Method, device and system for identifying posture states of a person

Abstract

To provide a method and an apparatus for identifying a posture condition of a person.SOLUTION: There are provided a method and an apparatus for identifying a posture condition of a person. The method comprises the steps of: detecting a rotation of one or more body parts of the person around an upright center axis of the person; and calculating an asymmetric score of the one or more body parts based on the rotation of the one or more body parts of the person, the asymmetric score relating to a level of the posture condition of the person.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to, but is not limited to, a method and apparatus for identifying a person's pose state. [Background technology]

[0002] Posture is related to the alignment of various body segments relative to one another. Good posture means that the body's positioning is balanced to minimize stress on the body segments, while poor posture means that the body's positioning is unbalanced, causing abnormal stress on various body segments, which can lead to abnormal anatomical adaptations, altered performance, and reduced efficiency.

[0003] Postural analysis is the evaluation of the function of the motor system (bones, muscles, ligaments) and its control by the nervous system. It covers not only the evaluation of bones and muscles, but also the alignment of the spinal cord. Postural analysis allows the examination of the correct standing alignment of a person from the front, back, and side views.

[0004] Posture analysis allows experienced and skilled physical therapists to assess a patient's posture more accurately and confidently, while junior physical therapists may struggle to assess a patient's posture accurately and confidently. To date, there are several limitations to posture assessment, such as manual assessment by identifying landmarks on images, placing reflective markers at specific body positions, or utilizing pressure mats under the patient's feet. Most calculated deviations from images relate to the shift (e.g., left / right) and tilt (e.g., up / down) of the body's major joints from the reference, but do not consider the rotation of the body's joints.

[0005] Furthermore, body, facial, and spinal landmarks (e.g., anatomical points) are important for assessing overall postural deviations. In particular, spinal landmarks are provided manually by the clinician. Each individual's posture is different and typically begins to form at a young age. Furthermore, deformations of bone structures can also cause postural deformations in the form of shifts, tilts, and rotations. Therefore, it is important to assess individual posture based on identifying not only shifts and / or tilts but also rotations of body parts for all age groups, from children to the elderly and from healthy individuals to patients, in order to obtain early intervention to correct an individual's posture. Summary of the Invention [Problem to be solved by the invention]

[0006] The present invention is an embodiment of a method and apparatus for identifying a postural state of a person that addresses one or more of the above problems.

[0007] Furthermore, other desirable features and characteristics will become apparent from the subsequent detailed description and appended claims, taken in conjunction with the accompanying drawings and this background of the disclosure. [Means for solving the problem]

[0008] In a first aspect, the present disclosure provides a method for identifying a postural state of a person, the method comprising detecting a rotation of one or more body parts of a person around an upright central axis of the person and calculating an asymmetry score for the one or more body parts based on the rotation of the one or more body parts of the person, the asymmetry score being related to a level of the postural state of the person.

[0009] In a second aspect, the present disclosure provides an apparatus for identifying a postural state, comprising at least one processor and at least one memory including computer program code, the at least one memory and the computer program code being configured to: use the at least one processor to detect rotation of one or more body parts of a person around the person's upright central axis; and calculate an asymmetry score for the one or more body parts based on the rotation of the one or more body parts of the person; and the asymmetry score is configured to cause a server to perform a step related to a level of the postural state of the person.

[0010] In a third aspect, the present disclosure provides a system for identifying a postural state of a person, comprising the apparatus of the above aspect and one or more image capture devices configured to capture one or more images of the person, wherein the one or more images comprise an image of the person viewed from the front and / or an image of the person across a sagittal plane.

[0011] Additional benefits and advantages of the disclosed embodiments will become apparent from the specification and drawings. Benefits and / or advantages may be obtained individually from various embodiments and features of the specification and drawings, and not all of them need be provided to obtain one or more of such benefits and / or advantages.

[0012] The accompanying drawings, in which like numerals refer to identical or functionally similar elements throughout the different views, and which, together with the following detailed description, are incorporated into and constitute a part of the specification, illustrate various embodiments and serve to explain, by way of non-limiting example only, various principles and advantages according to the present embodiments.

[0013] Embodiments of the present invention will be better understood and readily apparent to those skilled in the art from the following written description, by way of example only, and in conjunction with the drawings in which: [Brief explanation of the drawings]

[0014] [Figure 1]An example of estimating shift, tilt and rotation from an image of a person is given below.

[0015] [Figure 2] 1 shows an illustration of unbalanced weight bearing due to hip rotation of a person.

[0016] [Figure 3] 1 shows a flowchart illustrating a method for identifying a pose state of a person according to various embodiments of the present disclosure.

[0017] [Figure 4] 1 shows a block diagram of an apparatus for identifying a person's pose state according to an embodiment of the present disclosure.

[0018] [Figure 5] 1 illustrates a workflow diagram of an overall system for identifying pose states of a person according to an embodiment of the present disclosure.

[0019] [Figure 6] 1 shows a flowchart for identifying a pose state of a person according to an embodiment of the present disclosure.

[0020] [Figure 7] 1 shows a flowchart of a landmark detection system according to an embodiment of the present disclosure.

[0021] [Figure 8] 1 shows an exemplary diagram of landmark detection according to an embodiment of the present disclosure.

[0022] [Figure 9] 10 shows another exemplary diagram of landmark detection according to an embodiment of the present disclosure.

[0023] [Figure 10] 1 shows a diagram for detection of the jugular notch (JN) from a frontal view according to an embodiment of the present disclosure.

[0024] [Figure 11]1 shows a diagram for detection of the jugular notch (JN) from a lateral view according to an embodiment of the present disclosure.

[0025] [Figure 12] 1 shows an exemplary front view for detection of the jugular notch (JN) according to an embodiment of the present disclosure.

[0026] [Figure 13] 1 shows an exemplary side view for detecting the jugular notch (JN) according to an embodiment of the present disclosure.

[0027] [Figure 14] 1 illustrates a front view for identifying a centerline (CR) location based on a detected jugular notch (JN) according to an embodiment of the present disclosure. FIG.

[0028] [Figure 15A] 1 illustrates a right side view for detecting the centerline (CR) location, chest, and mid-chest (e.g., T7) according to an embodiment of the present disclosure. [Figure 15B] 1 illustrates a left side view for detecting the centerline (CR) location, chest, and mid-chest (e.g., T7) according to an embodiment of the present disclosure.

[0029] [Figure 16] 1 shows a right side view for estimating the position of the cervical (C7), thoracic (T2, T7, T10), lumbar (L1 and L4) and sacral (S2) vertebrae of the spine according to an embodiment of the present disclosure.

[0030] [Figure 17] 1 shows a left side view for estimating the position of the cervical (C7), thoracic (T2, T7, T10), lumbar (L1 and L4) and sacral (S2) vertebrae of the spine according to an embodiment of the present disclosure.

[0031] [Figure 18] 1 illustrates a front / front view with detected full-body landmarks according to an embodiment of the present disclosure.

[0032] [Figure 19A]FIG. 1 illustrates a left side view with detected full body landmarks according to an embodiment of the present disclosure. [Figure 19B] 1 illustrates a right side view with detected full body landmarks according to an embodiment of the present disclosure.

[0033] [Figure 20] 1 shows a flowchart illustrating a posture analysis device according to an embodiment of the present disclosure.

[0034] [Figure 21] 10 shows a table for calculating deviation and kinematic parameters for a front view of a body image according to an embodiment of the present disclosure.

[0035] [Figure 22A] 10 illustrates an analysis of various body part positions according to an embodiment of the present disclosure. [Figure 22B] 10 illustrates an analysis of angles for various body part positions according to an embodiment of the present disclosure.

[0036] [Figure 23A] 1 illustrates an analysis of knee placement according to an embodiment of the present disclosure. [Figure 23B] 1 illustrates an analysis of foot placement according to an embodiment of the present disclosure.

[0037] [Figure 24] 1 shows a diagram for determining body part rotation according to an embodiment of the present disclosure;

[0038] [Figure 25] 10 shows a table for calculating deviation and kinematic parameters of a side view of a body image according to an embodiment of the present disclosure.

[0039] [Figure 26A] 10A-10C show diagrams for determining joint shift from a side view according to an embodiment of the present disclosure;

[0040] [Figure 26B] 10A-10C show diagrams for determining joint angles from a side view according to an embodiment of the present disclosure;

[0041] [Figure 26C] FIG. 10 shows a diagram for determining hip and knee joint angles from a side view according to an embodiment of the present disclosure.

[0042] [Figure 27A] 1 shows an exemplary diagram of a "forward head / straight neck" posture according to an embodiment of the present disclosure.

[0043] [Figure 27B] 1A-1C show exemplary diagrams of chin-down postures according to embodiments of the present disclosure;

[0044] [Figure 28A] 10A-10C illustrate exemplary diagrams of posture analyzer output based on a frontal body view according to embodiments of the present disclosure. [Figure 28B] 10A-10C illustrate exemplary diagrams of posture analyzer output based on a frontal body view according to embodiments of the present disclosure.

[0045] [Figure 29A] 10A-10C each show an exemplary diagram of a posture analyzer output based on a right-side body view according to an embodiment of the present disclosure. [Figure 29B] 10A-10C show exemplary diagrams of posture analyzer output based on a left body view, respectively, according to an embodiment of the present disclosure.

[0046] [Figure 30] 10 shows a table for calculating an asymmetry score to provide a level of postural abnormality according to an embodiment of the present disclosure.

[0047] [Figure 31] 1 shows a flowchart illustrating a posture corrective exercise recommendation device according to an embodiment of the present disclosure.

[0048] [Figure 32] 4 shows a schematic diagram of an exemplary computing device suitable for use in performing the method of FIG. 3. DETAILED DESCRIPTION OF THE INVENTION

[0049] A person's posture state relates to how the person's body parts (e.g., shoulders, chin, hips, head, ears, eyebrows, ankles, and other similar body parts) deviate from their standard positions (e.g., a reference position). This may relate to a shift (e.g., a left or right shift), tilt (e.g., an upward or downward tilt), or rotation from the reference position. Identification can utilize images of front, frontal, and / or side views of the person's body (e.g., an image of the person across the sagittal plane, such as a right and / or left side view). Based on the images, each body part position is identified and indicated by a respective landmark (e.g., an anatomical point), and any shift, tilt, or rotation is detected based on detecting and calculating distances measured between body part positions and / or comparing lines connecting the body part positions, and angles obtained from the line comparison. For example, rotation of one or more body parts of the person around the person's upright central axis is determined based on detecting the angle of a line connecting two body part positions relative to a reference line. The reference line may be another line connecting two other body part positions of the person. Different landmarks, lines and / or angles may be used depending on the body part for which the posture is to be determined.

[0050] An asymmetry score for the person's level of postural state can be calculated based on the detected rotation of one or more body parts of the person. For example, deviations and kinematic parameters can be calculated based on the identified landmarks, lines, and / or angles to derive the score. For example, a higher asymmetry score may indicate a greater severity of deviation for that person's body part, and postural correction of that body part may be given more emphasis compared to other body parts with lower asymmetry scores.

[0051] Embodiments of the present invention will now be described, by way of example only, with reference to the drawings in which like reference numerals and letters indicate like or equivalent elements.

[0052] Some portions of the description which follow are presented explicitly or implicitly in terms of algorithms and functional or symbolic representations of operations on data within a computer memory. These algorithmic descriptions and functional or symbolic representations are the means used by those skilled in the data processing arts to most effectively convey the substance of their work to others skilled in the art. An algorithm is here, and generally, conceived to be a self-consistent sequence of steps leading to a desired result. The steps involve physical manipulations of physical quantities, such as electrical, magnetic, or optical signals capable of being stored, transferred, combined, compared, and otherwise manipulated.

[0053] Unless otherwise indicated, and as will become apparent below, discussions throughout this specification using terms such as "detecting," "estimating," "comparing," "receiving," "calculating," "determining," "updating," "generating," "initializing," "outputting," "receiving," "retrieving," "identifying," "distributing," "authenticating," and the like, will be understood to refer to the operations and processes of a computer system or similar electronic device or other information storage, transmission, or display device that manipulates and transforms data, represented as physical quantities within the computer system, into other data, also represented as physical quantities within the computer system.

[0054] Apparatus for performing these method operations is also disclosed herein. Such apparatus may be specially constructed for the required purposes, or may comprise a computer or other device selectively activated or reconfigured by a computer program stored in the computer. The algorithms and displays presented herein are not inherently related to any particular computer or other device. Various machines may be used with programs in accordance with the teachings described herein. Alternatively, the construction of more specialized apparatus to perform the required method steps may be appropriate. The structure of the computer will emerge from the description below.

[0055] Furthermore, this specification also implicitly discloses a computer program in that it would be obvious to one skilled in the art that the individual steps of the methods described herein could be implemented by computer code. The computer program is not intended to be limited to a particular programming language and its implementation. It will be understood that a variety of programming languages ​​and coding thereof can be used to implement the teachings of the disclosure contained herein. Furthermore, the computer program is not intended to be limited to a particular control flow. There are many other variations of the computer program that can use different control flows without departing from the spirit or scope of the present invention.

[0056] Furthermore, one or more steps of the computer program may be executed in parallel rather than sequentially. Such a computer program may be stored on any computer-readable medium. The computer-readable medium may include storage devices such as magnetic or optical disks, memory chips, or other storage devices suitable for interfacing with a computer. The computer-readable medium may also include wired media, such as those exemplified by the Internet system, or wireless media, such as those exemplified by the GSM mobile telephone system. Such a computer program loaded and executed on a computer effectively provides an apparatus that implements the steps of the preferred method.

[0057] Illustrative Embodiments Various embodiments of the present disclosure relate to methods and apparatus for identifying a pose state of a person.

[0058] FIG. 1 shows an exemplary body image 100 that can be used to determine whether there is a rotation in the body posture (e.g., in the direction indicated by arrow 102 relative to a vertical reference axis 104, such that the upper body rotates clockwise around reference axis 104, as indicated by arrow 102). FIG. 2 shows images 200, 202, and 204 illustrating changes in pressure through the feet of a person with pelvic rotation, for example, through the use of a pressure mat to detect pressure points under the feet. Image 200 shows an increase in pressure at arrow 206, for example, due to clockwise rotation of the pelvis, image 200 shows a neutral pelvic position where pressure through the feet is equal, and image 204 shows an increase in pressure at arrow 208, for example, due to counterclockwise rotation of the pelvis. These images in FIGS. 1 and 2 can be used to analyze the shift and tilt of a person's body parts. However, it is also useful to determine the rotation of the body parts so that the cause of the shift or tilt can be identified. Body parts can be better analyzed to determine how good posture can be achieved through corrective training so that an appropriate postural control training program can be recommended.

[0059] Furthermore, there is still a lack of postural control training recommendation systems for detecting anatomical landmarks, providing deviation and postural assessment, and comprehensive postural assessment for a person's recovery (e.g., from detected poor posture) and early intervention (e.g., to prevent further regression of poor posture). Therefore, a comprehensive system for assessing standing posture and providing postural correction training recommendations can include an anatomical landmark detection system, a posture analysis system for calculating body segment deviations for the person's anterior / posterior, posterior, and lateral sides, and a postural abnormality scoring system. An example of a comprehensive system is shown in Figure 4. These systems have the advantage of being able to indicate deviations in body segment alignment from a baseline. For example, a postural abnormality scoring system can identify the severity of the postural condition, and a postural correction training recommendation system can enable readjustment of the incorrect posture.

[0060] 3 shows a flowchart 300 illustrating a method for identifying a posture state of a person according to various embodiments of the present disclosure. In step 302, rotation of one or more body parts of a person about the person's central upright axis is detected. In step 304, an asymmetry score for one or more body parts is calculated based on the rotation of the one or more body parts of the person, the asymmetry score associated with the level of the person's posture state.

[0061] In an embodiment, the method further includes detecting an angle of a first line connecting the positions of two body parts of the person from the image of the person relative to a reference line, where detecting the rotation of one or more body parts of the person around the person's upright central axis is based on the angle. The first line can connect the positions of a left body part and a right body part of a first body part of the person when viewed from the front, and the reference line can include a second line connecting the positions of the left body part and the right body part of a second body part of the person when viewed from the front. The reference line can be a third line connecting the positions of another two body parts of the person.

[0062] In an embodiment, the method may include detecting a first distance of a position of the body part from a nearest point along a central standing axis of the person, wherein detecting a rotation of the one or more body parts of the person and / or calculating an asymmetry score for the one or more body parts of the person is based on the first distance.

[0063] In an embodiment, the method includes detecting a second distance of the body part position from one of (i) two body part positions of the person, (ii) a midpoint between the two body part positions, or (iii) a fourth line connecting the two body part positions, wherein detecting the rotation of the one or more body parts of the person and / or calculating an asymmetry score for the one or more body parts of the person is based on the second distance.

[0064] In an embodiment, the method includes detecting multiple body part positions of a person based on relative positions of multiple body parts in one or more images in which the person is detected, where each of the multiple body part positions corresponds to a body part of the person. Detecting the multiple body part positions of the person includes estimating a body part position of the person based on one of the multiple body part positions, where the multiple body part positions of the person further include the estimated body part position.

[0065] In implementations, the method may include receiving demographic data associated with the person, where the calculation of the asymmetry score is further based on the demographic data. The asymmetry score may be one of a plurality of asymmetry scores associated with a plurality of body parts, and the method may further include calculating a level of the person's postural state based on the plurality of asymmetry scores. The method may further include comparing each of the plurality of asymmetry scores with other asymmetry scores of the plurality of asymmetry scores, identifying one or more asymmetry scores having a higher score among the plurality of asymmetry scores, and identifying a set of posture correction programs based on the results of the identification.

[0066] 4 illustrates a block diagram of an apparatus 400 for identifying a postural state of a person according to an embodiment of the present disclosure. In an embodiment, the apparatus 400 can be generally described as a physical device including at least one processor and at least one memory containing computer program code. The at least one memory and the computer program code are configured to cause the physical device, using the at least one processor, to perform the operations described in FIG. 3 .

[0067] In an embodiment, device 400 can receive input video or images (e.g., video frames or images of front, front, and / or side views of a person's body, or other similar images) from source 402. For example, the input images or videos may be newly captured images or videos, may be from an existing image or video database, may be captured by a camera, or may be stored on a device such as a smartphone, camera, or other similar device. In one example, the input video may be decomposed into multiple still video image frames, such that each video image frame is analyzed by device 400. Device 400 can include a landmark detection system 404 configured to determine a set of two-dimensional and / or three-dimensional body landmark positions from the input video or images, and a pose analyzer 406 configured to determine a level of body alignment deviation and / or posture abnormality score (e.g., an asymmetry score) based on the set of two-dimensional and / or three-dimensional body landmark positions and other demographic data as input. In one example, the landmark detection system 404 can be configured to determine whether the image is a front view, a back view, or a side view of the person's body to determine the landmark locations. The landmark detection system 404 is further described in FIGS. 7-19, and the pose analyzer 406 is further described in FIGS. 20-30. The device 400 can also include a posture correction exercise recommender 410 configured to recommend an appropriate exercise or posture correction program based on the level of body alignment deviation and / or posture abnormality score (e.g., asymmetry score) as input. The posture correction exercise recommender 410 is further described in FIG. 31.

[0068] Device 400 may be configured to display video and / or images received from source 402 and / or the inputs and outputs of landmark detection system 404, pose analyzer 406, and posture-correcting exercise recommender 410 on user interface 412. A user of device 400 may also interact with user interface 412 with data entered via input / output interface display 414. It should be understood that each of landmark detection system 404, pose analyzer 406, and posture-correcting exercise recommender 410 may be part of device 400 or may be a standalone device or part of another device, and is in communication with device 400 via a connection. Such connection may be wired, wireless (e.g., via NFC communication, Bluetooth, etc.), or via a network (e.g., the Internet).

[0069] The device may include a data storage device 408 accessible by the device 400 for storing video and images from the source 402, as well as the inputs and outputs of the landmark detection system 404, the pose analyzer 406, and the posture-correcting exercise recommender 410. While the data storage device is shown in FIG. 4 as being part of the device 400, it should be understood that the data storage device 408 may not form part of the device 400, but may be in communication with the device 400 via a connection. Such a connection may be wired, wireless (e.g., via NFC communication, Bluetooth, etc.), over a network (e.g., the Internet), or via a cloud server. In another embodiment, each of the landmark detection system 404, the pose analyzer 406, the posture-correcting exercise recommender 410, and the source 402 may include its own data storage device for storing its input / output data.

[0070] 5 shows an overall system workflow diagram 500 for identifying a person's postural state according to an embodiment of the present disclosure. In an embodiment, video or images (e.g., video frames or images of front, front, and / or side views of a person's body, or other similar images) are received from a source (e.g., an image and / or video capture device, a database, the Internet, or other similar source) and stored in data storage 508 or sent directly to landmark detection system 504 as input before being used as input by landmark detection system 504. Landmark detection system 504 can determine a set of two-dimensional and / or three-dimensional body landmark locations from the input video or images, and pose analyzer 506 can determine a level of body alignment deviation and / or postural abnormality score (e.g., an asymmetry score) based on the set of two-dimensional and / or three-dimensional body landmark locations and other demographic data as input. The posture correction exercise recommender 410 can recommend an appropriate exercise or posture correction program based on the body alignment deviation and / or the level of the posture abnormality score (e.g., asymmetry score) as input. The recommended exercise or posture correction program can then be displayed on the input / output user interface display of the user interface 512.

[0071] In an embodiment, a system for identifying a postural state of a person may include apparatus 400 and one or more image capture devices configured to capture one or more images of the person, the one or more images including an image of the person viewed from the front and / or an image of the person across a sagittal plane.

[0072] FIG. 6 shows a flowchart 600 for identifying a person's postural state according to an embodiment of the present disclosure. In step 602, images of the front, back, and side of the person's body can be acquired. In one example, it can be determined whether the acquired images are front, back, or side images of the person's body. In one example, an input video is acquired, the video can be decomposed into multiple still video image frames, and it can be determined whether each video image frame is a front, back, or side image of the person's body. In step 604, anatomical landmarks (e.g., spatial coordinates of each landmark) for each body part in the image can be obtained. In step 606, the obtained anatomical landmarks can be reviewed with a clinician or healthcare professional. In step 608, deviations and kinematic parameters of the front, back, and side images of the image are calculated. In step 610, an asymmetry score is calculated to provide a level of postural abnormality. In step 612, one or more sets of personalized postural correction exercise programs are provided. In step 614, the deviations, postural abnormalities, and one or more sets of exercise programs are reviewed with a clinician or healthcare professional. In step 616, one or more exercises and appropriate exercises for the selected deviations (eg, based on the asymmetry score) are selected as prescriptions for the patient.

[0073] FIG. 7 illustrates a flowchart 700 of a landmark detection system according to an embodiment of the present disclosure. In step 702, images of the front, back, and side of a person's body may be acquired. In one example, it may be determined whether the acquired image is a front image, a back image, or a side image of the person's body. In an example of acquiring input video, the video may be broken down into multiple still video image frames, and each video image frame may be determined to be a front image, a back image, or a side image of the person's body. In step 704, anatomical landmarks (e.g., spatial coordinates of each landmark) for each body part in the image may be acquired. In step 706, facial landmark coordinates are acquired. In step 708, landmark coordinates for the person's jugular notch are acquired. In step 710, thoracic landmark coordinates are acquired. In step 712, it is determined whether the acquired image is a side image of the person's body. If it is not determined that this is the case, the process proceeds to step 714, where the acquired anatomical landmarks are reviewed with a clinician or medical professional, and the process ends. Otherwise, processing proceeds to step 716, where spinal landmark coordinates are obtained, and then to step 714, where the obtained anatomical landmarks (e.g., including spinal landmark coordinates) are reviewed with a clinician or medical professional, and processing ends.

[0074] 8 shows an exemplary diagram 800 for landmark detection according to an embodiment of the present disclosure. For example, referring to flowchart 700, landmark coordinates for each body part in section 802 can be obtained in step 704. Landmark coordinates for face section 806 can be obtained in step 706. Landmark coordinates for chest section 804 can be obtained in step 710. Spinal landmark coordinates for spinal section 808 (e.g., as can be seen from a side view of the person's body) can be obtained in step 716. Additionally, landmark coordinates 810 indicating the cervical notch can be obtained in step 708.

[0075] In one example, an open-source body and facial landmark detection engine, such as the MediaPipe Holistic Engine, can be utilized to obtain the required anatomical landmark coordinates. However, open-source engines typically cannot provide some of the landmarks required by a clinician, as shown in landmark diagram 900 of FIG. 9. For example, the jugular notch landmark, as shown at 902, is typically provided manually by a clinician. It will be appreciated that the required anatomical landmark coordinates can also be obtained by other non-open-source, proprietary landmark detection engines.

[0076] FIG. 10 shows a diagram 1000 for detecting the jugular notch (JN) from a front view of a person's body (e.g., an image of a person viewed from the front) according to an embodiment of the present disclosure. The respective positions of the person's chin, left shoulder, and right shoulder may be determined based on multiple body part positions and / or facial part positions detected, for example, in steps 704 and 706 of flowchart 700. For example, the chin landmark coordinate (C) may be defined as C:[Cx, Cy] (e.g., obtained in step 706 of flowchart 700), and the shoulder midpoint landmark coordinate (Sm) may be defined as Sm:[Smx, Smy] (e.g., identifying the midpoint between the left and right shoulder positions based on the landmark coordinates corresponding to the person's left and right shoulders obtained in step 704 of flowchart 700). Furthermore, the distance from the chin position to the shoulder midpoint position may be calculated and defined as dCSm. Then, the position of the person's jugular notch may be estimated based on the chin position and the calculated distance. For example, the landmark coordinates of the jugular notch may be defined as JN:[JNx, JNy]. Therefore, based on the landmark coordinates of the chin, shoulder, and midpoint of the shoulder, it can be determined that JNx corresponds to Smx, and JNy corresponds to Smy-25%dCSm.

[0077] Furthermore, as shown in diagram 1100 of FIG. 11, it is also possible to determine the position of the jugular notch from a side view of a person's body (e.g., an image of the person across the sagittal plane). The respective positions of the person's chin and left or right shoulder (depending on the side of the person shown in the image) can be determined based on multiple body part positions detected, for example, in steps 704 and 706 of flowchart 700. The shoulder landmark coordinate (S) can be defined as S:[Sx, Sy] (e.g., based on the landmark coordinates corresponding to the person's left or right shoulder obtained in step 704 of flowchart 700). A midpoint position along a vertical line starting from the chin position (e.g., the landmark coordinates corresponding to the chin position obtained in step 706 of flowchart 700) and ending at a horizontal line passing through the shoulder position can be identified. The vertical distance from the jaw position (C) to the horizontal line is defined as dCS, and 50% of this vertical distance (e.g., the distance from the jaw position to the midpoint) is calculated and defined as mCS. The distance from the shoulder to the midpoint is calculated and defined as l. θ is calculated and defined as the angle between the line from the shoulder position to the midpoint and the horizontal line passing through the shoulder position (S). Based on the calculated distance l and the calculated angle θ, the jugular notch (JN) can be detected and defined as JN:[JNx, JNy], where JNx corresponds to Sx±x, x = cos(θ)*l / 2, and JNy corresponds to Sy-y, y = 25% dCS. Based on these techniques and calculations, it is possible to identify the cervical notch based on a front view image (e.g., cervical notch landmark 1202 from exemplary front view image 1200 of FIG. 12, respectively) and a side view image (e.g., cervical notch landmarks 1304 and 1306 from exemplary side view images 1300 and 1302 of FIG. 13, respectively).

[0078] FIG. 14 illustrates a front view for identifying the midline (CR) location based on the detected jugular notch (JN) according to an embodiment of the present disclosure. Based on the jugular notch (JN), the midline (CR) location can be calculated, where the midline location can be 3-4 inches (e.g., 8-10 cm) below the jugular notch location. To detect the midline (CR) location, chest, and mid-thorax (e.g., T7) based on the jugular notch (JN), a right lateral view 1500 and / or a left lateral view 1502 of the person's body can be used to detect the CR location or a mirror image of the mid-thorax or T7. For example, a landmark used to locate the midline (CR) is at T7 (e.g., mid-thorax). The level of T7 is 3-4 inches (e.g., 8-10 cm) below the jugular notch.

[0079] 16 and 17 illustrate right and left lateral views 1600 and 1700, respectively, for estimating the location of the cervical (C7), thoracic (T2, T7, T10), lumbar (L1 and L4), and sacral (S2) vertebrae of the spine according to an embodiment of the present disclosure. The location of vertebrae along a person's spinal cord can be estimated based on the estimated location of the person's mid-thorax and the relative distance between each vertebra along the spinal cord. For example, based on the approximate distance between T7 and C7 and previous research by Ernst et al. on the ratio of the thoracic spine to the posterior superior iliac spine (PSIS), the locations of C7, T2, T7, T10, L1, L4, and S2 (PSIS) can be estimated using the lateral view images 1600 and 1700. In one example, a flexible ruler can be placed on the person's back to determine the distance between various segments of the spinal cord. The estimated distances can be used to calculate the percentage position of the overall segment distance from C7 to the PSIS.

[0080] 18 , 19A, and 19B show the results of a landmark detection system from a front / front view image 1800, a left side view image 1900, and a right side view image 1902, respectively, according to an embodiment of the present disclosure, where each landmark indicates a body part and is represented by a dot on the person's body in the images. For example, the person's jugular notch is indicated by landmark 1802 in image 1800 and by landmark 1904 in images 1900 and 1902. Further results of the landmark detection system are also shown in body edge image 1602 of FIG. 16 and body edge image 1702 of FIG. 17, where, for example, based on landmarks detected for the person's entire body in images 1900 and 1902, detected landmarks for the spine are selected and displayed on body edge images 1702 and 1602, respectively; for example, detected spinal cord landmarks are shown as white dots in body edge images 1602 and 1702.

[0081] FIG. 20 shows a flowchart 2000 illustrating a posture analyzer according to an embodiment of the present disclosure. In step 2002, anatomical landmark coordinates are obtained from one or more images showing, for example, front, back, and side views of a person's body. For example, the landmark coordinates are obtained as input from landmark detection system 404. Additionally, other demographic data about the person is also obtained. In step 2004, deviations and kinematic parameters are obtained for the front, back, and side views. In one example, the deviations and kinematic parameters may be based on the landmark coordinates or other demographic data associated with the person. In step 2006, an asymmetry score is calculated to provide a level representative of one or more posture abnormalities.

[0082] FIG. 21 shows a table for calculating deviations and kinematic parameters for a front view of a body image (e.g., as calculated in step 2004 of flowchart 2000) according to an embodiment of the present disclosure. Based on a forward view of a person's body, deviations and kinematic parameters can be calculated for various posture abnormalities, such as left-right shift (e.g., in cm), up-down tilt (e.g., in degrees), joint angle (e.g., the angle (in degrees) between a first line connecting two body parts and a second line connecting two other body parts), and rotation (e.g., in degrees). In an embodiment, the calculation for body part shift may be based on the distance of the body part's midpoint from the midpoint of a reference segment joint, such as between the hip and ankle, between the shoulder and hip, between the chin and shoulder, between the ear and the chin, between the eyebrow and the ear, and other similar distances. In another embodiment, the movement of a body part may be calculated by calculating the distance of the body part's midpoint from a reference vertical line or center of gravity line. In embodiments, calculations for the tilt of a body part may refer to body parts on the right or left side of a person's body, such as toes, heels, ankles, hips, shoulders, ears, eyebrows, elbows, wrists, and other similar body parts.

[0083] In another embodiment, calculations for joint angles may reference a first line connecting left and right body parts and a second line connecting other left and right body parts (e.g., a line connecting left and right shoulders, left and right hips, left and right knees, left and right toes, or other similar body parts). For example, left and right joint angles are obtained by referencing the left and right body parts connected by the first or second line and calculating the left and right angles, respectively, of a straight third line passing through the midpoint of the first line and the midpoint of the second line. Details of joint angles are shown in FIG. 24. Furthermore, calculations for the rotation of a body part (e.g., hips, shoulders, ears, or other similar body parts) may be based on the calculated left and right joint angles.

[0084] Based on a front view of a person's body in a standing position, the following analysis can be provided. For example, left and right shifts can be determined based on deviations of body segment distances from reference body segments, such as the hip distance from the ankle, the shoulder distance from the hip, the ear to the chin distance, the eyebrow distance from the ear, the chin distance from the shoulder, and other similar distances. In the present disclosure, "-" means a shift to the left side of the body, and "+" means a shift to the right side of the body. For example, with reference to the front view image 2200 in FIG. 22A , the ear position (indicated by reference numeral 2202) shifts 0.2 cm left relative to the chin, the chin position (indicated by reference numeral 2204) shifts 0.05 cm right relative to the shoulder, the shoulder position (indicated by reference numeral 2206) shifts 0.2 cm left relative to the hip, the hip position (indicated by reference numeral 2208) shifts 0.2 cm left from the ankle, and the ankle position is indicated by reference numeral 2210.

[0085] In another example, the left and right tilt angles are determined based on how the positions of the body joints (e.g., ankle, knee, hip, shoulder, ear angle) are tilted from a reference side. In the present disclosure, "-" means tilting downward from the reference plane, and "+" means tilting upward from the reference plane. For example, with reference to the front view image 2212 of FIG. 22B , the neck tilt can be determined based on the ear angle 2214, the shoulder tilt can be determined based on the shoulder angle 2216, the wrist tilt can be determined based on the wrist angle 2218, the hip tilt can be determined based on the hip angle 2220, the knee tilt can be determined based on the knee angle 2222, and the ankle tilt can be determined based on the ankle angle 2224.

[0086] Lower limb analysis may be performed from the front view to determine bowed or knocked legs, with reference to diagram 2300 in FIG. 23A (e.g., based on right knee angle 2304 and left knee angle 2306 in image 2302). "-" means bowed legs, and "+" means knocked legs. Further, with reference to diagrams 2308 and 2312 in FIG. 23A, whether bowed or knocked legs are present (and the severity of the bowed or knocked legs) may be determined, for example, based on the distance 2310 between the right knee and the line connecting the right hip and right ankle for bowed leg image 2308, or the distance 2314 between the left knee and the line connecting the left hip and left ankle for bowed leg image 2312. Further, with reference to the diagram in FIG. 23B, toe-in and toe-out positions of the feet may be determined. For example, referring to diagram 2316 in FIG. 23B, the placement of a person's feet can be determined based on toe angle with reference to reference line 2322, where a positive (+) toe angle means toe in angle 2318 and a negative (-) toe angle means toe out angle 2320.

[0087] 24 shows a diagram 2400 for determining body part rotation according to an embodiment of the present disclosure. For example, head rotation may be based on analysis of ear-shoulder joint angles 2402 and 2404 calculated from a first line 2406 connecting the ears and a second line 2408 connecting the shoulders with reference to a third line 2410. Shoulder or upper body rotation may be based on analysis of shoulder-hip joint angles 2412 and 2414 calculated from a first line 2408 connecting the shoulders and a second line 2416 connecting the hips with reference to the third line 2410. Additionally, hip or lower body rotation may be based on analysis of hip-ankle joint angles 2418 and 2420 calculated from a first line 2416 connecting the hips and a second line 2422 connecting the ankles with reference to the third line 2410. The resulting joint angles may be analyzed as shown in table 2424. For example, if the left ear-shoulder joint angle is smaller than the right ear-shoulder joint angle, it indicates a right-to-left head rotation. If the left shoulder-hip joint angle is larger than the right shoulder-hip joint angle, it indicates a left-to-right shoulder or upper body rotation. Furthermore, if the left hip-ankle joint angle is larger than the right hip-ankle joint angle, it indicates a left-to-right hip or lower body rotation.

[0088] FIG. 25 illustrates a table 2500 for calculating deviation and kinematic parameters of a side view of a body image according to an embodiment of the present disclosure. For example, anterior-posterior movements of body parts such as the knees, hips, shoulders, chin, ears, and eyebrows can be determined based on their distance from a reference vertical line. With reference to the side view image 2600 of FIG. 26A , and a vertical line 2602 passing vertically 2 cm forward from the ankle position 2612, it can be determined that the ear position (indicated by reference numeral 2604) shifts forward 2.0 cm, the shoulder position (indicated by reference numeral 2606) shifts backward 3.0 cm, the hip position (indicated by reference numeral 2608) shifts forward 0.50 cm, and the knee position (indicated by reference numeral 2610) shifts backward 3.0 cm.

[0089] Additionally, joint angles for the ankle, knee, hip-shoulder, hip angle, head, and forward head and jaw drop postures can be measured in degrees with reference to a reference line. For example, with reference to side view image 2614 of FIG. 26B , a head or shoulder-ear angle 2616 can be determined with reference to a line 2620 connecting the ear and shoulder and a vertical line 2618. A hip-shoulder angle 2622 can be determined with reference to a line 2624 connecting the hip and shoulder and a vertical line 2626. In another example shown in side view image 2634 of FIG. 26C , parameters such as a hip angle 2636 and a knee angle 2638 can be determined. Furthermore, a joint angle 2702 for a forward head posture can be determined based on side view image 2700 of FIG. 27A , and corresponding deviations and kinematic parameters can also be calculated from jaw drop posture side view image 2704 of FIG. 27B .

[0090] 28A and 28B show exemplary diagrams 2800 and 2808 of posture analyzer output based on a front body view according to an embodiment of the present disclosure. For example, in the front body view of diagram 2800, symbols indicating left and right shift and tilt of each body joint (e.g., symbol 2804 indicating shift and symbol 2806 indicating the joint angle of each body part) are displayed on the body image. Also, in the front body view of diagram 2808, symbols indicating left and right lower limb analysis (e.g., symbol 2810 indicating hip angle, symbol 2812 indicating knee angle, symbol 2814 indicating bowlegs, symbol 2816 indicating ankle alignment, and symbol 2818 indicating toe-out) are displayed on the body image.

[0091] 29A and 29B show exemplary diagrams 2900 and 2904 of posture analyzer output based on right and left body views, respectively, according to an embodiment of the present disclosure. For example, for the right body view in diagram 2900, a symbol 2902 indicating a right shift from vertical at a specific joint angle is displayed on the body image. Additionally, for the left body view in diagram 2904, a symbol 2906 indicating a left shift from vertical at a specific joint angle is displayed on the body image.

[0092] 30 shows a table 3000 for calculating asymmetry scores to provide a level of postural abnormality according to an embodiment of the present disclosure. For example, an asymmetry index (ASI) can be calculated for each of head rotation, shoulder rotation, hip rotation, bow or bow legs, toes-in or toes-out, ear distance, chin distance, shoulder distance, hip distance, and knee distance. Each ASI score may be calculated according to the following formula and ranges from 0 to 10:

number

[0093] FIG. 31 shows a flowchart 3100 illustrating a posture correction exercise recommendation device according to an embodiment of the present disclosure. In step 3102, a pose analyzer (e.g., pose analyzer 406 / 506) can be utilized to calculate a postural abnormality score (e.g., ASI score) level and obtain all relevant body alignment deviations. In step 3104, the obtained deviations are ranked based on the ASI score. In step 3106, the top five deviations (e.g., the top five deviations with the highest ASI scores) are selected. In step 3108, one or more sets of personalized posture correction exercise programs are provided based on the top five deviations (e.g., recommending a personalized posture correction exercise program to correct the top five deviations). For example, exercises such as three-position toe raises, 45-degree neck stretches, hip rotations, press-ups, scapular presses, and triceps stretches can be recommended based on the ASI score for a personalized posture correction exercise program. In step 3110, the deviations, posture abnormalities, and set of exercise programs are reviewed with a clinician or healthcare professional.

[0094] Figure 32 illustrates an exemplary computing device 3200, hereinafter synonymous with computer system 3200, which may be used to perform the method of Figure 3. The exemplary computing device 3200 may be used to implement the device 400 shown in Figure 4. The following description of computing device 3200 is provided by way of example only and is not intended to be limiting.

[0095] 32, the exemplary computing device 3200 includes a processor 3204 for executing software routines. While a single processor is shown for clarity, the computing device 3200 may also include a multi-processor system. The processor 3204 is connected to a communications infrastructure 3206 for communicating with other components of the computing device 3200. The communications infrastructure 3206 may include, for example, a communications bus, crossbar, or network.

[0096] The computing device 3200 further includes a main memory 3208, such as random access memory (RAM), and a secondary memory 3210. The secondary memory 3210 may include a storage drive 3212, which may be, for example, a hard disk drive, a solid state drive, or a hybrid drive, and / or a removable storage drive 3214, which may include a magnetic tape drive, an optical disk drive, a solid state storage drive (such as a USB flash drive, a flash memory device, a solid state drive, a memory card, etc.). The removable storage drive 3214 reads from and / or writes to the removable storage medium 3218 in a well-known manner. The removable storage medium 3218 may include a magnetic tape, an optical disk, a non-volatile memory storage medium, etc., which is read from and written to by the removable storage drive 3214. As will be appreciated by those skilled in the art, the removable storage medium 3218 includes a computer-readable storage medium having computer-executable program code instructions and / or data stored thereon.

[0097] In alternative embodiments, secondary memory 3210 may additionally or alternatively include other similar means for allowing computer programs or other instructions to be loaded into computer device 3200. Such means may include, for example, removable storage 3222 and interface 3220. Examples of removable storage 3222 and interface 3220 include program cartridges and cartridge interfaces (such as those found in video game console devices), removable storage chips (such as EPROMs or PROMs) and associated sockets, removable solid-state storage devices (such as USB flash drives, flash memory devices, solid-state drives, memory cards, etc.), and other removable storage devices 3222 and interfaces 3220 that allow software and data to be transferred from removable storage 3222 to computer system 3200.

[0098] The computing device 3200 also includes at least one communications interface 3224. The communications interface 3224 allows software and data to be transferred between the computing device 3200 and external devices via communications path 3226. In various embodiments of the present invention, the communications interface 3224 allows data to be transferred between the computing device 3200 and a data communications network, such as a public or private data communications network. The communications interface 3224 can be used to exchange data between different computing devices 3200, where such computing devices 3200 form part of an interconnected computer network. Examples of communications interfaces 3224 can include a modem, a network interface (e.g., an Ethernet card), a communications port (e.g., serial, parallel, printer, GPIB, IEEE 1394, RJ45, USB), an antenna with associated circuitry, or the like. The communications interface 3224 can be wired or wireless. The software and data transferred via the communications interface 3224 are in the form of signals, which can be electronic, electromagnetic, optical, or other signals receivable by the communications interface 3224. These signals are provided to the communications interface via communications path 3226.

[0099] As shown in FIG. 32, the computing device 3200 further includes a display interface 3202 that performs operations to display images on an associated display 3230, and an audio interface 3232 that performs operations to play audio content via an associated speaker 3234.

[0100] The term "computer program product" as used herein may refer, in part, to removable storage medium 3218, removable storage device 3222, a hard disk attached to storage drive 3212, or software carrying a carrier wave over communications path 3226 (wireless link or cable) to communications interface 3224. A computer-readable storage medium refers to any non-transitory, non-volatile, tangible storage medium that provides recorded instructions and / or data to computing device 3200 for execution and / or processing. Examples of such storage media include magnetic tape, CD-ROM, DVD, Blu-ray disc, hard disk drive, ROM or integrated circuit, solid-state storage drive (such as a USB flash drive, flash memory device, solid-state drive, memory card), hybrid drive, magneto-optical disk, or computer-readable card such as a PCMCIA card, whether such device is internal or external to computing device 3200. Examples of transitory or non-tangible computer-readable transmission media that may also be involved in providing software, application programs, instructions and / or data to computing device 3200 include wireless or infrared transmission channels, network connections to other computers or networked devices, and the Internet or intranet, including email transmissions and recorded information on websites and the like.

[0101] Computer programs (also referred to as computer program code) are stored in main memory 3208 and / or secondary memory 3210. Computer programs may also be received via communications interface 3224. When such computer programs are executed, they enable computing device 3200 to perform one or more features of the embodiments described herein. In various embodiments, when the computer programs are executed, they enable processor 3204 to perform the features of the above-described embodiments. Thus, such computer programs represent controllers of computer system 3200.

[0102] The software may be stored on a computer program product and loaded into the computing device 3200 using the removable storage drive 3214, the storage drive 3212, or the interface 3220. The computer program product may be a non-transitory computer-readable medium. Alternatively, the computer program product may be downloaded to the computing system 3200 via communications path 3226. The software, when executed by the processor 3204, causes the computing device 3200 to perform the operations necessary to implement the methods shown in FIG.

[0103] It should be understood that the embodiment of Figure 32 is presented merely as an example to illustrate the operation and structure of device 400. Thus, in some embodiments, one or more features of computer device 3200 may be omitted. Also, in some embodiments, one or more features of computer device 3200 may be combined together. Furthermore, in some embodiments, one or more features of computer device 3200 may be split into one or more component parts.

[0104] It will be appreciated by those skilled in the art that numerous variations and / or modifications may be made to the present invention as illustrated in the specific embodiments without departing from the spirit or scope of the invention as broadly described, and the present embodiments are therefore considered in all respects to be illustrative and not restrictive. A part or all of the above-described embodiments can be described as, but not limited to, the following supplementary notes. (Appendix 1) detecting a rotation of one or more body parts of the person about the person's upright central axis; A method for identifying a postural state of a person, wherein an asymmetry score is calculated for one or more body parts based on a rotation of the one or more body parts of the person, the asymmetry score being associated with a level of the postural state of the person. (Appendix 2) 2. The method of claim 1, further comprising detecting an angle of a first line connecting two body part positions of the person from an image of the person relative to a reference line, and detecting a rotation of one or more body parts of the person around the person's upright central axis based on the angle. (Appendix 3) the first line connects a left body part position and a right body part position of a first body part of the person in a front view; 3. The method of claim 2, wherein the reference line includes a second line connecting a left body part position and a right body part position of a second body part of the person when viewed from the front. (Appendix 4) 3. The method of claim 2, wherein the reference line is a third line connecting two other body part positions of the person. (Appendix 5) 5. The method of any one of claims 1 to 4, further comprising detecting a first distance of a body part position from a nearest point along the person's standing axis, and detecting a rotation of one or more body parts of the person and / or calculating an asymmetry score for one or more body parts of the person based on the first distance. (Appendix 6) 5. The method of any one of claims 1 to 4, further comprising detecting a second distance of a body part position from one of (i) two body part positions of the person, (ii) a midpoint between the two body part positions, or (iii) a fourth line connecting the two body part positions, and detecting the rotation of one or more body parts of the person and / or calculating an asymmetry score for one or more body parts of the person is based on the second distance. (Appendix 7) 7. The method of any one of claims 1 to 6, further comprising detecting a plurality of body part positions of the person based on relative positions of a plurality of body parts in one or more images in which the person is detected, each of the plurality of body part positions corresponding to a body part of the person. (Appendix 8) 8. The method of claim 7, wherein detecting multiple body part positions of the person includes estimating a body part position of the person based on one of the multiple body part positions, and the multiple body part positions of the person further include the estimated body part position. (Appendix 9) 9. The method of any one of claims 1 to 8, further comprising receiving demographic data about the person, and wherein calculation of the asymmetry score is further based on the demographic data. (Appendix 10) the asymmetry score is one of a plurality of asymmetry scores associated with a plurality of body parts; 10. The method of any one of claims 1 to 9, further comprising calculating a level of the person's posture state based on the plurality of asymmetry scores. (Appendix 11) further comparing each of the plurality of asymmetry scores with other asymmetry scores of the plurality of asymmetry scores; identifying one or more asymmetry scores having a higher score among the plurality of asymmetry scores; 10. The method of claim 9, further comprising identifying a set of posture correction programs based on the results of the identification. (Appendix 12) at least one processor; at least one memory containing computer program code; At least one of the memories and computer program code are used by the at least one of the processors to provide the server with at least: detecting a rotation of one or more body parts of the person about a central upright axis of the person; 1. An apparatus for identifying a postural state of a person configured to: calculate an asymmetry score for one or more body parts based on a rotation of the one or more body parts of the person, the asymmetry score being related to a level of the postural state of the person. (Appendix 13) The at least one memory and the computer program code are transmitted to a server using at least one processor, and at least: Detecting an angle of a first line connecting two body part positions of the person from the image of the person relative to a reference line; 13. The apparatus of claim 12, further configured to detect a rotation of one or more body parts of the person about an upright central axis of the person based on the angle. (Appendix 14) the first line connects the position of a left body part and the position of a right body part of the first body of the person when viewed from the front; 14. The device of claim 13, wherein the reference line includes a second line connecting the position of a left body part and a right body part of the person's second body when viewed from the front. (Appendix 15) 14. The apparatus of claim 13, wherein the reference line is a third line connecting the positions of two other body parts of the person. (Appendix 16) The at least one memory and the computer program code are transmitted to the server using at least one processor, at least: Detecting a first distance of a body part position from a nearest point along the person's standing center axis; 16. The apparatus of any one of claims 12 to 15, further configured to detect a rotation of one or more body parts of the person and / or calculate an asymmetry score for one or more body parts of the person based on the first distance. (Appendix 17) The at least one memory and the computer program code are transmitted to a server using at least one processor, and at least: Detecting a second distance of the body part location from one of (i) two body part locations of the person, (ii) a midpoint between the two body part locations, or (iii) a fourth line connecting the two body part locations; 16. The apparatus of any one of claims 12 to 15, further configured to detect a rotation of one or more body parts of the person, and / or calculate an asymmetry score for one or more body parts of the person based on the second distance. (Appendix 18) The at least one memory and the computer program code are transmitted to a server using at least one processor, and at least: 18. The apparatus of any one of claims 12 to 17, further configured to detect a plurality of body part positions of the person based on relative positions of a plurality of body parts in one or more images in which the person is detected, each of the plurality of body part positions corresponding to a body part of the person. (Appendix 19) The at least one memory and the computer program code are transmitted to a server using at least one processor, and at least: 19. The apparatus of claim 18, further configured to estimate a body part position of the person based on one of the plurality of body part positions, wherein the plurality of body part positions of the person further comprises the estimated body part position. (Appendix 20) The at least one memory and the computer program code are provided to the server using at least one processor to perform at least: receiving demographic data about the person; 20. The apparatus of any one of claims 12 to 19, further configured to further calculate the asymmetry score based on the demographic data. (Appendix 21) At least one of the memories and the computer program code are provided to a server using at least one processor to execute at least: 21. The apparatus of any one of claims 12 to 20, further configured to calculate a level of a posture state of the person based on the plurality of asymmetry scores. (Appendix 22) The at least one memory and the computer program code are transmitted to a server using at least one processor, and at least: comparing each of the plurality of asymmetry scores to other asymmetry scores of the plurality of asymmetry scores; identifying one or more asymmetry scores having a higher score among the plurality of asymmetry scores; 22. The apparatus of claim 21, further configured to identify a set of posture correction programs based on a result of the identification. (Appendix 23) An apparatus according to any one of claims 12 to 22; one or more image capture devices configured to capture one or more images of the person; A system for identifying a postural state of the person, wherein the one or more images include an image of the person viewed from the front and / or an image of the person viewed across a sagittal plane.

Claims

1. a landmark detection system determining a set of two-dimensional and / or three-dimensional body landmark locations based on the one or more captured images (excluding the one two-dimensional image); a computing device detecting a rotation of one or more body parts of the person about the person's central upright axis based on the set of body landmark positions; 1. A method of computing an asymmetry score for one or more body parts of the person based on a rotation of the one or more body parts, the asymmetry score identifying a postural state of the person associated with a level of the postural state of the person, the method comprising: A method comprising: detecting an angle of a first line connecting a left body part position and a right body part position of a first body part of the person when viewed from the front, from an image of the person relative to a reference line including a second line connecting a left body part position and a right body part position of a second body part of the person when viewed from the front; and detecting a rotation of one or more body parts of the person around the person's upright central axis based on the angle.

2. a landmark detection system determining a set of two-dimensional and / or three-dimensional body landmark locations based on the one or more captured images (excluding the one two-dimensional image); a computing device detecting a rotation of one or more body parts of the person about the person's central upright axis based on the set of body landmark positions; 1. A method of computing an asymmetry score for one or more body parts of the person based on a rotation of the one or more body parts, the asymmetry score identifying a postural state of the person associated with a level of the postural state of the person, the method comprising: A method comprising: detecting an angle of a first line connecting a left body part position and a right body part position of a first body part of the person when viewed from the front from an image of the person relative to a reference line, the reference line being a third line connecting two other body part positions of the person; and detecting a rotation of one or more body parts of the person around an upright central axis of the person based on the angle.

3. 3. The method of claim 1 or 2, further comprising detecting a first distance from the person's central upright axis to a body part position that is the shortest, and detecting a rotation of one or more body parts of the person and / or calculating an asymmetry score for one or more body parts of the person based on the first distance.

4. 3. The method of claim 1 or 2, further comprising detecting a second distance according to a distance between two body part positions of the person, and detecting a rotation of one or more body parts of the person and / or calculating an asymmetry score for one or more body parts of the person based on the second distance.

5. 3. The method of claim 1, further comprising detecting a plurality of body part positions of the person based on relative positions of a plurality of body parts in one or more images in which the person is detected, each of the plurality of body part positions corresponding to a body part of the person.

6. 5. The method of claim 4, wherein detecting a plurality of body part positions of the person estimates a body part position of the person based on one of the plurality of body part positions, and the plurality of body part positions of the person further comprises the estimated body part position.

7. at least one processor; and at least one memory containing computer program code, the computer device comprising at least: a landmark detection system detecting a rotation of one or more body parts of the person about a central upright axis of the person based on a set of two-dimensional and / or three-dimensional body landmark positions determined based on one or more captured images (excluding one two-dimensional image); 1. An apparatus for identifying a postural state of a person configured to: calculate an asymmetry score for one or more body parts based on a rotation of the one or more body parts of the person, the asymmetry score being related to a level of the postural state of the person; An apparatus for detecting an angle of a first line connecting a left body part position and a right body part position of a first body part of the person when viewed from the front, from an image of the person relative to a reference line including a second line connecting a left body part position and a right body part position of a second body part of the person when viewed from the front, and detecting a rotation of one or more body parts of the person around an upright central axis of the person based on the angle.

8. at least one processor; and at least one memory containing computer program code, the computer device comprising at least: a landmark detection system detecting a rotation of one or more body parts of the person about a central upright axis of the person based on a set of two-dimensional and / or three-dimensional body landmark positions determined based on one or more captured images (excluding one two-dimensional image); 1. An apparatus for identifying a postural state of a person configured to: calculate an asymmetry score for one or more body parts based on a rotation of the one or more body parts of the person, the asymmetry score being related to a level of the postural state of the person; An apparatus for detecting an angle of a first line connecting a left body part position and a right body part position of a first body part of the person when viewed from the front from an image of the person relative to a reference line, the reference line being a third line connecting two other body part positions of the person, and detecting a rotation of one or more body parts of the person around an upright central axis of the person is based on the angle.

9. An apparatus according to claim 7 or 8; one or more image capture devices configured to capture one or more images of the person; A system for identifying a postural state of the person, wherein the one or more images include a frontal image of the person and / or a sagittal image of the person.

Images