360 results about "Human motion" patented technology

Provided in the present invention is a reconstruction and driving method based on a multi-view motion video of a clothed human body. The method comprises: using a two-layer three-dimensional Gaussian representation to model a human body and corresponding clothes; defining geometric attributes of three-dimensional Gaussian ellipsoids for any face on coarse three-dimensional meshes of the human body and the clothes; at a reconstruction stage, transforming coarse triangular meshes of the clothes and the body from a canonical space to an observation space, and updating Gaussian geometric attributes; and on the basis of a camera pose and parameters of an input video, rasterizing a three-dimensional Gaussian representation of the observation space to an image space, and rendering a picture on the basis of a Gaussian splatting formula. Self-supervised training can be performed on a model by means of calculating the color difference between a rendered picture and a real picture and using same as an optimization objective; when reconstruction is completed and driving is performed, a geometric result of a current frame is used to calculate the position of the coarse triangular mesh of the clothes in the next frame; and the reconstruction-stage method and related parameters are used to render a driving picture in a new posture.

The invention relates to a human body posture estimation key point correction method based on skeleton proportion constraint, and belongs to the technical field of human body posture estimation key point detection. The method comprises the following steps: extracting key points of a human body in a human body motion video, forming a skeleton based on the key points, calculating a skeleton proportion value of each skeleton length and a thoracolumbar spine length in the skeleton, and defining a frame corresponding to the skeleton proportion value not within a preset threshold as an invalid frame; iterating the positions of the key points in the invalid frame until the bone proportion values formed by all key point sequences in the invalid frame are within a preset threshold value or reach a preset maximum number of iterations; and detecting the widths of the four limbs based on the iterated key points, adjusting the positions of the key points to the axis positions corresponding to the width centers of the four limbs to obtain the adjusted key points, and reconstructing a key point sequence to realize key point correction. The objective of the invention is to solve the technical problem of low key point detection accuracy caused by deviation in key point positioning in the prior art.

Approaches presented herein provide for the use of reinforcement learning to fine-tune a generative model, such as a motion diffusion model, for a specific objective, such as to generate representations of human motion corresponding to provided text input. A discriminator can be used to guide the training of the generative model. In at least one embodiment, the discriminator can compare the input text and generated motion representation (or embeddings of each) to determine an alignment value or match score, for example, which can then be used to adjust the network parameters or weights of the generative model to improve the alignment between input text and generated motion.

The invention relates to the technical field of pressure monitoring, and discloses a method and system for motion state analysis according to plantar pressure, and the method comprises the steps: receiving multi-modal biomechanical data related to a landing event; extracting a corresponding motion feature vector from the multi-modal biomechanical data; and inputting the motion feature vector into a motion state analysis model for identification to obtain a corresponding damage risk index, and outputting the corresponding damage risk index. By receiving pressure distribution data detected by the plantar pressure sensing array, shear force data detected by the shear force measuring and calculating module and a kinematics data sequence detected by the inertial measurement unit, biomechanical information in the movement process can be comprehensively captured from multiple dimensions; compared with a single data acquisition mode, the fusion of the multi-modal data can reflect the real state of the human body in motion more accurately, and a richer and more accurate data basis is provided for subsequent motion state analysis.

The invention relates to the technical field of posture recognition, in particular to a motion posture recognition method and system based on deep learning, and the method comprises the following steps: based on a human body motion image sequence, analyzing a skeleton center space trajectory, adjusting an acquisition visual angle, recognizing staged skeleton nodes, and generating a node time sequence trigger sequence; and optimizing joint point angle feature grouping and expression to obtain a hierarchical feature expression structure. In the invention, through detection for interference dynamic change characteristics, visual angle adjustment and node staged response in a time sequence are combined, compensation reasoning of a space trajectory is enhanced, and the hierarchical expression capability of motion capture is improved; the input features have the advantages of actively screening background disturbance, dynamically adapting the motion direction, separating action core nodes and perfecting node response association and grouping angle time sequence structures, the integrity and identification degree of action feature structure expression and hierarchical modeling are guaranteed, and higher identification accuracy and complex environment adaptability are brought.

The invention relates to a tumble risk assessment method and system based on a multi-modal deep learning network, and relates to the technical field of human body posture recognition, and the method comprises the steps: collecting a scene motion image sequence, a plantar pressure feature sequence and personnel basic information; analyzing the scene motion image sequence by using a human body posture estimation network model to determine a human body key feature point sequence; performing feature extraction on the human body key feature point sequence and the scene motion image sequence to generate human body posture sequence features; performing feature extraction fusion on the human body key feature point sequence and the plantar pressure feature sequence to generate human body motion fusion time sequence features; performing feature extraction on the personnel basic information to generate personnel basic information features; and inputting the human body posture sequence features, the human body motion fusion time sequence features and the personnel basic information features into a multi-modal deep learning network for analysis so as to determine a fall risk score. The method and the device have the effect of improving the precision of fall risk assessment.

The invention provides a body function evaluation method and device based on multi-modal sensing fusion. Real-time detection and analysis of human motion and postures are achieved through a multi-modal sensing technology. According to the method, the test process is standardized, operation is automatic, scoring is objective, the accuracy and repeatability of the test are improved, and manual intervention is reduced. Through visualization and voice guidance, action demonstration and prompts are provided on a large screen, it is ensured that a testee completes a test autonomously, and the operation burden of workers is reduced. Meanwhile, the system automatically calculates scores and generates reports according to a standard SPPB scoring rule, provides result interpretation and rehabilitation suggestions, and remarkably improves the evaluation efficiency and practicability. In addition, a personalized video exercise prescription is generated according to the test score, so that the testee can track the training progress conveniently. The closed loop of detection and intervention services is ensured, and the rehabilitation effect is improved.

The invention discloses an end joint precision enhancement method and device of an unmarked motion capture system, and relates to the technical field of human motion capture, and the method comprises the steps that a main system collects original motion capture data to obtain a joint initial pose of the whole body of a detected object; an enhancement module is installed at the tail end joint of the measured object, and the enhancement module at least comprises an optical mark point and an inertial sensor; performing spatial position identification on the tail end joint of the detected object to obtain first tail end joint position data, and synchronously acquiring second tail end joint position data of the enhancement module; performing fusion filtering processing on the first tail end joint position data and the second tail end joint position data to obtain high-precision position data of the tail end joint; and the enhanced tail end joint motion data of the enhancement module is collected, and the target tail end joint motion data is output by combining the enhanced tail end joint motion data and the high-precision position data. The method has the effect of improving the accuracy and robustness of the motion capture system.

The invention provides an intelligent motion generation method, an intelligent motion generation system, electronic equipment and a computer readable storage medium, in the intelligent motion generation method, motion feature data are mapped and aligned to a virtual human skeleton structure, a preliminary motion sequence is obtained, and in addition, the motion feature data are mapped and aligned to the virtual human skeleton structure; and according to the virtual human skeleton structure and the preliminary action sequence, optimizing the action of the virtual human skeleton structure, and outputting an optimized action sequence, so that after the preliminary action sequence is obtained according to a redirection strategy based on the frame sequence of the video of the human body motion, the motion of the virtual human skeleton structure is optimized. And an optimized action sequence obtained by optimizing the preliminary action sequence can be more matched with the body shape of the virtual human.

The invention discloses a video generation method and system based on a scene adaptive track and mirror moving control, and belongs to the technical field of computer vision. The method comprises the following steps: firstly, aligning a scene point cloud and a parameterized human motion sequence to a unified coordinate system, generating a ground adaptive three-dimensional track according to a two-dimensional track input by a user and scene geometry, and redirecting the motion; projecting the three-dimensional information into a two-dimensional condition graph based on camera parameters; and finally, in a conditional diffusion model generation process, selectively fusing scene geometric condition information after noise addition into the current denoising latent variable by utilizing a visibility mask generated by the depth map, so as to realize view angle self-adaptive mirror operation control. According to the method, the problems of conflict between character motion and scene, inflexible track control and inconsistent background under camera motion are solved, and a controllable video which is physically consistent and visually real can be generated.

The invention discloses a 3D human body action recognition method based on geometry-contour multi-frequency, relates to the technical field of computer vision and behavior recognition, and provides an end-to-end learning framework from point cloud sequence input to action recognition output. The framework innovatively fuses two core processes of geometry-contour space structure modeling and multi-frequency time decomposition modeling: the geometry-contour space structure modeling accurately depicts macroscopic contour postures and microscopic geometry details of a human body to comprehensively capture motion information; the latter decomposes the time attitude sequence into a low-frequency trend flow and a high-frequency detail flow, and digs action mode rules under different frequencies; and through cooperative work of the two, the accuracy of point cloud action recognition and the robustness in a complex scene are remarkably improved.

The present invention discloses a method and an apparatus for rebuilding a relightable implicit human body model. A human body is represented as a deformable implicit neural representation, and a geometric shape, material attributes, and ambient lighting of the human body are decomposed to obtain a relightable and drivable implicit human body model. In addition, a volumetric lighting grid including a plurality of spherical Gaussian models is introduced to represent complex lighting with spatial variation, and visible probes capable of changing in position with change in human pose are introduced to record dynamic self-occlusion caused by human motion. With the method, drivable implicit models capable of being used for high-fidelity human body relighting may be generated in cases of sparse video input and even monocular input.

The invention relates to the technical field of human body kinematics parameter measurement, in particular to an adult joint motion range evaluation method and system based on a multi-view video, and the method comprises the steps: obtaining a multi-view video sequence of a testee, and generating an observation data set containing two-dimensional key point observation, a human body segmentation mask and camera parameter information; constructing an individualized joint geometric model containing a bone segment length, a joint center, a joint principal axis and a joint angle coordinate system based on the observation data set, and generating a three-dimensional symbol distance field; constructing a factor graph and performing incremental optimization solution, and outputting a joint angle sequence and an abnormal frame set; and determining an activity range interval and carrying out anti-fact consistency check, and if the validity is not satisfied, calculating an information matrix by a Jacobian matrix to generate a supplementary collection instruction to update a result, thereby improving robustness and verifiability.

The invention discloses a robot action training method and system based on a human motion video, and the method comprises the steps: firstly carrying out the preprocessing of the human motion video, and extracting and enhancing a video frame; then detecting two-dimensional coordinates of the motion key points, mapping the two-dimensional coordinates into three-dimensional space relative coordinates, and calculating a joint rotation angle; constructing and optimizing a motion time sequence, and fitting a reward function; and finally, based on reinforcement learning algorithm training, a target action model is obtained, and a system core depends on a deep learning network containing a ResNet backbone layer and the like and a DDPG / PPO reinforcement learning framework. Existing pain points are solved on the basis of a complete technical process, precise action simulation of the robot is achieved through rigorous steps, the scheme can be reproduced and expanded, and the robot action training requirements in the fields of industrial cooperation, service, rehabilitation and the like are met.

The invention provides an intelligent motion generation method and system of a human body, and electronic equipment, and the method comprises the steps: obtaining motion capture data based on video data, obtaining target motion information and an initial human body skeleton key point sequence based on the motion capture data, obtaining source skeleton motion data based on a human body structure prior key point sequence of the human body skeleton, and obtaining the target motion information and the initial human body skeleton key point sequence; obtaining a virtual human image based on the target motion information, obtaining a virtual human skeleton structure based on the virtual human image, obtaining redirected skeleton data, obtaining an optical flow graph and a mask graph based on the virtual human image and the redirected skeleton data, and obtaining an initial virtual human motion image sequence based on the optical flow graph and the mask graph; and acquiring a virtual human motion image sequence based on the initial virtual human motion image sequence, and acquiring motion output of the living agent based on the virtual human motion image sequence. Therefore, coherent and natural action generation is realized.

The invention discloses a multi-fingered dexterous hand operation reinforcement learning method based on a human action prediction model, and belongs to the field of humanoid robot dexterous hand tool body operation. A human action prediction model is trained by using data with three-dimensional pose marks of a human hand and an object, and is used for generalizedly generating a three-dimensional track of interaction between the human hand and the object. And a universal reward function is constructed by using the trajectory and is used for training a dexterous hand reinforcement learning strategy in simulation. Then, the migration ability of the strategy from simulation to reality is improved through simulation parameter domain randomization and course learning, and the reinforcement learning strategy obtained through training is deployed on a real world robot. According to the scheme, understanding of the future attitude of the operated object is introduced, the interaction process of the human hand and the object is predicted as a whole, the unified, simple and efficient reward function is constructed based on the prediction, the method is suitable for different operation tasks and different types of dexterous hands, and the method has the advantages of being high in generalization, high in success rate, high in reward function universality and the like.

The invention discloses a non-contact screening system for motion coordination based on computer vision, which relates to the field of computer vision, and comprises the following steps: acquiring a human body motion image through a non-contact camera, extracting human body key point coordinates frame by frame by using a deep learning algorithm after noise reduction and distortion correction, and constructing a time sequence data set; performing smooth jitter removal and interpolation processing to form standardized data; calculating track, rhythm and stability multi-dimensional motion coordination quantitative characteristics based on the data, and performing comparative analysis with a personal historical database to identify a motion function degradation trend; and based on age groups and scenes, customizing screening parameters, coordinating workflows of the components, and realizing full-process automatic screening. The method has the advantages that invasive and cross infection risks are avoided through non-contact acquisition, accurate and objective assessment of the exercise coordination is realized by means of deep learning and time series data analysis, exercise function degradation can be early warned, and an efficient and reliable scheme is provided for various exercise coordination screening scenes.

1. The name of the design product: physiological signal detector. 2. The use of the design product: for continuous acquisition of human motion, skin temperature, pulse, blood oxygen, electrocardiogram, electroencephalogram, ambient light, air pressure and other signals. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view.

The application discloses a 3D cartoon character motion video generation method oriented to a text sketch guide, takes a text prompt and a sketch as explicit input, takes noise input and a time step as implicit input, and correspondingly obtains a text vector and a motion feature; the noise input, the time step, the text vector and the motion feature are taken as input of a diffusion model to obtain noise of a next time step; after denoising by one step, a motion feature obtained from the sketch is used to calculate a key frame posture loss through a regressor, so that the key frame posture is more suitable for the sketch posture; a natural guide module is responsible for adjusting the natural degree of the key frame and adjacent frame postures to obtain denoised noise input; and after T-step continuous denoising, a final motion sequence is obtained. Through sketch feature extraction and sketch control information fusion branches, the application realizes the guiding function of the sketch on human motion sequence generation, and through the natural guide module, realizes the natural and smooth adjacent frame postures and motion diversity in the human motion sequence.

The invention belongs to the technical field of man-machine cooperation interactive motion monitoring, and provides a man-machine cooperation-oriented human trajectory prediction method and monitoring system, and the method comprises the steps: obtaining multi-view image data; based on the multi-view image data, reconstructing three-dimensional positions, linear velocities and angular velocities of a plurality of joint points of a human body and the tail ends of all connecting rods of the robot in real time, and generating a motion state sequence; inputting human body joint point three-dimensional coordinates of past T frames in the motion state sequence into a deep neural network model, and generating K multi-modal human body motion track prediction results of future S frames in a single step; calculating a potential risk index based on the multi-modal human body motion trajectory prediction result, the current human-machine relative pose and the motion state, and triggering a graded early warning signal according to a preset threshold value; and sending the graded early warning signal to a robot controller to execute deceleration, pause or avoidance actions, and synchronously displaying a man-machine digital twin, a predicted trajectory cloud and a risk grade on a three-dimensional visual interface.

The approaches presented here provide the use of reinforcement learning to fine-tune a generative model, such as a motion diffusion model, for a specific goal: to generate representations of human motion that correspond to the provided text input. A discriminator can be used to guide the training of the generative model. In at least one embodiment, the discriminator can compare the input text and the generated motion representation (or embeddings of both) to determine, for example, an alignment value or match score, which can then be used to adjust the network parameters or weights of the generative model to improve the alignment between the input text and the generated motion.

The application provides a human motion sequence prediction method and system based on skeleton-enhanced Transformer, comprising the following steps: S1: sampling human motion to obtain original motion sequence data; S2: generating virtual joints based on real joints, adding the virtual joints to the sequence, constructing enhanced motion sequence data, and generating an enhanced skeleton graph; S3: inputting the enhanced skeleton graph into a graph convolution network, establishing the skeletal correlation between virtual joints and real joints, and optimizing feature extraction through a spatiotemporal attention mechanism; S4: decomposing the input motion sequence data and dividing it into a trend part and a residual part, wherein the trend part is used to capture the trend information of the overall motion, and the residual part is used to retain local details; S5: combining the extracted features with the trend part and the residual part to generate future motion sequence data to obtain a prediction result; the skeleton features are enhanced by simulating joint combinations and introducing virtual joints.

The invention provides a field human body action recognition system based on sparse IMU and edge AI. The system comprises an acquisition module and an embedded action recognition module. The acquisition module is used for acquiring and processing human body motion data; after the system is started, human body motion data processed by the acquisition module activate an AI model deployed in the embedded motion recognition module for reasoning, and recognition tasks of more than 20 whole body motion categories are completed, the system is flexible, light and low in cost, end-to-end real-time and high-precision field motion recognition is successfully achieved, and the system is suitable for popularization and application. The situation awareness response speed and precision in the field complex environment are improved, powerful technical support can be provided for field digital application of human body action recognition, and the method has the advantages of being low in cost, easy to deploy, free of site limitation and the like.

This invention discloses a method, device, and medium for correcting human motion posture based on AI vision, relating to the field of AI vision technology. The method includes: acquiring multi-channel image sequences of human posture using multiple cameras; performing temporal alignment and normalization on the multi-channel image sequences; obtaining a human probability map using a pre-trained human semantic segmentation network; performing morphological cleanup to obtain a human silhouette sequence; performing back-projection according to camera intrinsic and extrinsic parameters to obtain a 3D joint sequence from a 3D AI visual volume; extracting the human motion manifold to perform contact-sensing neural motion priors and outputting prior signals; and using posture feasibility scores and joint importance weights to weightedly fuse the geometric difference vector and the center of gravity correction component to obtain a comprehensive joint correction vector. This invention generates a 3D AI visual volume through back-projection and extracts the central skeleton, using constant bone segment lengths for constraint and correction, thus enhancing the reliability of posture stability analysis and alignment with standard action templates.

This invention discloses a method, device, intelligent equipment, and storage medium for predicting human motion. The method includes: acquiring kinematic information of a target human body; performing a pre-defined simplified dynamic calculation on the kinematic information to obtain dynamic information of the target human body; inputting the kinematic and dynamic information into a neural network encoder to obtain kinematic spatiotemporal features and dynamic spatiotemporal features, respectively; and inputting the kinematic and dynamic spatiotemporal features into a neural network decoder to obtain a predicted human motion result for the target human body. Through the human motion prediction method of this invention, kinematic and dynamic information are complementary and coupled in their representation of human motion, comprehensively describing human motion from different perspectives, and enabling more accurate and longer-term predictions of human motion posture over a certain period of time in the future.

A method and system for evaluation of object motion repetitiveness as a basis to distinguish between human motion and non-human motion, in order to facilitate control of device operation. An example method includes (i) a computing system receiving sensor data representing object motion detected by at least one sensor, the object motion defining motion of an object, (ii) the computing system making a determination, based at least on an evaluation of the received sensor data, of whether the detected object motion is repetitive, and (iii) based at least on the determination being that the detected object motion is not repetitive, the computing system responding to the detected object motion by causing a device to take an action corresponding with a prediction that the object is a human being.

The ability to effectively communicate information to people without causing them any discomfort or unease. [Solution] The robot operation method includes a detection step in which a detection device detects human movement, and an operation step in which a plurality of robots perform actions in accordance with the movement detected by the detection device, wherein in the operation step, at least one of the plurality of robots performs an action that is deviated from the actions of the other robots.

The application discloses a visual-based human motion intention recognition interaction method and system, and relates to the technical field of computer vision. The method comprises the following steps: acquiring three-dimensional scene modeling data, detecting object spatial distribution to obtain target spatial granularity, and constructing an adaptive three-dimensional spatial anchor frame accordingly; collecting facial and binocular feature points, and predicting a visual gaze track through a binocular parallax algorithm; calibrating and optimizing the track by using the three-dimensional anchor frame, outputting a target, generating an instruction after analyzing the motion semantics of the target, and sending the instruction to a robot interaction terminal. The technical problem of low computational efficiency and ambiguous intention association caused by anchor frame redundancy in a complex physical scene is solved for the traditional visual-based intention recognition method, the technical effects of self-adaptive matching of a three-dimensional spatial anchor frame with a scene physical scale, introduction of a calibration and optimization mechanism for a visual predicted track, accurate association of human motion intention and a target object, and improvement of the accuracy, real-time performance and reliability of intention recognition are achieved.

The invention relates to a passive driving microcapsule movement color development method, and belongs to the field of sensor detection. According to the method, sweat generated during movement of a human body drives a photovoltaic device to spontaneously generate electric potential, the electric potential directly acts on microcapsule dispersion liquid to drive directional movement of black and white particles in the microcapsule dispersion liquid, and visual color development response without an external power source is achieved. The method mainly comprises the steps of sweat induced power generation, particle electrophoresis driving and ion concentration associated color development. The problems that an existing sweat detection technology depends on an external power source, the equipment size is large, and real-time visual reading cannot be achieved are solved, and the device is mainly applied to the fields of real-time sweat electrolyte analysis in exercise physiological monitoring, autonomous color development change of wearable medical equipment, passive sensing display integrated systems based on ion concentration gradient and the like.

System and methods of a mobile application to remotely monitor residents with device-free sensing of human motion over Wi-Fi is provided. A detailed behavioral analysis of various functions related to the sleep and activity of the subject has been provided. The Remote Care Application (RCA) is a mobile application on several platforms that displays the results of all the analysis and behavioral patterns obtained from the motion from the subject. The system relies on existing wireless communication signals and machine learning techniques in order to automatically detect the motion of the subject, sleep analysis of the subject, activity analysis of the subject.