51 results about "Camera tracking" patented technology

The invention discloses a multi-camera collaborative pedestrian and vehicle cross-view tracking method and system, and relates to the technical field of intelligent monitoring. The method comprises the following steps: carrying out video stream acquisition on a monitoring area to obtain a multi-view video data set; multi-target detection of pedestrians and vehicles is carried out, tracking is carried out according to a plurality of detection targets, and a local tracking trajectory set of a plurality of cameras is generated; carrying out collaborative analysis on the plurality of cameras, constructing a multi-camera collaborative strategy to carry out cross-view trajectory matching on the local tracking trajectory set, and generating a global tracking trajectory set; and updating target states of the plurality of detection targets according to the global tracking trajectory set, and generating a multi-target cross-view tracking result of the pedestrian and the vehicle. The technical problem that in the prior art, information of different visual angles is difficult to integrate effectively in a multi-camera scene, so that target loss and track discontinuity easily occur in cross-camera tracking of pedestrians and vehicles is solved, and the technical effect of improving the continuity and accuracy of cross-camera tracking is achieved.

Methods provide computer-navigation assisted total hip arthroplasty (THA) procedures. Such procedures include the use of a navigated instrument, e.g., a navigated reamer construct or a navigated inserter construct to prepare the acetabulum and insert an acetabular shell, a camera tracking system adapted to track a pose of the navigated inserter or reamer construct relative to the patient in use, and a computer platform adapted to receive pose tracking information and generate and display navigational guidance for the user.

This application provides a camera tracking control system and intelligent care device, belonging to the field of intelligent elderly care technology. The system includes a motion drive module, a posture sensing module, a face recognition module, and a controller communicating with these three modules. The posture sensing module outputs corresponding drive signals based on the left and right displacement of the human body. The face recognition module outputs a stop signal after recognizing a face. The controller drives the camera to adjust its position according to the human body's movement, locking and recording upon face recognition. The camera tracking control system and intelligent care device provided by this invention can achieve fully automatic tracking and recording of the elderly, solving the problem of blind spots that are common with fixed cameras, effectively improving the reliability of home-based elderly care, and reducing the cost of manual care.

The present invention provides a system for monitoring and improving toothbrushing effectiveness. The system comprises a camera device comprising a camera or similar visual sensor configured to capture real-time footage of a user brushing their teeth an AI-powered software module is designed to analyse the footage using object recognition, pose estimation, and motion tracking algorithms to detect the user's hand, toothbrush, and face, a feedback device that comprises a feedback interface that provides real-time visual, auditory, or haptic feedback based on brushing technique, including parameters such as angle, coverage, and movement speed and movement patterns. Wherein the feedback provided comprises instructions to the user to alter their toothbrushing technique to bring the parameters closer to a predetermined set of ideal values. Thereby providing a system operates independently of sensors, accelerometers, or physical markers embedded in or attached to the toothbrush.

The invention discloses a semantic 3D Gaussian sputtering SLAM (Simultaneous Localization and Mapping) method based on illumination invariance. The method comprises the following steps: step 1, performing 3D Gaussian-based scene representation and joint micro-rendering on a to-be-constructed target area to obtain a multi-dimensional rendered initial scene model; 2, an internal appearance normalization IAN module is constructed, and decoupling of internal attributes of an initial scene model and transient illumination is achieved; 3, performing offset correction on a radiation field of a rendering graph in the initial scene model, and designing DRB-Loss loss to dynamically adjust correction intensity; step 4, through camera tracking optimization, combining a stable map and exposure compensation to realize robust pose estimation, and through double-stage key frame screening, improving SLAM system efficiency; step 5, constructing joint map loss, and optimizing all parameters of the 3D Gaussian primitive; and map updating is completed, invalid 3D Gaussian primitives are pruned, and a complete synchronous positioning and map building cycle is formed.

Paving machines and paving machine control systems utilize sensors to identify the location and orientation of paving machines and corresponding attachments. The control system implements a design profile by automatically adjusting actuators. The control system receives transition values to calculate a smooth transition including continuously adjusting actuator values with reference to the location and orientation of the paving machine. The control system may identify resurfacing values to repair a damaged surface by identifying an offset value of an existing surface to an ideal surface with reference to a disposition of the paving machine to minimize material usage.

The application relates to the technical field of computer vision, and discloses a vehicle cross-camera tracking method and system based on a twin network, which comprises the following steps: collecting a track segment of a target vehicle under a camera; performing Kalman filtering on the track segment to obtain a motion state sequence, performing time series interpolation on the motion state sequence to obtain a speed fluctuation curve; constructing a directed connected graph based on the layout characteristics of the camera, mapping the speed fluctuation curve to the directed connected graph to obtain a topological behavior correlation graph of the target vehicle; jointly encoding the speed fluctuation curve and the topological behavior correlation graph to obtain a behavior feature vector of the target vehicle, and inputting the behavior feature vector into a twin network to obtain a behavior consistency score of the target vehicle; and associating the identity of the target vehicle based on the behavior consistency score, and determining that the target vehicle is the same target vehicle if the behavior consistency score is greater than a preset association threshold; and the application can improve the accuracy and robustness of vehicle cross-camera tracking.

A surgical system for computer assisted navigation during surgery, includes at least one processor that obtains a 3D radiological representation of a targeted anatomical structure of a patient and a set of fiducials of a registration fixture. The operations attempt to register locations of the set of fiducials in the 3D radiological representation to a 3D imaging space tracked by a camera tracking system. Based on determining one of the fiducials of the set has a location that was not successfully registered to the 3D imaging space, the operations display at least one view of the 3D radiological representation with a graphical overlay indicating the fiducial has not been successfully registered to the 3D imaging space, receive user-supplied location information identifying where the fiducial is located in the 3D radiological representation, and register the location of the fiducial to the 3D imaging space based on the user-supplied location information.

The invention provides a visual SLAM (Simultaneous Localization and Mapping) algorithm based on multi-scale neural tensor representation, which belongs to the technical field of indoor scene real-time reconstruction, and can efficiently capture the global structure and fine details of a scene by utilizing a tensor decomposition technology, and reduce the memory usage amount and the calculation overhead at the same time. In addition, the invention further introduces an efficient pixel-based color query method, and each pixel only needs neural network forward regression once in the method, so that the real-time performance of the neural radiation field-based SLAM (Simultaneous Localization and Mapping) is further improved. Experiments on a synthetic data set and a real data set show that the method is superior to the most advanced method in the aspects of reconstruction quality and camera tracking precision, and the efficiency in the aspects of running time and memory use is very high.

The invention relates to a wireless camera motion control method and device, computer equipment and a storage medium. According to the application, a complete closed loop of image acquisition, wireless transmission, upper computer processing and motion control feedback can be constructed, so that the camera can automatically perform motion adjustment according to the difference between the actual position and the expected position of the target in the three-dimensional space. According to the method, an image is obtained through an image sensor, preprocessed and then wirelessly sent to an upper computer, the upper computer calculates the three-dimensional pose of a target and generates a control instruction, and finally, a core processing unit drives a camera to complete actions, so that the target in a shot picture is always kept at a specified position. Wireless, real-time and adaptive camera tracking control is realized in the whole process, manual operation is greatly reduced, and target capturing stability and shooting precision are improved.

A camera tracking system receives patient reference tracking information indicating pose of a patient reference array tracked by a patient tracking camera relative to a patient reference frame. A local XR headset view pose transform is determined between a local XR headset reference frame and the patient reference frame. Remote reference tracking information is received indicating pose of a remote reference array tracked by a remote reference tracking camera. A remote XR headset view pose transform is determined between a remote XR headset reference frame of a remote XR headset and the remote reference array. A 3D computer image is transformed from a local pose determined using the local XR headset view pose transform to a remote pose determined using the remote XR headset view pose transform. The transformed 3D computer image is provided to the remote XR headset for display with the remote pose relative to the remote XR headset reference frame.

The present invention provides a system for monitoring and improving toothbrushing effectiveness. The system comprises a camera device 10 comprising a camera or similar visual sensor configured to cap

The invention discloses a multi-target tracking method and device based on cross-camera trajectory association and a medium, for a target needing to be tracked, when the target appears in a video of a single scene for the first time, positioning is carried out, then the follow-up position of the target is predicted by combining the speed and the initial position of the target, and the follow-up position of the target is compared with that in an actual video to determine the target. And then, tracking targets across scenes and drawing a trajectory node space-time diagram, and associating trajectories in different scenes by using cosine similarity to realize multi-target tracking of cross-camera trajectory association. According to the invention, the actual moving tracks of the vehicle at different moments and different road sections can be visually displayed. Meanwhile, a correlation algorithm based on cosine similarity of a space-time diagram of track nodes is adopted, and track matching is carried out by using vehicle image blocks stored in the track nodes, so that the problem that a traditional cross-camera tracking method is sensitive to appearance changes of targets is effectively solved, and the accuracy and the real-time performance of cross-camera multi-target tracking are greatly improved.

The invention relates to a method for real-time multi-camera tracking of a target object comprising, for at least one new tracklet captured by at least one camera, calculating at least a first distance between the new tracklet and at least one reference tracklet belonging to the target object. The method also includes, when the smallest calculated first distance is between a first predetermined threshold value, called re-identification threshold, and a second predetermined threshold value, called human threshold, presenting the new tracklet to an operator, for the operator to assign or not assign the new tracklet to the target object, manually. The human threshold is a function of the number of reference tracklets in the reference list. The invention also relates to a computer program, to an object re-identification device and to an object re-identification system implementing such a method.

To synchronize the position change of a camera for photographing the live action of an actor with a video displayed on a background digital screen in virtual production.SOLUTION: The computer system identifying a first position of the physical camera corresponding to a first time period, rendering a first virtual scene for the first time period, projecting the first scene onto a display surface to determine a first rendered image for the first time period, and receiving a first camera image of the display surface from the camera during the first time period; Determining a first corrected position of the camera by comparing the first rendered image and the first camera image, predicting a second position of the camera corresponding to a second time period, rendering a second virtual scene for the second time period, and projecting the second virtual scene onto the display surface to determine a second rendered image for the second time period.SELECTED DRAWING: Figure 9

This application relates to the field of monitoring equipment control technology, and provides a control method, system, device, medium, and program product for a camera-PTZ linkage system. The method includes: after performing coordinate association calibration on the image pixels of the camera and PTZ camera in the camera-PTZ linkage system, detecting the pitch angle of the camera during operation; when a change in the camera's pitch angle is detected, correcting the initial coordinate mapping relationship between the calibrated camera and PTZ camera based on the change in the camera's pitch angle, and determining the corrected camera-PTZ coordinate mapping relationship; and controlling the PTZ camera to perform linkage tracking of the camera's monitoring image based on the corrected camera-PTZ coordinate mapping relationship. In this embodiment, when the camera's pitch angle changes, the coordinate mapping relationship between the camera and PTZ camera can be corrected according to the change in the camera's pitch angle, reducing the possibility of inaccurate PTZ camera tracking due to changes in the camera's lens angle and improving the accuracy of image tracking.

A camera tracking system for computer assisted navigation during surgery. The camera tracking system includes a processor operative to receive streams of video frames from tracking cameras which image a plurality of physical objects arranged as a reference array. For each of the physical objects imaged in a sequence of the video frames, that processor determines a set of coordinates for the physical object over the sequence of the video frames. For each of the physical objects, the processor generates an arithmetic combination of the set of coordinates for the physical object. The processor generates an array template identifying coordinates of the physical objects based on the arithmetic combinations of the sets of coordinates for the physical objects, and tracks pose of the physical objects of the reference array over time based on comparison of the array template to the reference array imaged in the streams of video frames.

The invention provides a dynamic SLAM system and method based on Gaussian representation. The dynamic SLAM system comprises a dynamic and static separation module, a pose estimation module and a dynamic and static mixed dense mapping module. Firstly, a rough dynamic area is obtained by utilizing geometric consistency characteristics represented by 3D Gaussian, the rough dynamic area is used as prompt information to guide accurate obtaining of a dynamic mask, then dynamic and static area division is carried out on a scene by utilizing the dynamic mask, and accurate camera pose estimation is realized based on the static area. And modeling of dynamic and static objects in the scene is realized through a mixed representation method combining static 3D Gaussian and dynamic 4D Gaussian in combination with backtracking frame optimization. The method can be applied to the fields of robot positioning and navigation, augmented reality and the like in a dynamic environment, and accurate reconstruction and robust camera tracking of a dynamic object can be completed at the same time.

A production process involves predetermined number of cameras simultaneously filming a background at predetermined angles, and filming actors in a studio with the same number of cameras and the same angles, used in conjunction with a virtual studio system. In studio, the actors perform before a green screen and the virtual studio system composites the actors onto the background in real-time. Camera tracking allows the in-studio cameras to pan, tilt, focus, zoom, and make limited other movements as the virtual studio system adjusts display of the background in a corresponding manner, resulting in a realistic scene without transporting actors and crew to the background location.

Embodiments of the present application provide a target tracking method and device, comprising: obtaining image position information of a target to be tracked in an image coordinate system of a first camera; converting the image position information from the image coordinate system to a world coordinate system to obtain world position information; determining a driving-out direction of the target to be tracked when the target drives out of a collection range of the first camera according to the world position information; obtaining a plurality of to-be-matched cameras adjacent to the first camera, and determining a relative direction of each to-be-matched camera relative to the first camera; selecting at least one second camera from each to-be-matched camera whose relative direction matches the driving-out direction; and performing cross-camera tracking on the target to be tracked according to image data of the first camera and image data of the second camera. Through the above method, the robustness of cross-camera tracking is improved.

This invention provides an automatic calibration method for master-slave camera tracking, comprising: S1 Initializing device state: manually controlling the PTZ camera to place the center of the PTZ camera image within the center range of the wide-angle camera image; S2 Data acquisition: the algorithm automatically controls the PTZ camera, simultaneously acquiring images from both the PTZ and wide-angle cameras and the P and T values ​​of the PTZ camera at each position, and saving them as a set of image pairs; S3 Data matching: using the SIFT matching algorithm to match the image pairs acquired in step S2, estimating the homography matrix of each image pair using the RANSAC method based on the matching results, and then solving for the coordinates u and v of the PTZ camera image center point corresponding to the wide-angle camera image; at this point, the PTZ camera P and T values ​​and u and v are matched to form a data pair (u... i v i P i T i ), n>6; S4 performs coordinate mapping solution: establish a geometric space model, and after solving, finally solve for the coordinate rotation matrix R.

The invention discloses a tracking control method and device of a camera, a storage medium and an electronic device, and the method comprises the steps: determining the type of a target object through employing a pre-trained target detection model under the condition of recognizing that the target object exists in the visual field of the camera; and matching a corresponding camera tracking parameter and a tracking prediction path according to the type of the target object. Controlling the rotation step length of a camera holder when the camera tracks the target object according to the camera tracking parameters, and controlling the rotation direction of the camera holder at the next moment when the camera tracks the target object according to the tracking prediction path; and controlling the camera to search the target object along the tracking prediction path under the condition of determining that the visual field of the target object is lost. Through intelligent target detection, path prediction of a matching target, holder control and a target search mechanism, active and accurate tracking of a target object is realized.

The invention discloses a mine inclined shaft vehicle safety pre-warning system and a mine inclined shaft vehicle safety pre-warning method based on multi-camera collaborative analysis and a dynamic locking mechanism, and the system realizes multi-camera linkage collaborative work by constructing a roadway topological map and establishing a spatial relationship and a locking state matrix among cameras. The system adopts a deep learning technology to carry out vehicle and personnel target detection, vehicle appearance features are extracted through a ReID network, and cross-camera tracking and motion direction judgment of the vehicle are realized in combination with a Kalman filter. When it is detected that a vehicle enters a specific road section, the system automatically updates the locking state of the area, the personnel detection module is started, and once it is found that personnel intrude into the locking area, sound-light alarm is triggered immediately and reported to the monitoring platform. According to the technical scheme, a curve monitoring blind area is effectively eliminated, the accuracy rate and the response real-time performance of man-vehicle conflict early warning are remarkably improved, reliable technical guarantee is provided for mine transportation safety, and good practicability and popularization value are achieved.

The invention discloses a tower perimeter security and protection method and system based on multi-view fusion, relates to the field of high-voltage power transmission infrastructure security and protection, and aims to solve the problems of limited monitoring range, low recognition precision, rigid alarm and difficulty in deployment, operation and maintenance of an existing tower security and protection system. The method comprises the steps of video acquisition, preliminary screening of motion, target detection and tracking, dynamic defense area decision making and the like, is suitable for a remote scene without commercial power, and can realize accurate and intelligent security and protection of the periphery of a tower. The system comprises a multi-view perception layer, an edge calculation and communication layer, a cloud intelligent analysis platform and an intelligent power supply layer, wherein the multi-view perception layer realizes 360-degree dead-angle-free coverage through a high-definition intelligent camera; the edge calculation unit executes motion detection, improved YOLOv8 target detection and single-camera tracking; the cloud platform realizes cross-camera target association through a graph neural network, and dynamically adjusts the defense area level; the intelligent power supply layer ensures long-term field operation.

A camera tracking system receives patient reference tracking information indicating pose of a patient reference array tracked by a patient tracking camera relative to a patient reference frame. A local XR headset view pose transform is determined between a local XR headset reference frame and the patient reference frame. Remote reference tracking information is received indicating pose of a remote reference array tracked by a remote reference tracking camera. A remote XR headset view pose transform is determined between a remote XR headset reference frame of a remote XR headset and the remote reference array. A 3D computer image is transformed from a local pose determined using the local XR headset view pose transform to a remote pose determined using the remote XR headset view pose transform. The transformed 3D computer image is provided to the remote XR headset for display with the remote pose relative to the remote XR headset reference frame.

The application relates to the field of computer vision and navigation, and provides a SLAM acceleration method based on three-dimensional Gaussian rendering, which comprises the following steps: initializing a three-dimensional Gaussian point cloud and camera pose parameters based on input data, performing camera tracking on a target frame in the input data, and outputting optimized camera pose parameters based on a staged switching strategy and an iteration mechanism, wherein the camera tracking utilizes a reduced-precision floating-point format; performing a hierarchical screening mechanism in the optimized camera pose parameters to determine key frames from color images and depth images, using the key frames and the optimized camera pose parameters, and performing mapping through a renderer supporting multi-channel data to update the three-dimensional Gaussian point cloud. The method can quickly converge the pose optimization through staged switching and iteration, uses the reduced-precision floating-point format to reduce the numerical processing burden, and the hierarchical screening mechanism can reduce the calculation cost of frame-by-frame fine comparison, thereby solving the problem of low efficiency of the SLAM process.

A camera tracking system for computer assisted navigation during surgery. Operations identify stray markers in a frame of tracking data from tracking cameras, and identify stray markers of a reference array. Stray markers of the reference array are designated assigned status and, otherwise, designated unknown status. The operations designate other of the assigned status stray markers and any of the unknown status stray markers along a same epipolar line of the tracking cameras as one of the assigned status stray markers as being epipolar ambiguous status. For each one of the epipolar ambiguous status stray markers, the operations estimate 3D locations where phantom markers can appear in the frame based on epipolar ambiguity of the tracking cameras when determining location of the one of the epipolar ambiguous status stray markers. The operations designate the unknown status stray markers within a threshold distance of the estimated 3D locations of phantom markers as being phantom status.

The application discloses a kind of cascading array microphone based on DOA's camera tracking method and system, comprising the following steps: setting cascading array microphone in meeting place, including main array microphone and slave array microphone;Camera ball machine is set around cascading array microphone;It also includes several azimuth zones, each azimuth zone is bound with the preset position of a camera ball machine, and corresponding information is stored to database, camera ball machine is aimed at preset position;Sound source information in meeting place is acquired by cascading array microphone;The DOA information and sound source energy value of slave array microphone are uploaded to main array microphone;DOA information is handled by main array microphone, and all sound source energy values collected are compared, the azimuth zone of the array microphone corresponding to the maximum sound source energy value is selected, and is uploaded to host computer;Host computer calculates the preset position corresponding to azimuth zone, and after anti-interference processing, the camera of camera ball machine is called to shoot preset position.

The application discloses a real-time cross-camera vehicle tracking method, comprising the following steps: performing single-camera vehicle detection and tracking based on a video stream collected in real time by multiple cameras to obtain local vehicle trajectories corresponding to each camera respectively; constructing a global trajectory buffer and storing the obtained local vehicle trajectories in the global trajectory buffer; performing trajectory association matching on the local vehicle trajectories and existing clustered trajectories in the global trajectory buffer; updating the clustered trajectories or creating new clustered trajectories according to the association matching result to obtain multi-camera vehicle trajectories with unique global IDs. The application has the beneficial effects that: the incremental clustering algorithm and the sliding window management are adopted to realize cross-camera tracking with a response of seconds, and the delay is greatly reduced compared with the traditional batch processing method, thereby meeting the strict requirements of real-time monitoring.