49 results about "Motion reconstruction" patented technology

Disclosed herein is a motion capture apparatus and method. The motion capture apparatus includes a three-dimensional (3D) appearance reconstruction unit for reconstructing a 3D appearance of an entity using a multi-view image obtained by capturing the entity at a T+1-th frame. A 3D skeleton generation unit generates a 3D skeleton of the T+1-th frame using the 3D appearance and a model of the entity having a motion of a T-th frame. A motion reconstruction unit reconstructs a motion of the T+1-th frame of the model by changing the motion of the model based on the 3D skeleton of the T+1-th frame.

A multifunctional intelligent rehabilitation training and assessment system comprises a control terminal, a mechanical arm, a force sensor, a position sensor, a myoelectricity sensor and a 3D printing support, wherein the mechanical arm is used for rehabilitation training of upper limbs and lower limbs; the 3D printing support is used for fixing different joints; the force sensor is arranged at one end, which is in contact with a patient, of the mechanical arm; the position sensor is mounted on the mechanical arm and can record the space position of the mechanical arm; the myoelectricity sensor is attached to the muscle body of the patient; the control terminal sends down a rehabilitation instruction to the mechanical arm by acquiring and analyzing the information of the force sensor and the information of the myoelectricity sensor on the basis of built-in parameters, and applies corresponding rehabilitation training actions to the patient; in a rehabilitation training action process, the control terminal acquires the position information of the mechanical arm in real time, and records a muscle force, which is fed back by the myoelectricity sensor, of the patient and a force which is fed back by the force sensor and is applied by the patient in a training process, and the rehabilitation training mode is adjusted in real time on that basis. The mechanical arm is small in size, low in cost and high in universality, and has good nerve and motion reconstruction rehabilitation effect due to the force and myoelectricity feedback function.

The invention relates to a dance training evaluation system based on a digitized place and a wireless motion capture device. Two-dimensional plantar pressure distribution information and space posture information of a dancer are synchronously collected through the digitized place and the wireless motion capture device in real time, and motion information of the dancer is obtained through information fusion and sent to a computer through the Ethernet; the computer drives a human body model to move in a three-dimensional space through data of the motion information, the motion reconstruction of the dancer is achieved, and meanwhile biomechanics parameters and kinematics parameters of motions are extracted, analyzed and displayed; dancing postures of the dancer are evaluated online by establishing a dancing motion template and a knowledge base, and guidance and demonstration are achieved for the dancer in real time. On the basis of the movable digitized place which can be spliced and the wearable wireless motion capture device, the motion information of the dancer can be captured on the premise of not influencing the dancing motions of the dancer, and the dance training evaluation system is suitable for teaching training and evaluation of various dances.

The invention relates to a human motion prediction method based on attention mechanism, belonging to the technical field of human-computer interaction and virtual reality. The method of the inventionreads all the data from the human body movement posture data set and performs standardization processing, then reads the training data batch according to the generated random number according to the length of the frame needed to be trained in the instruction, inputs the data into the attention model for data processing, and performs motion reconstruction according to the processed data. The invention adopts two kinds of attention mechanisms, namely global attention mechanism and local attention mechanism. Compared with the prior art, the method of the invention can avoid information redundancyin a certain time and distribute attention under a certain window at the same time.

The invention relates to a rat leg motion reconstruction test method based on spinal nerve function electrical stimulation. Corresponded three-dimensional coordinates based on spinal segments are established while rat spinal cord is focused, a tester can find out an electrode implanting position rapidly and accurately, and the success rate of tests is increased; excitation pulses are transmitted to two electrodes inserted into the rat spinal cord alternately, coherent stretching motions of the rat leg can be implemented, and the tester can grasp the process of rat spinal cord electrical stimulation more directly; meanwhile, the electromyographic signals of rat leg muscles can be collected through an electromyographic monitor and are used for comparison and analysis. By the aid of test method, the rat leg stretching motion reconstruction based on the spinal nerve function electrical stimulation can be completed, and important significance is provided for the fields of scientific research and teach; the operating ability of the tester is improved, the tester can grasp complex spinal cord stimulation tests which are difficult to grasp more easily, and the further academic research of the tester can be benefited.

The invention discloses a complex behavior recognition method. The complex behavior recognition method comprises: acquiring three-dimensional bone joint point information of a target motion by using asensor; pre-processing the joint point information, and normalizing a coordinate system; extracting a motion trajectory of each joint point, projecting the motion trajectory to three two-dimensionalplanes; extracting a motion vector between each two frames and the length and the direction angle thereof, using the k-means algorithm to perform clustering to acquire motion primitives, and performing counting to acquire histograms; calculating the weight of each joint point is calculated to form a descriptor according to the time pyramid, the time information, and values of the clusters of allthe histograms; performing SVM classification to achieve motion recognition. The invention can perform feature extraction and effective represent on the motion bone joint point information, and can improve the accuracy of the motion recognition; all the motion information is completely retained, and the motion reconstruction can be performed; all motion classes are clustered, and motion features of the human are captured globally; low-level features are used, the computational difficulty is lowered, the motion recognition efficiency is improved, and the the real-time requirements of the systemis satisfied.

The invention discloses a method and a system for restructuring the motion of a three-dimensional gesture. The method comprises the following steps of: performing region segmentation on a first frame image of an acquired gesture image sequence; generating an affine transformation matrix of each segmentation region; generating projection coefficient from three dimensions to two dimensions; obtaining a three-dimensional gesture model corresponding to the first frame image according to the projection coefficient and the affine transformation matrix; determining a framework node of the obtained three-dimensional gesture model corresponding to the first frame image and the freedom degree of the framework node; and performing simulated annealing particle filtering operation by combining the current frame image based on the framework node and the freedom degree of the framework node of the three-dimensional gesture model corresponding to a front frame image in allusion to follow frame imagesto obtain the three-dimensional gesture model of the current frame image, and thus realizing the restructuring of the three-dimensional gesture. According to the method and the system, the accurate modeling of a three-dimensional hand by using a laser scanner is not required, and the complexity of the motion restructuring process is reduced.

The invention discloses a reinsman motion posture capturing and three-dimensional reconstruction system based on the inertial sensing technology, and the system comprises a reinsman motion posture capturing unit which comprises a plurality of inertial sensors distributed on the body of a reinsman, a controller, and a wireless transmitting module for data output; a three-dimensional virtual character motion reconstruction unit which comprises a wireless receiving module for receiving the information, and a human body model real-time reconstruction module connected with the wireless receiving module; and a reinsman evaluating logic frame unit which is connected with the human body model real-time reconstruction module and comprises a reinsman state data display module and a posture calibration module. According to the invention, the system prestores the standard motion postures of the Chinese reinsman, drives the movement of the virtual three-dimensional model through the data of sensor posture analysis, finds out the wrong postures of the reinsman, carries out the targeted adjustment, achieves the coupling research of the human and horse motion postures, and improves the score of a sportsman.

Systems and methods for conformal imaging vibrometry capable of real-time measurements of the dynamic motions of any arbitrary two-dimensional or three-dimensional structure. The systems and methods are able to fully characterize the dynamic behavior of an object of any arbitrary geometry. The test object is illuminated with multiple laser beams whose directions conform to the local normal axis of the surface. The approach enables high-speed vibration imaging of whole-body dynamics of arbitrarily shaped structures in real-time, with no multiplexed data capture or synthesized motion reconstruction, as is currently practiced. By measuring the object's vibrations simultaneously at multiple points, the disclosed systems and methods are able to reproduce the structural behavior under operational conditions, which can then be spectrally decomposed to determine the modal, complex modal and transient nature of the true structural dynamics.

The present application discloses a method and a device for obtaining video human body movements. The method comprises the following steps of: acquiring calibration information of a camera; acquiringvideo information photographed by a camera, wherein the video information comprises a plurality of different unique identification points of an encoded pattern, the identification points being arranged on a target part of a human body for marking a moving state of the target part of the human body, each encoded pattern comprising at least four feature points; obtaining target information of an identification point, wherein the target information comprises at least: a coding pattern of the identification point and spatial structure information of the coding pattern; according to the calibrationinformation and the target information of the marking point, the video information is reconstructed to obtain the spatial posture of the marking point in the camera coordinate system, so as to determine the movement of the target part of the human body. In the related art of human motion capture, the invention solves the technical problems that the motion capture system cannot simultaneously giveconsideration to the simple structure, the convenient operation and the accurate motion reconstruction.

Systems and methods for conformal imaging vibrometry capable of real-time measurements of the dynamic motions of any arbitrary two-dimensional or three-dimensional structure. The disclosed systems and methods are able to fully characterize the dynamic behavior of an object of any arbitrary geometry. The test object is illuminated with multiple laser beams whose directions conform to the local normal axis of the surface. The approach enables high-speed vibration imaging of whole-body dynamics of arbitrarily shaped structures in real-time, with no multiplexed data capture or synthesized motion reconstruction, as is currently practiced. By measuring the object's vibrations simultaneously at multiple points, the disclosed systems and methods are able to reproduce the structural behavior under operational conditions, which can then be spectrally decomposed to determine the modal, complex modal and transient nature of the true structural dynamics.

The invention discloses a video saliency detection method, which comprises the following steps: determining a background candidate region by considering background clues, color compactness and color uniqueness, forming a static reconstruction dictionary to reconstruct superpixels in each video frame, and obtaining a static saliency map; determining a background seed point set by considering the motion compactness and the motion uniqueness, forming a motion reconstruction dictionary to reconstruct superpixels in each video frame, and obtaining a motion saliency map; fusing the static saliency map and the motion saliency map to obtain a single-frame saliency result; obtaining an inter-frame saliency map by utilizing bidirectional sparse propagation; and constructing an energy function composed of the unitary data item, the space-time smoothing item, the space mutual exclusion item and the global item, and optimizing the single frame and inter-frame saliency result through the energy function. According to the method, the salient target in the video sequence can be accurately extracted, the background inhibition capability is high, and the contour of the salient target is clear. The designed model has good robustness, and can process a plurality of challenging scenes.

A control system for a movement reconstruction and / or restoration system for a patient, comprising a CNS-Stimulation Module, especially an EES-Module, configured and arranged to provide CNS-Stimulation to a patient, and / or a PNS-Stimulation Module, especially an FES-Module, configured and arranged to provide PNS-Stimulation to a patient, a controller configured and arranged to control the CNS-Stimulation Module and / or the PNS-Stimulation Module, and at least one sensor configured and arranged to measure at least one parameter indicative of the movement of at least one limb and / or part of a limb of a patient.

A method and apparatus for obtaining a video action based on an identification point are disclosed. The method comprises the following steps of: acquiring calibration information of a camera, wherein,the calibration information of the camera at least comprises a lens focal length, an imaging center, a tilt factor and distortion parameters; acquiring video information captured by a camera, whereinthe video information includes at least an encoded pattern of identification points arranged on a target object for marking a moving state of the target object; obtaining target information of an identification point, wherein the target information comprises at least: a coding pattern of the identification point and spatial structure information of the coding pattern; according to the calibrationinformation and the target information of the identification point, the three-dimensional reconstruction of the video information being carried out, and the spatial posture change of the identification point in the camera coordinate system being obtained. The invention solves the problems that the motion capture system in the related art cannot simultaneously give consideration to the simple structure, the convenient operation and the accurate motion reconstruction.

The invention discloses an interactive motion data acquisition method based on a portable terminal. A common portable tablet device is adopted, and a series of program modules provided for users to complete image acquisition, image marking, marking information recognition and motion reconstruction are arranged. The interactive motion data acquisition method enables a user to conveniently make animation reproducing various creatures without expensive motion capture equipment, breaks through the limitation of acquisition objects of traditional motion capture equipment and provides professionals with a method of carrying out motion data acquisition and reconstruction work more quickly. The tool interface is friendly. The method is applicable to various levels of demands from entertainment to development.

A control system for a movement reconstruction and / or restoration system for a patient, comprisinga sampling module configured and arranged to sample signals describing directly and / or indirectly motion at a sampling rate of at least 50 Hz;at least one stimulation system configured and arranged to provide stimulation for movement reconstruction and / or restoration to the patient;a prediction module configured and arranged to provide a prediction of at least a next movement, especially movement stage and / or sequence, to reduce latency and to synchronize stimulation to the movement phase,wherein the control system further comprises at least one controller, the controller being configured and arranged to provide stimulation control signals to the stimulation system on the basis of the information obtained by the sampling module and the prediction provided by the prediction module.

Systems and methods for conformal imaging vibrometry capable of real-time measurements of the dynamic motions of any arbitrary two-dimensional or three-dimensional structure. The systems and methods are able to fully characterize the dynamic behavior of an object of any arbitrary geometry. The test object is illuminated with multiple laser beams whose directions conform to the local normal axis of the surface. The approach enables high-speed vibration imaging of whole-body dynamics of arbitrarily shaped structures in real-time, with no multiplexed data capture or synthesized motion reconstruction, as is currently practiced. By measuring the object's vibrations simultaneously at multiple points, the disclosed systems and methods are able to reproduce the structural behavior under operational conditions, which can then be spectrally decomposed to determine the modal, complex modal and transient nature of the true structural dynamics.

The invention discloses a hand geometric motion reconstruction method and device based on a multi-voxel block. The method comprises a step of acquiring a human hand skeleton according to depth data ofa first frame and initializing the multi-voxel block, a step of performing motion estimation on a hand through a reconstructed hand model in the multi-voxel block to obtain new depth data, and a stepof fusing the new depth data into the multi-voxel block and updating a geometric model of the hand so as to simultaneously reconstruct the motion and geometric models of the hand. According to the method, the geometric model and motion information of the human hand can be reconstructed at the same time, the dependence on a parametric model in a traditional human hand motion reconstruction processis avoided, the robustness of the human hand reconstruction technology is enhanced and the application range of the human hand reconstruction technology is expanded, and the development of the humanhand reconstruction technology is promoted.

Systems and methods for conformal imaging vibrometry capable of real-time measurements of the dynamic motions of any arbitrary two-dimensional or three-dimensional structure. The disclosed systems and methods are able to fully characterize the dynamic behavior of an object of any arbitrary geometry. The test object is illuminated with multiple laser beams whose directions conform to the local normal axis of the surface. The approach enables high-speed vibration imaging of whole-body dynamics of arbitrarily shaped structures in real-time, with no multiplexed data capture or synthesized motion reconstruction, as is currently practiced. By measuring the object's vibrations simultaneously at multiple points, the disclosed systems and methods are able to reproduce the structural behavior under operational conditions, which can then be spectrally decomposed to determine the modal, complex modal and transient nature of the true structural dynamics.

A control system for a movement reconstruction and / or restoration system for a patient, comprisingat least one sensor,at least one controller,at least one programmer,at least one stimulation system,wherein the controller is connected with the sensor, the programmer and the stimulation system,wherein the sensor is part of or attached to a training entity in order tocreate and / or guide a movement model for a patient and / oradjust stimulation settings based on sensor input.

The invention discloses a motion data redirection method based on a neural network. The motion data redirection method comprises the following steps of based on motion data represented by joint coordinates, training by taking a motion reconstruction error as a loss function to obtain an auto-encoder; acquiring source motion data, calculating hidden variables corresponding to the source motion datathrough the auto-encoder, and reconstructing motion; applying a bone length constraint, a footprint constraint, a root joint position constraint and a bone angle constraint to the reconstruction motion, and reversely propagating constraint loss to an implicit variable space to optimize the implicit variable; through multiple iterations, the auto-encoder completing motion redirection from the source bone to the target bone. According to the method, the hidden variable corresponding to the source motion data is calculated through the auto-encoder, the motion is reconstructed, then various constraints are applied to the reconstructed motion, motion reconstruction loss is propagated back to the hidden variable space to optimize the hidden variable of the auto-encoder, the redirected motion isfinally obtained through multiple iterations, and a good motion redirection effect is achieved.

The invention provides a temporal-mandibular joint motion reconstruction method based on multi-modal information fusion. The temporal-mandibular joint motion reconstruction method is characterized by comprising the following steps: making an upper and lower jaw marking device; acquiring and processing mandibular movement data; acquiring and processing CBCT data images; fusing the mandibular movement data and the CBCT image data; a multi-jaw-position biteplate is manufactured; acquiring and processing an MRI data image; the mandibular movement data, CBCT image data and MRI image data are registered and fused, and mandibular movement is visually reproduced in a 3D model and image mode. According to the method, the ternary relative motion relation and the space posture of the temporomandibular glenoid fossa, the joint disc and the condyle process are accurately recorded, the real functional motion state of the temporomandibular joint is revealed in a visual and quantitative mode, and the method can be used for clinical research of the temporomandibular joint, serves as an auxiliary means for diagnosis, treatment and prognosis evaluation of temporomandibular joint diseases and has a good application prospect. Clinical operation is easy and convenient, the difficulty that a clinician judges the position of the joint disc can be lowered, and the accuracy of diagnosis and treatment of temporomandibular joint diseases is improved.

The invention discloses a method and a system for extracting a key frame from motion capture data and reconstructing motion. According to the invention, the distance between quaternions is used to represent the human body posture difference; the total change on each joint of the human body is used as a frame spacing; the first motion frame is taken as a first key frame; the difference between thecurrent frame and the last key frame is calculated through continuous iteration, frames of which the difference is smaller than the threshold value are eliminated, frames of which the difference exceeds the threshold value are used as key frames and are extracted and stored, and then the extracted key frame set is subjected to motion reconstruction by adopting a quaternion spherical interpolationmethod, so that the original motion capture data has a relatively high compression ratio, and the trueness and fidelity of original animation restoration are ensured.

The invention belongs to the technical field of human body measurement, and relates to a human body measurement method based on motion reconstruction. The process comprises the following steps: step 1, collecting images around a human body by using a mobile phone camera; step 2, calculating a scene three-dimensional model of the acquired images around the human body by using a motion reconstruction method; step 3, removing a background and the ground in the scene three-dimensional model so as to extract a human body three-dimensional model; step 4, zooming the human body three-dimensional model to a real scale; step 5, selecting a to-be-measured part on the human body model for measurement; according to the method, professional image information acquisition equipment is not needed, no specific skill requirement is needed for information acquisition personnel, the human body three-dimensional model can be extracted and adjusted to the human body real proportion, and the human body integrating degree and comfort degree of clothes personalized design are improved; the method has a wide application prospect in the fields of clothes personalized design, electronic commerce, virtual fitting, medicine and the like.

The invention relates to a large-scale real-time human motion acquisition reconstruction system, which comprises a plurality of sets of node sensors, a hub, a data receiver, a processor, a motion reconstruction module, a memory and a cloud server, wherein each group of node sensors is used for acquiring a gravity acceleration, a rotation rate and magnetic data and transmitting the data to the hub;the hub transmits the collected data to the data receiver; the data receiver transmits the received data to the processor through a USB virtual serial port; and the processor performs fusion solutionon the sensor data, and the data are transmitted to the motion reconstruction module for real-time human motion reconstruction, the data are compressed at the same time, the compressed data are savedto the memory and uploaded to the cloud server. The system provided by the invention can realize large-scale human motion acquisition with low cost and high precision, real-time monitoring on the acquisition process is carried out, and real-time efficient storage and transmission of human motion data are realized.

The invention discloses a virtual evaluation method, device and equipment for pedestrian head protection performance of a vehicle. The method comprises the steps that accident information in an accident site is collected; wherein the accident information comprises pedestrian information and vehicle information in an accident site; person according to accident information Vehicle collision motionreconstruction to reproduce people-in the accident scene A collision form of vehicle collision; analysis of real person through collision morphology The distribution rule of vehicle collision initialboundary conditions; depending on pedestrian models and the real person- And establishing a virtual evaluation model of the pedestrian head protection performance of the vehicle according to the distribution rule of the vehicle collision initial boundary conditions so as to evaluate the pedestrian head protection performance of the real vehicle. By establishing the virtual evaluation model of thepedestrian head protection performance of the vehicle, the pedestrian head protection performance of the real vehicle is evaluated, and the traffic damage of the vehicle to the pedestrian head can beeffectively prevented.

The invention relates to a rat leg motion reconstruction test method based on spinal nerve function electrical stimulation. Corresponded three-dimensional coordinates based on spinal segments are established while rat spinal cord is focused, a tester can find out an electrode implanting position rapidly and accurately, and the success rate of tests is increased; excitation pulses are transmitted to two electrodes inserted into the rat spinal cord alternately, coherent stretching motions of the rat leg can be implemented, and the tester can grasp the process of rat spinal cord electrical stimulation more directly; meanwhile, the electromyographic signals of rat leg muscles can be collected through an electromyographic monitor and are used for comparison and analysis. By the aid of test method, the rat leg stretching motion reconstruction based on the spinal nerve function electrical stimulation can be completed, and important significance is provided for the fields of scientific research and teach; the operating ability of the tester is improved, the tester can grasp complex spinal cord stimulation tests which are difficult to grasp more easily, and the further academic research of the tester can be benefited.

The invention relates to the technical field of computer vision, in particular to a mouse movement reconstruction method and device based on a multi-view camera. The mouse motion reconstruction method based on the multi-view camera comprises the following steps: acquiring a color image of a mouse and a mouse point cloud under a world coordinate system; detecting key points of a mouse in the color image based on a convolutional neural network; and performing three-dimensional reconstruction on the movement of the mouse by using a reconstruction algorithm based on an LM algorithm. According to the technical scheme, the movement data of the mouse is automatically collected, the data collection process is digitalized, and the movement of the mouse is reconstructed in combination with the point cloud data and the key point detection data of the mouse, so that biological researchers are helped to better capture the movement information of the mouse, and the accuracy of the researchers is improved. And downstream mouse motion analysis tasks of researchers are also facilitated.

The invention discloses a space nutation target parameter estimation method based on broadband radar observation, which mainly solves the problem of nutation parameter estimation of an unknown target shape or an unknown scattering center position. The method is realized by the following steps that a broadband echo sequence of a space nutation target is obtained by a radar; after motion compensation is performed on the echo sequence, a high-resolution radial-distance sequence of a plurality of static scattering centers of the target is obtained, and spectral estimation is performed on the high-resolution radial-distance sequence, so as to obtain initial values of coning angular frequency, spin angular frequency and nutation angular frequency; a space motion model of the target is established in accordance with a geometric orientation diagram of the target and the line of sight of the radar; European three-dimensional reconstruction is performed on the high-resolution radial-distance sequence, so as to obtain a European motion reconstruction matrix; and according to the European motion reconstruction matrix, the space motion model and the initial values of the angular frequencies, an iterative optimization algorithm is adopted to estimate the coning angular frequency, the spin angular frequency, the nutation angular frequency, an amplitude of swing, an initial phase and a central angle of swing. The space nutation target parameter estimation method has the advantage of good estimation performance and can be used for space target recognition of the radar.