448 results about "Articulation points" patented technology

The invention discloses a human body motion recognition method based on three-dimensional bone information. The method comprises the steps that color data stream, depth data stream and bone data stream of a number of individual samples of different sex and height are processed, wherein the individual samples make various motions; SVM models corresponding to all motions are constructed; the bone data stream of a recognized target which makes any motion within the acquisition range of a Kinect depth sensor is acquired; through the bone data stream, the distance between a normalized human body bone articulation point and a reference point and 14 vector angles are acquired; the data are respectively input to multiple SVM models; and a motion corresponding to the SVM model with the highest output probability is the motion of a recognition target. According to the invention, the Kinect depth sensor is used to acquire images; the method is less affected by light condition, shadow and other factors; the depth map and the bone information of a human body motion are acquired in real time; and a human body target in a scene can be accurately located.

The invention relates to a fall detection method based on a video. The method comprises steps that S1, video images are processed, and human body areas of the images are identified and positioned; S2, articulation point extraction based on the cascade regression network is carried out for the human body areas to acquire a set of human body articulation points, the first stage network is cascaded back with multiple regression networks having identical structures so as to precisely adjusting coordinate positions of the human body articulation points; and S3, a motion vector of each human body articulation point is taken as a human body motion characteristic, through analyzing articulation point change, whether a human body falls is dynamically analyzed. The method is advantaged in that whether the elderly people living alone falls can be accurately and conveniently determined through video image analysis.

Shoe (1), especially an athletic shoe, with a midsole (2) and an upper (3) connected to the midsole (2), the midsole (2) having a base body (2a) which is connected to the upper (3), and a bottom plate (2b). To improve the adjustability of the spring properties and / or the damping properties of the shoe sole, it is provided that at least one part (4, 5) of the bottom plate (2b) is connected to the base body (2a) by a hinge at an articulation point (7). Furthermore, at least one spring element (8) is provided between the side of the part (4, 5) of the bottom plate (2b) facing the base body (2a) and the side of the base body (2a) facing the part (4, 5) of the bottom plate (2b). Preferably, one spring is provided in the heel area and a plurality of springs are provided in the forefoot area.

The invention discloses a pedestrian falling recognition method based on skeleton detection. The method comprises the following steps: S1, acquiring a monitoring area image by using a camera; S2, segmenting the image to obtain a pedestrian body area image, and detecting to obtain pedestrian skeleton feature point distribution information; S3, analyzing pedestrian skeleton feature point distribution information, extracting articulation point coordinates of key parts of the human body of the pedestrian, acquiring pedestrian human body articulation point space position features and pedestrian posture geometrical quantities according to the articulation point coordinates, and the key parts are preset parts on the human body; S4, establishing a tumble detection model according to the spatial position characteristics of the joint points of the human body of the pedestrian and the pedestrian posture geometric quantity; and S5, judging the posture of the pedestrian by using the fall detectionmodel wherein the posture of the pedestrian comprises normal walking and fall, and the detection and recognition of the posture of the pedestrian are realized according to the judgment result. The method can actively detect abnormal conditions such as pedestrian falling in a monitoring video, and can improve the monitoring capability of pedestrian safety events by combining with an early warning system.

The invention discloses a hand posture estimation method based on depth information and a calibration method. The method comprises the steps that at the step S1, hand depth data is acquired, and a hand area is partitioned from the hand depth data; at the step S2, a palm posture is detected according to the hand area; at the step S3, positions of each articulation point of the hand are computed in combination with the palm posture and a hand standard bone model; at the step S4, projection characteristics of each articulation point of the hand are computed; and at the step S5, finger posture calibration is carried out according to the projection characteristics of each articulation point of the hand. The method disclosed by the invention is characterized in that the depth data is taken as the basis; the hand area is partitioned; the palm posture is computed; and the finger posture is estimated according to depth images through the posture calibration. The method is simple and practical.

A method, information processing system, and computer program storage product manage connections between a virtual world and a social network. A set of virtual world information and a set of social network information are analyzed. A graph including a plurality of vertices is generated. Each vertex represents one of virtual world information and social network information. Each vertex is coupled to at least one other vertex by a respective edge. At least one edge of the graph couples a first vertex representing virtual world information and a second vertex representing social network information. At least one vertex is determined to be an articulation point having a respective edge. The removal of the respective edge of the articulation point causes a disconnection of the virtual world information from the social network information within the graph. A user is notified via a graphical user interface that the graph comprises the articulation point.

The invention discloses a fall-down behavior real-time detection method based on a depth image. The method comprises steps of depth image obtainment, human body image identification, depth difference feature extraction, human body part analysis, articulation point extraction, height feature extraction and fall-down behavior detection. Based on a depth image, a specific depth difference feature is selected on an identified human body image; a random forest classifier is used for analyzing human body parts; a human body is divided into a head part and a trunk part; articulation points are detected, and then a height feature vector is extracted; and a support vector machine classifier is used for detecting whether a detected object is in fall-down state. The invention provides the fall-down behavior detection method, improves the operation speed, and achieves timeliness of fall-down behavior detection. The depth image is utilized for fall-down behavior detection. On one hand, the method is free of influences of illumination and can be operated in an all-weather manner, and on the other hand, personal privacy can be protected compared with a colorful image. Only one depth sensor is required in terms of hardware support, and the advantage of low cost is achieved.

The invention relates to a no-mark human body motion capture method; the method uses human surface three-dimensional voxel of human skeleton sleeve pattern tracing reconstruction, realizes right classification of voxel robustly and extracts articulation points automatically to get human motion parameter. The method comprises the following steps: (1) extracting three-dimensional voxel of human surface; (2) initializing human skeleton pattern and matching with a first frame voxel data; (3) using overall optimization algorithm to realize skeleton pattern tracing voxel data; (4) classifying the voxel based on overall classification distribution histogram of each voxel in tracing process; (5) extracting three-dimensional articulation points from classified voxels; and (6) getting articulation angle based on the coordinate of the articulation points and then getting human motion parameter. The invention has the advantages of easy implementation, relatively low cost, strong robustness and wide application and the like.

The invention provides a real-time human skeleton articulation point detection method and belongs to the field of image processing. The invention designs a hierarchical context attitude detection network, digital image processing and deep learning are adopted for realizing real-time detection on coordinates of skeleton articulation points of human bodies under the conditions of different attitudes, different illuminations and different scale dimensions, a real-time detection method is realized while precision is guaranteed, calculation amount of a model can be obviously reduced, and model training, deployment and usage efficiency is guaranteed.

The invention relates to an embedded platform real-time tumble detection method based on an improved attitude estimation algorithm. According to the method, an attitude estimation network is established by using depth separable convolution, an attention mechanism and an inverse residual error structure and is used for fall detection. The precision of the attitude estimation network is further improved; the parameter quantity and the calculation quantity are greatly reduced; the distance of each articulation point of the human body between different video frames is calculated to track the humanbody; acceleration of human body articulation points is calculated by using front and rear video frames, and whether falling occurs is judged according to the acceleration, the relative positions ofthe articulation points and the like, so that the attitude estimation network is more suitable for being deployed on an embedded platform, and a real-time effect can be achieved by deploying on a TX2embedded platform. According to the method, human body tracking is carried out by using multi-person human body joint point coordinates and skeleton information obtained in front and back frames, posture estimation is more stable due to multi-person tracking, and the problem of fall detection in a multi-person scene can be better solved.

The present invention discloses a human body behavior prediction method based on human body skeleton movement information, comprising the following steps: calculating a normalized relative orientative feature of each articulation point of each limb by using human body skeleton information extracted from an RGB D image; carrying out dynamic segmentation on a feature sequence by using a segmentation method based on feature sequence potential difference so as to acquire a posture feature sub-sequence and an action feature sub-sequence; extracting a key posture and an atomic action from the posture feature sub-sequence and the action feature sub-sequence, and constructing a multi-layer graph model based on the key posture and the atomic action; extracting a human body sub-behavior pattern contained in the multi-layer graph model, and constructing a context probability statistical model of the human body sub-behavior pattern; and carrying out identification and prediction of the human body sub-behavior pattern. The human body behavior prediction method based on the human body skeleton movement information has strong robustness for body differences, spatial position differences and the like between different individuals, has strong generalization capability for action differences between different individuals in the same category of behaviors, and has strong identification capability for the action similarity between different categories of behaviors.

The invention discloses a method for creating a three-dimensional surface model by using a perspective sketch. The method comprises the following steps of: identifying and processing a hand-drawn sketch into a vertex-edge graph consisting of contour sides and articulation points, and identifying closed areas; identifying the symmetric relation of the three-dimensional surface model, and completing hidden contours for the contours according to symmetry and visibility; establishing an object coordinate system of the surface model according to a symmetrical relation, calculating three-dimensional coordinates of the contour sides and the articulation points of the surface model, and recovering the coordinates of three-dimensional space curves among the articulation points according to curve projection between the three-dimensional coordinates of the articulation points and the articulation points; categorizing the three-dimensional closed areas of the model into different closed area pieces according to the characteristics of a level modeling method and a B-spline surface fitting method; and modeling the closed area pieces into sub-surfaces according to the characteristics of the closed area pieces by the level modeling method or the B-spline surface fitting method and jointing the sub-surfaces into a three-dimensional model through dispersing, blending and splicing.

The invention relates to an in particular capless neck end for a container, in particular a fuel container, with a double-flap closure which is formed by an upper pivoting flap (12) and a lower sealing flap (2). Neck ends of this type are known from WO 2006 / 084908. These known neck ends require too much installation space and are not optimally protected from the contents of the container being pumped out. This is improved by the invention in that the lower sealing flap is arranged in such a manner that it is arranged at an angle to the transverse section plane arranged at right angles to the neck axis (S) when in the closed state, and either the articulation point or, when in the closed state, the end opposite the articulation point of the lower sealing flap is arranged above the lower end of the first circular segment path. Furthermore, an anti-theft means (25) in the form of a mechanical barrier is provided below the double-flap closure.

The invention relates to an action evaluation method based on human body three-dimensional articulation point detection, which belongs to the field of computer vision and comprises the following steps: S1, performing human body three-dimensional articulation point detection on a single-frame picture after video framing; S2, extracting key frames of a specified number of frames of the video; S3, constructing motion vector features and joint kinetic energy features, and extracting feature values; and S4, constructing a key frame action similarity comparison model through multi-feature fusion: fusing the sub-features in the step S3, and constructing a personalized model for different types of actions; constructing a motion vector feature similarity function based on cosine similarity, and constructing a joint kinetic energy similarity function based on a weighting function; and obtaining a key frame action similarity comparison model based on the two similarity functions, comparing the action to be detected with the key frame set of the standard action, and finally obtaining the action similarity of the motion video. The method is more accurate and scientific, and can be used for physical fitness action correction and teaching.

An embodiment of the invention provides a method and device used for pedestrian detection and gesture estimation. The method includes acquiring a to-be-processed image; inputting the to-be-processed image into a first full convolution network so as to acquire articulation point feature maps relevant to each type of pedestrian articulation points in a first preset number of types of pedestrian articulation points and pedestrian feature maps relevant to each pedestrian in at least one part of pedestrians in a to-be-processed image respectively; forming a combined feature map through combining all the acquired pedestal feature maps and the all the articulation point feature maps together so as to acquire the pedestrian position and the articulation point position of each pedestrian among theat least one part of the pedestrians. According to the embodiment, intermediate results of pedestrian detection can assist pedestrian gesture estimation and intermediate results of pedestrian gestureestimation can also be used for pedestrian detection, so that the accuracy of pedestrian detection and pedestrian gesture estimation can be improved.

The invention discloses a three-dimensional human body motion tracking method based on quantum immune clone algorithm. The tracking process comprises: detecting articulation points of the human body in a two dimension picture in a unary picture sequence with human body gesture existing, using a kalman filter to predict the articulation points not detected, and simultaneously establishing a three-dimensional human body model; applying quantum immune clone algorithm in human body tracking, initializing species group, setting initial parameters of human body motion, then carrying out clone operation, performing operation of immunizing genes and immunoselection to generate a new species group; substituting the state parameters of the new species group in the three-dimensional human body model, using likelihood function to calculate the distance affinity, retaining the optimal solution, performing multiple times of substitution and calculation to obtain the ideal human body motion parameters to restore the three-dimensional human body gesture. The invention has the advantages of lower calculation cost and capable of obtaining three-dimensional human body gesture quickly, and is applicable to tracking of human body motion gesture in scene monitoring, motion analysis and health assessment.

The invention discloses a multi-model human motion tracking method, and relates to the technical field of image processing, which mainly solves the problems that human motion ambiguity and time complexity are high and excellent three-dimensional body posture estimation cannot be obtained by increasing motion models purely in the conventional method. The multi-model human motion tracking method comprises the following steps of: (1) inputting a human motion video image to acquire a human silhouette and an edge and a skeleton line thereof; (2) detecting the positions of human articulation points; (3) training the motion models by a ridge regression method; (4) initializing a model set M1; (5) performing an interactive multi-model algorithm to acquire a human motion posture; and (6) activating the motion models which meet the activating condition, and terminating the motion models which meet the termination condition. The multi-model human motion tracking method has the advantages of low time complexity and good tracking effect, has high cost effectiveness, and can be applied in fields of sports training, animation production and the like.

The invention relates to a video-based method for tracking human body limb movement which only tracks the limb movement of a single walking person; the method has the following operation steps: (1) initialization: a human body image is divided into 10 parts and 15 articulation points are used to describe the positions of the 10 parts in the image, so that the positions of the 10 articulation points in each image are confirmed to realize tracking of human body limb movement; (2) prediction: a numerical value of a posture vector RA responding to the current image is predicted only by predicting the numerical value of the posture vector RA of the previous frame, so that tacking of human body limb movement can be realized. In the method of the invention, a camera is only needed; in addition, two cases of edge sampling and internal sampling are both taken into account when weight of particles is calculated, so that the method has the advantages of less restrictions, no need of camera calibration, simple application, accurate pose recovery and the like.

The invention discloses an end-to-end based human body articulation point detection and classification method. The realization method thereof comprises the steps of constructing a deep learning network and training and testing input picture data. The deep learning network includes a basic network layer structure, a pyramid pooling residual module and a stacked hourglass-type module. In a trainingphase: pre-processing operation of an input image and initialization of network parameters are carried out; the processed image is input into the deep learning network based on the stacked hourglass-type module and the conditional random field theory to undergo training; and network weight parameters are updated according to a Softmax loss function. In a test phase: forward-propagation of the testimage is performed by using the deep learning network model parameters obtained by the learning to obtain the probability distribution of the test articulation points, and the articulation points aresequentially connected by using known criteria to obtain the result image.

The present invention discloses a three dimensional human pose recognition method and apparatus, where the three dimensional human pose recognition method includes steps of: a three dimensional pose initial recognition step of performing three dimensional pose recognition on an input image containing a human image to obtain image-based three dimensional human pose information; a sensor information acquisition step of acquiring, by a motion sensor, motion information of human articulation points; and a three dimensional pose correction step of correcting, with the motion information acquired by the sensor information acquisition step, the image-based three dimensional human pose information recognized by the three dimensional pose initial recognition step. According to the technical solution of the invention, it is possible to improve accuracy of the three dimensional human pose recognition efficiently. Further, the present invention also discloses a three dimensional human pose recognition method and apparatus, in which three dimensional half-body pose recognition is proposed, thereby speed of three dimensional human pose recognition can be improved significantly while improving accuracy of three dimensional human pose recognition.

The invention provides a human body movement identification method based on a Kinect. The human body movement identification method includes the steps of obtaining spatial position information of a skeleton articulation point of a target human body through the Kinect, and then identifying a movement type of the target human body by judging whether the spatial position information meets a preset judging standard of various human body movements or not. By means of the human body movement identification method, the advantages of the Kinect are effectively used, the spatial position information of the skeleton articulation point of the human body can be obtained in real time, other sensing devices are not required to be used in an assisting mode, picture processing is not carried out, the systematic complexity is reduced, the human movement identification speed and accuracy is improved, and the man-machine interaction effect is improved.

The invention discloses a human body three-dimensional reconstruction method and device based on RGBD single-view images. The human body three-dimensional reconstruction method comprises the steps: collecting RGBD images of a human body through a depth camera, wherein the images comprise a single-view-angle color image and a depth image; respectively obtaining three-dimensional human body model parameters, human body segmentation information and two-dimensional joint point information according to the RGBD image; obtaining human body three-dimensional articulation point information according to the human body segmentation information, the two-dimensional articulation point information and the depth picture so as to restrain articulation points and body shapes of the three-dimensional humanbody model according to the human body three-dimensional articulation point information, and optimizing three-dimensional human body model parameters and the three-dimensional human body model; and rendering a depth picture according to the optimized three-dimensional human body model, and further optimizing the front surface of the optimized three-dimensional human body model by utilizing the single-view color picture to obtain a three-dimensional reconstruction result of the human body. According to the human body three-dimensional reconstruction method, three-dimensional reconstruction ofa human body can be carried out by utilizing single-frame single-view RGBD image information acquired by the depth camera.

A hearing protection device is provided including a stem, a protrusion formed on the stem, an articulation point formed on the stem, and a sound attenuating element including a cavity, where the protrusion is disposed in locking engagement within the cavity to releasably attach the stem to the sound attenuating element, and where the stem is configured to at least partially articulate about the articulation point.

A method for processing an inputted XPath query against an inputted XML document is provided. The method generates a summary index, an ancestor-descendant path index, and a value index from one or more inputted XML documents. The summary index and the ancestor-descendant path index and the value index have at least one path defined. XPath queries at articulation points are parsed into multiple partial queries. The cursor type index access methods are determined. Partial queries are executed against the SUM-Index to generate a list of path identifiers (PID) satisfies the partial query segments. A set of ancestor-descendant PID identifiers list is generated to provide a result sequence. The result sequence of nodes is filtered producing one or more outputted XML documents from the final result sequence of nodes. A related system and computer medium is also provided.

The invention discloses an attention scenario evaluation system based on Kinect interaction. The system comprises the implementation steps of (1) utilizing the multimedia analogue simulation technology for presenting a scenario similar to real life, introducing a lifelization interaction scenario into a test system, (2) utilizing a Kinect camera for conducting dynamic capture on a user, converting user information into a skeleton tracking system through image recognition, drawing a human skeleton which is placed into the scenario in the first step, (3) tracking motions of the user through changes of positions of articulation points of the human skeleton, adding event monitoring at the same time, constantly drawing new skeletons in the scenario, (4) achieving communication between hardware and designed scenario application programs through the AIR Kinect middleware, and (5) completing items in the Kinect interaction mode by the user, wherein the system evaluates the situation that the user completes the items by comparing the motions with standard motions. According to the attention scenario evaluation system based on the Kinect interaction, testing is completed in the Kinect interaction mode, attention of the detected user under the real-world situation can be reflected, and assistance by professionals is not required.

The invention discloses a rehabilitation training and evaluating method based on Kinect for upper limbs. The method comprises a method for testing upper limbs based on Kinect and an exercise rehabilitation training module and evaluation module for upper limbs. According to the method for testing upper limbs based on Kinect, the spatial positions of shoulders, elbows, wrist joints of upper limbs of rehabilitation physician and patients are extracted in real time through the Kinet; the motion angles of three degrees of freedom of the shoulders and the elbows in relative to reference articulation points; for the upper limb exercise rehabilitation training module and the evaluation module, a set of simulation rehabilitation environment of cleaning tables and cleaning glass is designed at first; secondly, a set of standard motions for completing virtual environment tasks is designed by a rehabilitation physician and a standard exercise angle data module is calculated and generated according to the testing method of upper limbs based on the Kinect; finally, a real exercise data module for the patient to complete the simulation rehabilitation tasks is obtained through the Kinect method for testing upper limbs; by matching the real exercise data module with standard exercise data, the rehabilitation physician evaluates real exercise quality of the patient and provides reasonable rehabilitation feedback for the patient.

The invention discloses a hand positioning method and device in a three-dimensional space, and intelligent equipment, and is used for lowering a calculated amount for hand positioning in the three-dimensional space, improving hand positioning robustness and reducing environment interference for the hand positioning. The hand positioning method in the three-dimensional space comprises the following steps that: collecting a depth image which contains a hand; intercepting a target depth image which only contains the hand from the depth image, and on the basis of a pre-trained hand articulation point model and the target depth image, determining the three-dimensional space coordinate, which is relative to a camera used for collecting the depth image, of each articulation point in the target depth image; and according to the pre-determined pose data of the camera in the space and the three-dimensional space coordinate, which is relative to a camera used for collecting the depth image, of each articulation point in the target depth image, determining the three-dimensional space coordinate of each articulation point of the hand in the space in the target depth image.

In order to improve a surgical holding device for holding a surgical instrument, with a frame, with a first linkage comprising at least two links, which articulatedly connects the frame to a first articulation point associated with the instrument, and with a second linkage comprising at least two links, which articulatedly connects the frame to a second articulation point associated with the instrument, so that a large range of action of the surgical instrument held by the holding device is adjustable as simply as possible, it is proposed that the links of the first linkage be mounted for pivotal movement about pivot axes relative to one another and relative to the frame, and that the pivot axes extend parallel to one another.

The invention discloses a shooting training system and method based on a smart phone and artificial intelligence. The system comprises a court positioning system, a basket detection system, a player identity recognition system, a player action analysis system, a basketball track recording system, a goal detection system and a visual playback system. During training, the court positioning system detects positions of a court boundary line, a center line, a penalty line, and side lines of a limiting area and a penalty area; the basket detection system performs basket position detection, and performs tracking and real-time detection based on target tracking; the player identity recognition system confirms the position and identity of the player; the player action analysis system acquires player body articulation point position information; the basketball track recording system records the ball position and the target size; the goal detection system judges whether a ball enters the basket or not; and the visual playback system stores and plays back the obtained information and the frames of the video. According to the basketball shooting training system, shooting training of basketballtrainees is facilitated, and the defects of an existing shooting training system are overcome.