107 results about "Pose tracking" patented technology

Dense range data obtained at real-time rates is employed to estimate the pose of an articulated figure. In one approach, the range data is used in combination with a model of connected patches. Each patch is the planar convex hull of two circles, and a recursive procedure is carried out to determine an estimate of pose which most closely correlates to the range data. In another aspect of the invention, the dense range data is used in conjunction with image intensity information to improve pose tracking performance. The range information is used to determine the shape of an object, rather than assume a generic model or estimate structure from motion. In this aspect of the invention, a depth constraint equation, which is a counterpart to the classic brightness change constraint equation, is employed. Both constraints are used to jointly solve for motion estimates.

The invention discloses a Kinect-based robot self-positioning method. The method includes the following steps that: the RGB image and depth image of an environment are acquired through the Kinect, and the relative motion of a robot is estimated through the information of visual fusion and a physical speedometer, and pose tracking can be realized according to the pose of the robot at a last time point; depth information is converted into three-dimensional point cloud, and a ground surface is extracted from the point cloud, and the height and pitch angle of the Kinect relative to the ground surface are automatically calibrated according to the ground surface, so that the three-dimensional point cloud can be projected to the ground surface, and therefore, two-dimensional point cloud similar to laser data can be obtained, and the two-dimensional point cloud is matched with pre-constructed environment raster map, and thus, accumulated errors in a robot tracking process can be corrected, and the pose of the robot can be estimated accurately. According to the Kinect-based robot self-positioning method of the invention, the Kinect is adopted to replace laser to perform positioning, and therefore, cost is low; image and depth information is fused, so that the method can have high precision; and the method is compatible with a laser map, and the mounting height and pose of the Kinect are not required to be calibrated in advance, and therefore, the method is convenient to use, and requirements for autonomous positioning and navigation of the robot can be satisfied.

Dense range data obtained at real-time rates is employed to estimate the pose of an articulated figure. In one approach, the range data is used in combination with a model of connected patches. Each patch is the planar convex hull of two circles, and a recursive procedure is carried out to determine an estimate of pose which most closely correlates to the range data. In another aspect of the invention, the dense range data is used in conjunction with image intensity information to improve pose tracking performance. The range information is used to determine the shape of an object, rather than assume a generic model or estimate structure from motion. In this aspect of the invention, a depth constraint equation, which is a counterpart to the classic brightness change constraint equation, is employed. Both constraints are used to jointly solve for motion estimates.

A measurement system and method for detecting and tracking the pose of an object displaceable in a coordinate reference frame. In the system and method a visible target pattern is provided on the object. The visible target pattern includes a series of contrast regions for providing at least two feature points. Each of these feature points is located at a juncture of an optically detectable plurality of edges. Each edge in the optically detectable plurality of edges separates different contrast regions. The method and system determine the location of the feature points by first locating the edges using the change in contrast between contrast regions, and then determining junctures of multiple edges. The method and system also involve a stereoscopic digital camera for generating a pair of digital images of the target pattern and a marker template comprising a set of reference characteristics including a relationship between the feature points.

The invention provides an unscented Kalman filter-based method for tracking an inertial pose according to acceleration compensation, which is used for an inertial measurement unit integrating a three-axis micro-gyroscope, a three-axis micro-accelerometer and a three-axis magnetoresistive sensor, and realizes pose tracking estimation on a device carrier by using rotary angular velocity vectors, acceleration vectors and magnetic field sensor vectors which are detected by the device by means of filter technology. The method comprises the following steps: 1) treating the acceleration vectors as combination of the acceleration vectors and gravity acceleration vectors of the device carrier self, and constructing observation equations respectively for amplitude and normalized direction vectors of the acceleration vectors and the gravity acceleration vectors; 2) describing quaternion, accumulated error vectors of the gyroscope and the acceleration vectors of the device carrier self by using the pose to construct a system state vector; and 3) realizing a filter estimating process of the system by using the unscented Kalman filter technology because of nonlinearity of the observation equations. Compared with the conventional method ignoring the acceleration of the carrier self, the method not only can provide a more accurate estimation result, but also widens the application range of the system.

The invention provides an optical measurement system and method for a three-dimensional shape of a large-scale complex curved surface member. The method is based on binocular optical grating projection measurement technologies, point cloud poses of all station positions during multiple station position measurement are obtained via a laser tracker and a corresponding target ball, point cloud obtained via the multiple station position measurement is converted to be under a unified laser tracker coordinate system according to corresponding pose data, and overall merging of point cloud data of a large-scale complex surface-shaped member can be realized; in the system, a six degree-of-freedom robot is used as a carrier for a point cloud space pose tracking unit and a binocular structured light measurement device, single station position measurement precision is ensured via calibration of the binocular structured light measurement device before measurement work starts, and data integrity and measurement efficiency are ensured via measurement route planning. Via the optical measurement system and method, all kinds of large-scale complex surface-shaped members can be accurately measured in non-contact conditions; practical, reliable and complete original three dimensional shape data can be provided for evaluation of all kinds of processing quality.

The invention relates to a robot positioning method with fusion of visual features and IMU information. The invention puts forward a method of fusion of monocular vision with IMU. Visual front-end pose tracking is performed and the post of a robot is estimated by using feature points; an IMU deviation model, an absolute scale and a gravity acceleration direction are estimated by using pure visualinformation; IMU solution is performed to obtain high-precision pose information and thus an initial reference is provided to optimize the search process, and the initial reference, a state quantity and visual navigation information are used for participating in optimization; and a rear end employs a sliding-window-based tightly coupled nonlinear optimization method to realize pose and map optimization. And the computational complexity is fixed while the speedometer is calculated based on a sliding window method, so that the robustness of the algorithm is enhanced.

A reference in an unknown environment is generated on the fly for positioning and tracking. The reference is produced in a top down process by capturing an image of a planar object with a predefined geometric shape, detecting edge pixels of the planar object, then detecting a plurality of line segments from the edge pixels. The plurality of line segments may then be used to detect the planar object in the image based on the predefined geometric shape. An initial pose of the camera with respect to the planar object is determined and tracked using the edges of the planar object.

The invention provides a SLAM method applied to a multi-lens combination panoramic camera, which well realizes the automatic positioning and map construction functions of a vehicle-mounted multi-lenscombination panoramic camera, and achieves a high positioning accuracy of 0.1 m level without resorting to an expensive GPS / IMU integrated navigation system. The method firstly performs camera calibration on a multi-lens combination device, and then establishes a parallel three-thread, wherein the pose tracking thread mainly completes system initialization, feature extraction on a panoramic image,matching and projection, pose solution optimization and selection of key frames; The sparse map construction thread is mainly responsible for establishing sparse map points according to the initial posture and matching points, optimizing the camera posture and map points locally, and eliminating the wrong map points and redundant key frames. The closed-loop correction thread is mainly responsiblefor detecting whether the motion of the camera has returned to the previous position (i.e. Closed-loop), adding closed-loop constraints to correct the posture and global posture and map point optimization (i.e. Global beam adjustment).

The invention discloses a camera pose tracking method and device, equipment and a system and belongs to the field of augmented reality. The method comprises the steps that second equipment receives aborn-image and born-pose parameters sent by first equipment; the second equipment acquires a current image collected by a camera; the second equipment acquires camera pose variation of the current image relative to the born-image; and the second equipment obtains current camera pose parameters generated when the camera collects the current image through calculation according to the born-pose parameters and the camera pose variation. Through the camera pose tracking method and device, the equipment and the system, the second equipment performs camera pose tracking on the image collected by itself relative to the born-image of the first equipment, it is guaranteed that camera pose tracking results of the two images are identical, and therefore the problem that a large amount of data needs tobe transmitted if it is needed to synchronize an environment map constructed on the first equipment to the second equipment is solved.

The invention discloses a method for three-dimensional human pose estimation, which can realize the real-time and high-precision 3D human pose estimation without high configuration hardware support and precise human body model. In this method for three-dimensional human pose estimation, including the following steps: (1) establishing a three-dimensional human body model matching the object, which is a cloud point human body model of visible spherical distribution constraint. (2) Matching and optimizing between human body model for human body pose tracking and depth point cloud. (3) Recovering for pose tracking error based on dynamic database retrieval.

The invention discloses a high-rise building external wall cleaning robot comprises a robot body, a cleaning mechanism, an external wall absorbing mechanism, a robot running and obstacle detouring mechanism, a robot supporting system and a GPS position and pose tracking controlling system, wherein the cleaning mechanism and the external wall absorbing mechanism are arranged on the robot body. The high-rise building external wall cleaning robot is characterized in that the robot running and obstacle detouring mechanism is a flying vehicle which is arranged on the upper part of the robot body and is connected with the robot body through a traction cable on the lower part, the lower part of the flying vehicle is provided with a propeller, and the robot supporting system is arranged in the flying vehicle. The invention has simple structure and low cost and is easy to operate and convenient to use and maintain.

In accordance with various embodiments of the invention, a user-borne computer input device is disclosed including a 5D or greater “mouse” or other tool or object containing or consisting of one or more permanent magnetic presenting permanent magnetic dipoles. The position and orientation of the device are determined by magnetic field strength measurements derived from at least two vector-magnetometers fixed in a reference frame connected to the computing device. The system allows a user to interact with the computing device in at least 5 dimensions by manipulating the position and orientation of the device within a measurement volume, which may be a few cubic meters.

A state estimation system that utilizes long-range stereo visual odometry that can degrade to a monocular system at high-altitude, and integrates GPS, Barometer and IMU measurements. The system has two main parts: An EKF that is loosely fused and a long-range visual odometry part. For visual odometry, the system takes the EKF information for robust camera pose tracking, and the visual odometry outputs will be the measurement for EKF state update.

Methods and systems for registering a manipulator assembly and independently positionable surgical table are provided herein. In one aspect, methods include reading a fiducial marker on the surgical table with a sensor associated with the manipulator assembly and localizing the manipulator assembly and surgical table with respect to a common reference frame. Methods may further include translating a 3D configuration of the surgical table to a 2D frame of reference so as to estimate a 3D pose of the surgical table relative the manipulator assembly for use in coordinating movements therebetween.

A method for obtaining a refined pose for a 3D sensor for online 3D modeling of a surface geometry of an object, the pose encompassing six degrees of freedom (DOF) including three translation parameters and three orientation parameters, the method comprising: providing the 3D sensor, the 3D sensor being adapted to capture 3D point measurements of the surface of the object from a viewpoint; providing a geometry model of at least part of the surface; observing a portion of the surface of the object with the 3D sensor; measuring an initialization pose for the 3D sensor by at least one of positioning device pose measurement, predicted pose tracking and target observation; finding a best fit arrangement of the 3D point measurements in the geometry model using the initialization pose; generating the refined pose for the 3D sensor using the best fit arrangement.

The invention relates to a monocular real-time three-dimensional reconstruction method based on loop testing, and belongs to the technical field of three-dimensional reconstruction. The method comprises: carrying out pairwise matching in an image sequence of a specified scene on the basis of an image feature point matching theory to obtain image matching point pairs; solving an essential matrix, and then utilizing a singular-value decomposition theory to acquire an initial pose; utilizing the initial pose or a previous-frame pose to obtain an estimated pose through a pose tracking model; judging whether a current frame is a key frame; then utilizing a random fern algorithm to calculate similarity of the current frame and the key frame, and if the similarity reaches a threshold value, it isconsidered that a loop is formed; utilizing a pose of the key frame to optimize the current pose if the loop is formed; utilizing the above-obtained pose to obtain a point cloud, and fusing the sameinto a TSDF global-model; and adopting a light ray projection algorithm to visualize a surface. According to the method, accuracy of the acquired pose is enabled to be high, the cumulative-error problem in three-dimensional-reconstruction processes is eliminated, and a real-time reconstruction result has higher accuracy.

A camera pose tracking apparatus may track a camera pose based on frames photographed using at least three cameras, may extract and track at least one first feature in multiple-frames, and may track a pose of each camera in each of the multiple-frames based on first features. When the first features are tracked in the multiple-frames, the camera pose tracking apparatus may track each camera pose in each of at least one single-frame based on at least one second feature of each of the at least one single-frame. Each of the at least one second feature may correspond to one of the at least one first feature, and each of the at least one single-frame may be a previous frame of an initial frame of which the number of tracked second features is less than a threshold, among frames consecutive to multiple-frames.

The invention relates to a real-time pose tracking method based on a target three-dimensional model. According to the method, based on a three-dimensional model of a target, contour features are utilized to establish a contour component model, and a contour is divided into a plurality of local contour sections according to the local contour recognition degree; For each contour segment, two dimensional-three dimensional correspondence is established in a gradient domain based on a phase matching criterion; a wrong matching pair is established by utilizing a geometrical relationship among all parts of the rigid body target; based on the established two dimensional-three dimensional matching pair, a pose parameter is solved through an RPNP algorithm, the whole process is iterated until convergence, and target pose tracking in the sequence image is realized. The algorithm is simple and clear in step and easy to apply, and drift-free tracking in the whole process can be achieved due to introduction of the three-dimensional model of the target.

A reference in an unknown environment is generated on the fly for positioning and tracking. The reference is produced in a top down process by capturing an image of a planar object with a predefined geometric shape, detecting edge pixels of the planar object, then detecting a plurality of line segments from the edge pixels. The plurality of line segments may then be used to detect the planar object in the image based on the predefined geometric shape. An initial pose of the camera with respect to the planar object is determined and tracked using the edges of the planar object.

The invention provides a pose tracking method and device. The pose tracking method comprises the steps that an image of a tracking object is acquired, and a mark flickering at a specific frequency isarranged on the tracking object; acquiring pixels of which the brightness changes from the acquired image; the six-degree-of-freedom posture of the tracking object is calculated based on the obtainedpixels, so that the dependence of posture tracking on the specific layout of LED marks is reduced, meanwhile, the delay of posture tracking is reduced, and the precision and efficiency of posture tracking are improved.

A system and a method for pose tracking, particularly for top-down, online, multi-person pose tracking. The system includes a computing device having a processor and a storage device storing computer executable code. The computer executable code, when executed at the processor, is configured to: provide a plurality of sequential frames of a video, the sequential frames comprising at least one keyframe and a plurality of non-keyframes; for each of the non-keyframes: receive a previous inference bounding box of an object inferred from a previous frame; estimate keypoints from the non-keyframe in an area defined by the previous inference bounding box to obtain estimated keypoints; determine object state based on the estimated keypoints, wherein the object state comprise a “tracked” state and a “lost” state; and when the object state is “tracked,” infer an inference bounding box based on the estimated keypoints to process a frame next to the non-keyframe.

The embodiment of the invention discloses a surgery navigation image display method and system based on augmented reality. When a mark point on a marker is blocked, a preset algorithm model is adoptedto process a gesture rotation matrix of the marker on a current moment, a three-dimensional space coordinate of the mark point on the marker and three-dimensional space displacement between the markpoint and an initially defined original point to obtain the three-dimensional space coordinate of the original point at the current moment; according to the position relationship between a target object and the initially defined original point and the three-dimensional space coordinate of the original point at the current moment, the position information of the target object at the current momentis obtained; finally, gesture information and position information are used as a pose tracking result of the target object, the pose tracking result is combined with a real scene to generate a surgerynavigation image. By use of the scheme, the position accuracy of a surgery tool in the real scene in the surgery navigation image can be improved.

This application discloses a repositioning method and apparatus in a camera pose tracking process, a device, and a storage medium, belonging to the field of augmented reality (AR). The method includes: obtaining a current image acquired after an ith anchor image in a plurality of anchor images; obtaining an initial feature point and an initial pose parameter in the first anchor image in the plurality of anchor images in a case that the current image satisfies a repositioning condition; performing feature point tracking on the current image relative to the first anchor image, to obtain a plurality of matching feature point pairs; filtering the plurality of matching feature point pairs according to a constraint condition, to obtain a filtered matching feature point pair; calculating a pose change amount of a camera from the initial pose parameter to a target pose parameter according to the filtered matching feature point pair; and performing repositioning according to the initial pose parameter and the pose change amount to obtain the target pose parameter of the camera.

Embodiments of this application disclose a method for determining camera pose information of a camera of a mobile terminal. The method includes: obtaining a first image, a second image, and a template image; performing feature point detection on a first feature point of the template image and a second feature point of the second image, to obtain a first homography; determining a first target homography according to a first optical flow feature point in the first image and a second optical flow feature point in the second image, and determining a second homography according to the first target homography and a second target homography; and performing complementary filtering processing on the first homography and the second homography, to obtain camera pose information of the camera. In the embodiments of this application, complementary filtering processing may be performed on two homographies obtained in a camera pose tracking process.

The invention discloses an integrated control method for a tethered robot approaching target pose. The method comprises the steps of: 1) establishing a tethered robot pose dynamic model; 2) designingan interference observer; 3) designing a controller; and 4) distributing control force / torque. According to the integrated control method for the tethered robot approaching target pose, the interference observer is designed to estimate un-modeled deviation and environmental interference, and the integrated control method is designed for target pose tracking. In addition, the pose-orbit couplingcharacteristic of the tethered robot is fully considered, so that the pose of the tethered robot can be integrally controlled by a tether in a target approaching process.

The invention provides an information detection system for rotation center movement of a glenohumeral joint of a human shoulder and belongs to the technical field of detection. The detection system comprises a human sternum position tracking sensor, a human humerus position tracking sensor, a human scapula pose tracking sensor and a data acquisition card, wherein the human sternum position tracking sensor is used for detecting the position of human sternum and provided with four point positions for acquiring human pose information in real time; the human humerus position tracking sensor is used for detecting the position of human humerus; the human scapula pose tracking sensor is used for detecting the pose of human scapula; the data acquisition card is used for collecting the positions ofthe human sternum and the human humerus and the pose of the human scapula; the rotation center movement information of the glenohumeral joint of the shoulder in one cycle is displayed by processing based on upper computer application and data processing subprogram. According to the acquisition system, the problems of data mark point loss of a VICON detection system and blocking of a human camerasystem are solved; the acquisition system is simple, high in adaptive capacity, accurate in data acquisition and good in real-time performance, and can be applied to shoulder function bionics, shoulder function simulation, ergonomics and the like.

This application discloses a repositioning method performed by an electronic device in a camera pose tracking process, belonging to the field of augmented reality (AR). The method includes: obtaining a current image acquired by the camera after an ith anchor image in a plurality of anchor images; selecting a target keyframe from a keyframe database according to Hash index information in a case that the current image satisfies a repositioning condition; performing second repositioning on the current image relative to the target keyframe; and calculating a camera pose parameter of a camera during acquisition of the current image according to a positioning result of the first repositioning and a positioning result of the second repositioning. In a case that there are different keyframes covering a surrounding area of a camera acquisition scene, it is highly probable that repositioning can succeed, thereby improving the success probability of a repositioning process.

The invention provides a double-rigid-body characteristic point saturation fixed time relative pose tracking control method. The method comprises the steps: establishing a pose kinetic equation of a tracker and a target device based on a Newton-Euler method; establishing a relative pose kinetic equation considering model uncertainty according to the established respective pose kinetic equations ofthe two rigid bodies and the relative pose and the relative speed between the two rigid body feature points; designing a fixed time disturbance observer according to the established relative pose kinetic equation; designing a non-linear saturation compensator according to the established fixed-time non-singular terminal sliding variable so as to solve the saturation effect of an actuator; and combining the established fixed-time disturbance observer and the nonlinear saturation compensator to obtain a model-based robust adaptive controller. The invention relates to the field of control of anautonomous moving body in a three-dimensional space and spacecrafts.