bone segment.

3. The spinal motion capture method according to claim 1, characterized in that, The calculation of the initial pose information of at least one shared child bone segment located between the initial bone segment and the shared parent bone segment, based on the pose information of the initial bone segment and the shared parent bone segment in the world coordinate system, includes: Spherical linear interpolation is performed on the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system to generate the rotational attitude of the at least one shared child bone segment, which serves as the initial attitude information of the at least one shared child bone segment. The interpolation ratio coefficient of the spherical linear interpolation is dynamically adjusted according to the overall bending angle between the starting bone segment and the shared parent bone segment.

4. The spinal motion capture method according to claim 3, characterized in that, The interpolation ratio coefficient of the spherical linear interpolation is dynamically adjusted based on the overall bending angle between the initial bone segment and the shared parent bone segment, including: When the overall bending angle is within the first angle range, the interpolation ratio coefficient is calculated according to the first function; When the overall bending angle is within the second angle range, the interpolation ratio coefficient is calculated according to the second function, which includes additional terms related to the angle. When the overall bending angle is within the third angle range, the interpolation ratio coefficient is calculated according to the third function; Wherein, the upper limit of the first angle interval is less than the lower limit of the second angle interval, and the upper limit of the second angle interval is less than the lower limit of the third angle interval.

5. The spinal motion capture method according to claim 1, characterized in that, The objective function of the optimization problem includes: The position error term includes the difference between the observed position based on the optical markers on the shared parent bone segment and the position calculated based on the bone segment pose in the optimization. The length constraint error term includes the difference between the total model length based on the optimized continuous bone segment chain and the preset total anatomical length.

6. The spinal motion capture method according to claim 5, characterized in that, The coordinated optimization of the pose information of each bone segment in the continuous bone segment chain includes: The Levenberg-Marquardt algorithm is used to iteratively solve the objective function of the optimization problem in order to optimize the pose information of each bone segment in the continuous bone segment chain, including the shared parent bone segment and the shared child bone segment; The iterative solution process terminates when at least one of the following convergence conditions is met: The positional error change of the shared parent bone segment is less than a first preset threshold. The number of iterations has reached the preset maximum number of iterations; The position error calculated in the first iteration is less than the second preset threshold.

7. The spinal motion capture method according to claim 6, characterized in that, The iterative solution process also includes: Dynamically adjusted weighting coefficients are applied to the position error term and the length constraint error term; In the initial stage of iteration, the weight of the position error term is greater than the weight of the length constraint error term; as the iteration progresses, the weight of the length constraint error term is gradually increased.

8. A spinal motion capture device, characterized in that, include: A definition module is used to divide a continuous bone segment chain into multiple bone segments, and to determine the starting bone segment, the shared parent bone segment, and the shared child bone segment located between the starting bone segment and the shared parent bone segment from the multiple bone segments; The acquisition module is used to set optical markers on the starting bone segment and the shared parent bone segment of the continuous bone segment chain, and to acquire the attitude information of the starting bone segment and the shared parent bone segment in the world coordinate system based on the three-dimensional coordinate data collected by the optical markers. The calculation module is used to calculate the initial posture information of at least one shared child bone segment located between the starting bone segment and the shared parent bone segment based on the posture information of the starting bone segment and the shared parent bone segment in the world coordinate system. The optimization module is used to construct an optimization problem objective function based on the initial posture information of the shared sub-bone segment as the initial value, and to collaboratively optimize the posture information of each bone segment in the continuous bone segment chain to obtain the optimized posture information of all bone segments in the continuous bone segment chain.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the spinal motion capture method as described in any one of claims 1 to 7.

10. A non-transitory readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the spinal motion capture method as described in any one of claims 1 to 7.

Images