570 results about "Monocular vision" patented technology

The invention discloses an end-to-end monocular visual odometer method fusing space-time semantic information. According to the method, continuous image sequence frames are collected through a color monocular camera, and a multi-information fusion end-to-end deep learning framework is constructed; a heterogeneous training domain is adopted to set various data set course sharing parameter fusion training, continuous image sequences are input, and the end-to-end deep learning framework is dynamically coupled with hidden state feature vectors output historically, so that a feature mapping relation of time sequence perception is formed; and interpretable feature decoupling of the static background elements and the dynamic entity objects in the scene is realized. After iterative feature fusion, the system outputs sparse depth and camera motion poses which conform to scene geometric constraints, so that a camera trajectory estimation model with high robustness and strong generalization ability in a complex environment is constructed. According to the method, the positioning precision and stability of the monocular vision odometer are remarkably improved.

The invention relates to a dynamic environment semantic SLAM method based on monocular vision and LiDAR fusion, and belongs to the technical field of robots. The method comprises the following steps: 1) performing instance segmentation on point clouds to obtain unknown motion state point clouds, pure static road sign point clouds and ground attribute point clouds, identifying semi-static and dynamic road sign point clouds through initial pose estimation and Kalman filtering, and then performing accurate pose estimation; 2) segmenting and enhancing the image instance, extracting SuperPoint feature points, projecting a dynamic road sign to the segmented image so as to distinguish static and non-static points, tracking static features by adopting a plurality of tracking strategies, and optimizing the pose by utilizing a local map; 3) carrying out heterogeneous sensor attitude fusion; 4) outputting an optimized track by combining two loopback detection results; and 5) constructing a global pure static semantic map. According to the method, the dynamic environment influence is considered, and the positioning precision and the mapping quality of the system in the dynamic complex environment are improved.

The invention discloses a monocular vision inertial odometer positioning method based on self-adaptive mixed vision residual errors, and belongs to the technical field of robot navigation positioning. The method comprises the following steps: firstly, respectively carrying out feature point extraction and pre-integration on data of a monocular camera and an inertial measurement unit (IMU), and selecting a strong pixel gradient region according to a feature extraction result; predicting a candidate feature point search window based on IMU (Inertial Measurement Unit) motion information, accelerating feature matching and reducing mismatching, and selecting a key frame at the same time; introducing a direct method model on the basis of feature extraction and matching to construct a luminosity error; a feature reliability evaluation method is designed, and a dynamic factor is adopted to adaptively select and optimize a hybrid visual residual projection error or a luminosity error in real time; and at the back-end optimization stage, a sliding window-based graph optimization method is adopted to optimize and combine with the optimization model to perform pose estimation, so that the purpose of improving the positioning accuracy of the visual inertial odometer in a low-light weak-texture scene is achieved.

The invention discloses a field personnel positioning method cooperating with multi-view images and spatial vectors of an unmanned aerial vehicle, and the method comprises the steps: detecting a personnel target in real time through a pre-trained lightweight model, obtaining a surrounding radius in combination with a self-adaptive surrounding radius algorithm, and achieving the collection of a multi-view image sequence; carrying out quality quantitative analysis on the multi-view image sequence by adopting a multi-source fixed weight quality evaluation algorithm; a DBOF coordinate system is established, and coordinate conversion is completed; extracting high-frequency matching points based on a cross-image bidirectional matching mechanism, and performing composite distortion correction on the high-frequency matching points; carrying out weighted least square method solution on the multi-view observation equation set to generate a three-dimensional space coordinate of the target point in a DBOF coordinate system; and introducing a robust iterative optimization strategy to eliminate abnormal values, and outputting geographic coordinates. On the basis of low-cost monocular vision, the real-time performance and robustness of personnel positioning in a complex field environment are improved through a dynamic surround sampling and geometric resolving mechanism, and spatial data support is provided for field rescue.

The invention discloses a three-dimensional color modeling method and system based on multi-sensor fusion, and the method comprises the steps: obtaining monocular vision pose information, an RGB image and radar three-dimensional point cloud information, enabling the monocular vision pose information to be used for correcting a radar pose, and generating corrected point cloud information; the corrected point cloud information is screened and then projected to a specific plane; proper feature points are extracted from the projection image and the RGB image, and according to the corresponding relation between the feature points of the projection image and the RGB image, the plane point cloud of the projection image and the pixel coordinate plane of the RGB image are registered; performing fusion modeling on the point cloud and the image data by using an MRF model; according to the method, the radar pose is corrected by using visual pose information, the accumulated error influence of a system is reduced, the corrected point cloud is projected to a plane, and feature points of a projection image and a color image are extracted to establish an MRF model for describing the relationship between a pixel plane and the corresponding point cloud, so that a point cloud image which is uniformly distributed in space and contains color information is obtained.

The invention relates to a monocular vision tower span measurement method based on RTK correction, and belongs to the technical field of space distance measurement. Comprising the following steps: acquiring ambient light intensity and quality in real time based on an environment where a power transmission and distribution tower is located, and obtaining ambient light adaptation parameters by analyzing ambient light data and adjusting ISO sensitivity and shutter speed of a camera. According to the method, real-time environment light data analysis and camera parameter automatic adjustment are introduced in the monocular vision tower span measurement process, the image quality is effectively improved, tower recognition is more accurate, background interference is reduced, the geographic position and the shooting angle of a shooting point are corrected through the RTK technology, and the measurement accuracy is improved. And perspective transformation correction is carried out on the image in cooperation with a perspective projection theory, and the accuracy of distance measurement data is improved to centimeter level.

The invention relates to the technical field of settlement observation, in particular to a non-contact monocular vision high-precision automatic settlement monitoring method, which comprises the following steps: firstly, performing target monitoring by using a YOLOv10s model, improving the real-time reasoning speed and recognition precision of target recognition, and solving the target recognition problem under a complex background; secondly, a digital image processing technology and an edge-based least square ellipse fitting algorithm are combined to realize accurate positioning of the target; thirdly, deriving based on a camera imaging principle, and introducing an improved world coordinate calculation method to solve world coordinates corresponding to the pixel coordinates so as to obtain three-dimensional world coordinates; and finally, calculating a settlement value of the monitoring area by taking the first frame image as a reference, predicting a future settlement value by using a deep learning model, and making an early warning for structure safety in advance. According to the invention, the installation demand and the calculation overhead are reduced, and the settlement monitoring precision is improved.

The embodiment of the invention provides an unmanned aerial vehicle landing system, an unmanned aerial vehicle landing control method and an unmanned aerial vehicle, the system comprises an unmanned aerial vehicle, a hangar monocular vision sensor and a hangar radar, the unmanned aerial vehicle comprises an unmanned aerial vehicle monocular vision sensor and a processor, and an unmanned aerial vehicle state machine runs on the processor. The hangar monocular vision sensor and the hangar radar are mounted on the hangar; a hangar two-dimensional code is arranged on the hangar, and a landing point is arranged on the hangar two-dimensional code; the unmanned aerial vehicle monocular vision sensor is used for determining a first pose of the unmanned aerial vehicle monocular vision sensor relative to the landing point; the hangar monocular vision sensor is used for determining a second pose of a marker on the unmanned aerial vehicle relative to the landing point; the hangar radar is used for determining the position of the unmanned aerial vehicle; and the unmanned aerial vehicle state machine is used for generating a control instruction according to the first pose, the unmanned aerial vehicle position and the second pose, and controlling the unmanned aerial vehicle to land to the landing point according to the control instruction. According to the invention, the problem that the unmanned aerial vehicle cannot land accurately is solved.

The invention relates to a tunnel surrounding rock three-dimensional deformation monitoring system integrating monocular vision and millimeter wave radar. The tunnel surrounding rock three-dimensional deformation monitoring system comprises in-tunnel sensing equipment, a tunnel opening terminal, a database and application server and a remote monitoring end. The invention also relates to a tunnel surrounding rock three-dimensional deformation monitoring method fusing monocular vision and millimeter wave radar. The method comprises the following steps: S1, setting a monitoring target; s2, multi-modal data acquisition is carried out; s3, visual and radar target detection; s4, performing visual and radar heterogeneous data matching; s5, solving pixel displacement and radial displacement; s6, calculating the three-dimensional deformation of the surrounding rock; and S7, alarm judgment and result output. According to the invention, the precision, the real-time performance and the automation level of tunnel surrounding rock deformation monitoring can be obviously improved, and the method is especially suitable for scenes with complex environment and high safety risk in the early stage of tunnel excavation; the invention aims to provide an efficient, accurate and automatic monitoring means capable of adapting to the early stage of excavation for tunnel surrounding rock deformation.

The invention belongs to the technical field of computer vision, and discloses a human motion capture method based on mask perceptual graph convolution and skeleton prior, and the method comprises the steps: obtaining an input image and a corresponding human mask; extracting features of the image and the mask and coding; constructing a mask perception graph convolutional network, constructing an adjacent matrix of the graph convolutional network by using mask information, and applying mask constraint loss to enhance the expression ability of the human body region features of the image; a skeleton prior decoupling network is constructed, skeleton and vertex information of an SMPL model is used as prior, a cross attention mechanism is combined, and multi-modal data enhancement of SMPL skeleton node features is guided through image features; and finally outputting a three-dimensional joint position coordinate and a shape grid vertex coordinate. According to the method, the local feature consistency is enhanced through mask perception image convolution, the geometric consistency is improved in combination with skeleton prior, and the precision of three-dimensional human body motion capture under monocular vision is effectively improved.

The invention provides a crack three-way measurement research method and system based on multi-coordinate system fusion, and relates to the technical field of earth and rockfill dam monitoring. The method comprises the following steps: shooting concentric circle target images at multiple angles through a monocular camera, extracting feature points through preprocessing, and correcting projection distortion of non-parallel shooting; establishing a mapping relation between pixels and a three-dimensional coordinate system, and optimizing camera pose parameters; reconstructing a three-dimensional coordinate point cloud based on the multi-view data, screening outliers and fitting plane optimization data; and registering the three-dimensional point cloud to a world coordinate system through a coordinate system conversion algorithm, and comparing two measurement results to calculate the three-way displacement of the crack. According to the method, concentric circle target constraint feature precision is utilized, projection correction and plane fitting are combined to compensate monocular vision depth deficiency, high-precision displacement detection is achieved, and engineering monitoring requirements are met.

The invention discloses a visual language navigation method and system based on a monocular camera and a language instruction, and the method comprises the steps: S1, feature field construction: employing a 3DGS to construct a 3DGS feature field by employing visual information obtained by a monocular RGB-D camera; s2, implicit partial complementation: inferring the representation of a missing region through a context feature relationship to generate a complete feature map; s3, waypoint prediction: generating an aerial view feature map based on the 3DGS feature field, and predicting nearby navigable waypoints through a waypoint predictor; s4, based on active perception of uncertainty, when the robot is uncertain in navigation decision, more visual information is acquired and missing visual information is supplemented by rotating a camera; and S5, constructing a topological map based on the constructed panoramic feature map and the predicted path points, and carrying out navigation decision making by using a panoramic visual language navigation planning model in combination with a language instruction. The problem that monocular vision information is incomplete is effectively solved, and the success rate and efficiency of navigation are remarkably improved.

The invention discloses a monocular vision depth estimation method and system suitable for a night scene, and aims to overcome the limitation problem of a basic model. According to the method, daytime images are synthesized into highly vivid night images, then a daytime-night data set is formed, and a low-light-parameter efficient fine tuning strategy is provided for adjusting a pre-trained basic model to adapt to a night scene. Finally, the method not only can effectively cope with challenges in a night scene, but also realizes robust depth estimation performance, thereby encouraging further exploration and innovation in the field.

The invention belongs to the technical field of ship management, and provides a combined intelligent ship water diversion system integrating differential positioning, a laser radar and a camera. In the shore end RTK-inertial navigation depth fusion positioning module, extended Kalman filtering fusion is carried out on data acquired by shipborne equipment, so that continuity of accurate positioning is realized; the construction of a color high-precision map is realized through a laser radar-monocular vision color high-precision modeling module, details such as wharf fender materials and signboard characters can be better distinguished, and the problem of shoreline modeling defects is solved; visualization of obstacles is achieved through the multi-scene dynamic obstacle detection and fusion module, meanwhile, through cooperation of multi-source data, the precision and stability of ship positioning are greatly improved, higher precision and reliability are achieved, and the precision of a guide instruction is improved; through cooperation of the 5G private network and the visualization module, the data transmission efficiency is improved, the timeliness is improved, and transmission delay can be avoided.

The invention discloses a robot joint layer sensing system fusing ToF and monocular vision, and belongs to the field of robot sensing, the system adopts an Eye-in-Hand architecture, and a ToF camera and an RGB camera are integrated; the quality of the depth map is improved by combining bilateral filtering and Kalman filtering; the six-degree-of-freedom pose estimation of the target is realized based on SURF feature matching and KD-Tree accelerated ICP registration; and planning a grabbing path in combination with an improved RRT algorithm and hierarchical collision detection. According to the robot joint layer sensing system fusing ToF and monocular vision provided by the invention, spatial alignment and fusion are carried out on two types of visual information, respective limitations are made up, and the robustness and precision of the whole sensing system are improved; in a single-arm and double-arm grabbing experiment, the positioning precision of the system reaches + / -3mm, and the dynamic scene grabbing success rate is high. The method is suitable for the fields of industrial automation, logistics sorting and the like.

The invention belongs to the technical field of robot positioning and three-dimensional mapping, and discloses a high-precision map reconstruction method and system based on monocular vision, a medium and equipment. Scene image data are acquired through a monocular image acquisition module, after feature extraction and matching are performed on each frame of image, attitude information acquired by an inertial measurement unit is fused, and pose calculation of a robot is completed through a sparse vision SLAM system; meanwhile, an image dense depth map is generated by a monocular depth estimation model based on an attention mechanism, and the image dense depth map is converted into a single-frame color dense point cloud in combination with image RGB color information. According to the invention, a loose coupling fusion strategy is adopted to carry out spatial registration on a single-frame colored dense point cloud and a robot pose at a corresponding moment, multi-frame data fusion is completed through point cloud splicing and optimization, and a globally consistent three-dimensional dense point cloud map is constructed to realize scene modeling. The method has the characteristics of high robustness and high reconstruction precision, and can be effectively applied to three-dimensional map construction in an outdoor complex environment.

The invention belongs to the related technical field of bridge hoisting, and discloses a monocular vision-based bridge hoisting pose real-time measurement method and system, and the method comprises the steps: calibrating the relative position between an optical calibration plate and the upper surface of a prefabricated bridge before hoisting operation: shooting a plurality of calibration images of the upper surface of the prefabricated bridge; obtaining a calibration plate coordinate system and an image coordinate system projection transformation matrix of any image; calculating projection coordinates of the vertex of the upper surface of the prefabricated bridge in the calibration plate coordinate system; on the basis of the multiple calibration images, shape features of the upper surface of the prefabricated bridge are obtained, and the relative position relation is calculated; and measurement in the hoisting process: obtaining the pose of the prefabricated bridge according to the real-time image and the relative position relation obtained through calibration. According to the method, the relative position mapping relation between the optical calibration plate and the prefabricated bridge is accurately established by utilizing multi-view image analysis and geometric constraint calibration before hoisting operation, so that the pose of the prefabricated bridge is accurately measured in real time on the basis of monocular vision.

The invention belongs to the technical field of conveyor coal flow monitoring, and aims at solving the problem that an existing coal flow monitoring method is insufficient in precision. According to the belt conveyor coal flow monitoring method based on the inspection robot and the monocular vision, coal flow images are dynamically collected through the inspection robot, accurate segmentation of a coal flow area is achieved through a deep learning algorithm, and a coal flow three-dimensional point cloud is reconstructed in combination with a monocular depth estimation technology; and finally, qualitative and quantitative analysis of the coal flow is realized. Compared with a traditional method, the method has the advantages of non-contact measurement, full conveying belt coverage, high calculation efficiency, adaptability to complex environments and the like, and reliable data support can be provided for a coal mine intelligent transportation system.

The invention discloses a material movement behavior monitoring method and system based on monocular vision, and relates to the field of material movement monitoring, and the method comprises the steps: firstly collecting a material movement video through a monocular high-speed camera, and extracting a continuous state sequence of a material on a two-dimensional image plane through employing an instance segmentation and time sequence tracking technology; the method is characterized in that a time sequence deep learning model is introduced to analyze a state sequence, and a key event that a material collides with a working surface of equipment is automatically identified, so that the collision moment and contact point coordinates are accurately obtained. In this way, the collision contact point is used as a geometric anchor point for three-dimensional space back calculation, a two-dimensional image track is reversely projected and reconstructed into a three-dimensional space motion track in combination with internal parameters of a camera and physical parameters of an equipment reference plane, then key kinetic parameters such as the speed and the recovery coefficient are calculated, and low-cost and high-precision automatic online monitoring is achieved.

The invention discloses a method for positioning a target on a plane based on monocular vision. The method comprises the following steps: determining related parameters of a shooting device on a target plane and an external parameter relationship between the shooting device and the target plane; calibrating the shooting device, determining an internal reference matrix and a distortion coefficient of the shooting device, and performing distortion correction on the internal reference of the shooting device; establishing a three-dimensional coordinate system on the target plane, shooting an image, establishing a pixel coordinate system on the image, and determining pixel coordinates of a target point; constructing a visual cone model of the shooting device, and calculating the coordinates of the target point in a target plane coordinate system according to the pixel coordinates of the target point, the related parameters and the external parameter relationship between the shooting device and the target plane; according to the coordinates of the target point in the target plane coordinate system, adjustment calculation is carried out through a linear interpolation method to obtain the final position coordinates of the target point in the target plane, and target positioning on the plane is completed.

The invention relates to the technical field of positioning of planar robots, in particular to a wheel type odometer and monocular vision fused SLAM positioning system with three driven omnidirectional wheel sensors. According to the method, firstly, internal reference calibration is conducted on a sensor, then the pose of a vehicle body is calculated according to data of a wheel type odometer sensor, joint initialization of a wheel type odometer and a monocular camera is conducted after timestamp synchronization is completed, and the pose of the vehicle body is calculated through vision-wheel type odometer tight coupling nonlinear sliding window optimization. A dynamic plane constraint self-adaptive method based on local geometric features and meta learning is added, a mixed strategy based on motion component decomposition and residual entropy dynamic adjustment is designed, back-end optimization is carried out, and whether a loop exists or not is detected to optimize a track and reduce errors.

The invention relates to a multi-modal perception fusion target following method and system. The method comprises the following steps: adopting an improved KCF algorithm to realize a closed-loop process of target tracking, multi-scale space construction, feature fusion, response calculation, optimal scale decision and model updating; designing a four-level shielding processing mechanism fusing motion prediction and depth verification, and realizing tracking recovery in a shielding scene through shielding judgment, motion prediction, fine search and template protection; constructing a target distance mapping function fusing geometric distortion correction and attitude compensation, and realizing high-precision distance estimation based on monocular vision; laser radar point cloud information is integrated, an obstacle threat degree model is constructed, and cooperative path planning of following and obstacle avoidance is realized in combination with an improved TEB algorithm; and designing a linear velocity control law and an angular velocity control law based on the distance deviation and the azimuth angle deviation, and driving the robot to complete target following motion. According to the invention, high-precision and robust following of the robot to the target can be realized.

The invention belongs to the technical field of visual angle follow-up vehicle panoramic look-around, and particularly relates to a visual angle follow-up vehicle panoramic look-around system based on dynamic prediction, which is characterized in that monocular vision and laser radar time sequence data are fused, and a dense and stable dynamic target depth sequence is generated through dynamic target initial perception, cross-modal time sequence alignment and inter-frame depth correlation modeling; through environment interaction modeling and target behavior intention prediction double-branch parallel processing, a dynamic self-adaptive accurate track is output in combination with a weighted fusion strategy, and the prediction weight dynamic adjustment requirement of a complex motion scene is adapted; after a virtual view angle image is preliminarily synthesized based on the predicted trajectory, deviation detection and coordinate dynamic correction are realized through inter-frame target consistency verification, and artifact area positioning and pixel-level restoration are synchronously completed; multi-modal preprocessing and algorithm tasks are deployed according to hardware, and a model lightweight optimization and inter-frame prediction result multiplexing strategy is combined, so that low-delay operation of the system in a high-resolution panoramic image output scene is ensured.

The invention relates to a monocular vision-based map construction method, system and equipment. The monocular vision-based map construction method comprises the following steps of: acquiring a current frame image, a key frame image and a sparse point cloud map; extracting image information of the current frame image, and determining a first camera attitude; determining a prior depth map of the current frame image on the key frame image; constructing a first Gaussian model according to the first camera attitude, the image information, the sparse point cloud map, the prior depth map and the current frame image, and optimizing the first camera attitude to obtain a second camera attitude; wherein the Gaussian model comprises Gaussian distribution of a plurality of scenes; and obtaining a second Gaussian model according to the sparse point cloud map, the prior depth map, the image information and the second camera attitude, and adding the second Gaussian model to the Gaussian map according to the second camera attitude. The method can improve the positioning and map building precision, and is applied to the technical field of computer vision and robots.

The invention provides a monocular vision and sparse IMU rehabilitation action whole body posture estimation method and system, and the method comprises the steps: synchronously collecting video data and inertial data of human body rehabilitation actions through a monocular RGB camera and a plurality of IMUs, and cutting and zooming an image to a preset resolution; extracting a key point thermodynamic diagram from continuous N frames of images by using a sliding window and a residual neural network, and calculating 2D key point pixel coordinates of each frame; splicing the N frames of 2D key point pixel coordinates, the rotation matrix of the IMU and the acceleration signal into an input sequence; cross-modal time sequence modeling is carried out on an input sequence through time Transform, and after high-dimensional features are extracted, weighted average is carried out through a convolutional layer, and 3D relative key point coordinates of the last frame are output through a regression head. According to the method, by fusing monocular vision and sparse IMU cross-modal data, the problem of visual information loss caused by limb self-shielding is effectively solved, and the defect that a traditional pure vision method is insufficient in precision in rehabilitation actions is overcome.

This invention discloses a lane departure warning method and system for use in nighttime scenarios. Developed primarily based on monocular vision and deep learning methods, the system includes: using an on-board camera to capture images of the road ahead and using the Res2Net50-VHA network to detect lane marking feature points on both sides of the current lane; using a Kalman filter to track lane marking feature points in preceding and following frame images; applying a combined fitting method of quadratic polynomials and linear equations to fit the upper and lower portions of the lane markings; calculating relevant parameters, including road curvature, lane centerline equation, vehicle lateral distance from the lane centerline, and yaw angle; and combining these parameters to determine the vehicle's driving state and implement a lane departure warning strategy. The algorithm employed in this invention boasts high recognition accuracy, high computational efficiency, and strong anti-interference capabilities, and the program outputs intuitive, real-time visualizations.

The invention provides an intelligent and accurate fruit picking method based on visual perception, relates to the field of fruit picking, and adopts a picking robot comprising a movable chassis (1), a walking mechanism (2), an industrial computer (3), a collecting frame (4), a picking mechanical arm (5), an end effector (6), a monocular vision camera (7), a binocular stereo camera (8) and a blowing assembly (9). According to the picking robot, through recognition of the binocular stereo camera (8), by means of blowing of the blowing assembly (9) and through a timestamp, the relative motion state between a picking point and sheltered fruits after blowing is obtained, and therefore sensing and precise clamping and shearing of the picking point in the motion state are achieved under the condition that the fruits are sheltered and subjected to complex motion after sheltering is removed; the picking efficiency is improved, the equipment damage probability is reduced, and the harvesting cost is controlled.

The invention discloses a monocular visual odometer method and system based on aerial view representation and differentiable weighted Procrustes solver, and the method comprises the steps: constructing a pose estimation model which can carry out the extraction and matching of key points in an aerial view, and carries out the relative pose estimation through an interpretable weighted Procrustes solver; the method comprises the following steps: constructing a loss function based on pose information, optimizing parameters of a pose estimation model by using the loss function, and performing pose information estimation between two frames of monocular images by using the parameter-optimized pose estimation model to realize the monocular visual odometer, so that the monocular visual odometer is superior in the aspect of reducing scale drift; and stable performance is shown under different weather and illumination conditions.

The invention discloses a rapid robust monocular vision inertial positioning method and system. The method comprises the following steps: configuring IMU and monocular camera sensor parameters; preprocessing the IMU data; iMU attitude initialization is carried out; image features of the monocular camera are extracted, abnormal matching points are removed, and IMU pre-integration is carried out; monocular vision initialization is carried out; odometer attitude optimization: judging whether the feature points are mismatched or not based on an IMU pre-integration result, constructing a feature point re-projection error jacobian matrix, and performing blocking and diagonalization processing on the re-projection error jacobian matrix to respectively optimize inverse depths and image frame attitudes of the feature points; selecting a key frame based on a key frame identification rule after the current frame attitude update is completed; and judging according to the current frame, and removing a certain frame in the sliding window to reserve a sliding window space for adding the latest frame. According to the method, the inverse depth of the feature points and the image frame attitude are optimized respectively based on the counterweight projection error Jacobian matrix partitioning and diagonalization processing, and the calculation complexity is reduced.

The invention relates to the technical field of multi-person posture estimation, in particular to an interactive scene data synthesis and key point visibility updating algorithm based on monocular vision, which comprises the following steps of: 1, arranging single-person picture data under monocular shooting vision, processing an input single-person picture under monocular shooting vision by using a GrondingDINO visual language open type target detection model, and obtaining a target detection result; and the Person is used as a retrieval keyword to identify a character individual in the picture. Through fine processing of the steps, especially introduction of target matching, target position adjustment and key point visibility updating methods, the quality of the synthesized image is greatly improved, the fusion degree of the target and the background is optimized through accurate matching and natural target pasting, and the image quality is improved. The space consistency and the visual naturalness of the synthetic image are enhanced, the problem of overfitting is effectively avoided through the improvement, the diversity of training data is enhanced, and therefore the generalization ability of the model is improved.