A5023955044- Robotics and Sensor-Based Localization
- Advanced Vision and Imaging
- Optical measurement and interference techniques
- Advanced Image and Video Retrieval Techniques
- 3D Surveying and Cultural Heritage
- Image and Object Detection Techniques
- Advanced Memory and Neural Computing
- Image Processing Techniques and Applications
- Satellite Image Processing and Photogrammetry
- Advanced Neural Network Applications
- Optical Systems and Laser Technology
- Structural Health Monitoring Techniques
- Robotic Path Planning Algorithms
- Advanced Measurement and Detection Methods
- Remote Sensing and LiDAR Applications
- Infrared Target Detection Methodologies
- Robotic Mechanisms and Dynamics
- Image Enhancement Techniques
- Advanced Measurement and Metrology Techniques
- Space Satellite Systems and Control
- Radiation Detection and Scintillator Technologies
- Autonomous Vehicle Technology and Safety
- Radiation Effects in Electronics
- Optical Coherence Tomography Applications
- Visual Attention and Saliency Detection
National University of Defense Technology
2016-2025
Institute of Navigation
2024
Pose estimation and tracking of objects is a fundamental application in 3D vision. Event cameras possess remarkable attributes such as high dynamic range, low latency, resilience against motion blur, which enables them to address challenging range scenes or high-speed motion. These features make event an ideal complement over standard for object pose estimation. In this work, we propose line-based robust method planar non-planar using camera. Firstly, extract lines directly from events, then...
Three-dimensional (3D) imaging with structured light is crucial in diverse scenarios, ranging from intelligent manufacturing and medicine to entertainment. However, current methods rely on projector–camera synchronization, limiting the use of affordable devices their consumer applications. In this work, we introduce an asynchronous approach based generative deep neural networks relax synchronization constraint, accomplishing challenges fringe pattern aliasing, without relying any a priori...
In this paper we present four cases of minimal solutions for two-view relative pose estimation by exploiting the affine transformation between feature points and demonstrate efficient solvers these cases. It is shown, that under planar motion assumption or with knowledge a vertical direction, single correspondence sufficient to recover camera pose. The considered are calibrated cameras as closed-form least-squares solution, solution unknown focal length case known direction. These algorithms...
In this paper we present minimal solutions for two-view relative motion estimation based on a homography formulation. By assuming known vertical direction (e.g. from an IMU) and dominant ground plane demonstrate that rotation translation can be decoupled. This result allows us to reduce the number of point matches needed compute hypothesis. We then derive different algorithms decoupling allow efficient estimation. also how these used efficiently optimal inlier set using exhaustive search or...
In this article, we present four cases of minimal solutions for two-view relative pose estimation by exploiting the affine transformation between feature points, and demonstrate efficient solvers these cases. It is shown that under planar motion assumption or with knowledge a vertical direction, single correspondence sufficient to recover camera pose. The considered are calibrated cameras as closed-form least-squares solutions, solution unknown focal length, case known direction. These...
In this paper, we propose a self-calibration approach to stereo cameras with radial distortion from image pairs of common 3D scene. Based on the epipolar constraint in pair, intrinsic and extrinsic parameters are estimated synchronously minimum number nine point correspondences. It is significant within random sample consensus (RANSAC) scheme cope outliers feature matches efficiently robustly. Then inliers pair that have been determined after RANSAC used optimize calibration cameras....
In this paper, we present a robust technique of stereo calibration using homography constraints. Our method is novel as performed by solving polynomial equation system including two radial distortion parameters, minimal number five image point correspondences. This enables us to calibrate from only pair images planar scene, and provide the exact algebraic solution problem. The case useful reduce computation time increase robustness when random sample consensus (RANSAC) correspondences pair....
基于多项式拟合的标定方法可在缺乏相机曝光时间的条件下,获取相机响应函数(CRF)曲线和图像曝光比,具备广泛的适用性。然而该方法存在迭代发散和标定精度不高问题,影响其实际应用。本文通过分析传统多项式拟合标定方法流程,发现在全局误差函数条件下,标定数据集合中存在大量无效项,既减少了有效标定数据,又降低了图像曝光比迭代计算精度。针对这一问题,提出了一种改进的联合局部误差函数标定方法,可在两幅曝光相近的图像间选取标定数据,避免引入无效项,使得计算多项式系数和曝光比的数据一致。在公开数据集和某工业相机拍摄数据集上的标定结果表明,改进方法具有较好的收敛性,相比于传统方法,颜色三通道CRF曲线分布更加紧凑,通道间曝光比平均偏差分别减少了49.83%和42.25%。
We propose three novel solvers for estimating the relative pose of a multi-camera system from affine correspondences (ACs). A new constraint is derived interpreting relationship ACs and generalized camera model. Using constraint, we demonstrate efficient two types motions assumed. Considering that cameras undergo planar motion, minimal solution using single AC solver with to overcome degenerate case. Also, known vertical direction, e.g., an IMU. Since proposed methods require significantly...
Calibration of the stereo vision systems is a crucial step for precise 3D measurements. Restricted by outdoors’ large field view (FOV), conventional method based on calibration boards not suitable since process time consuming and accuracy guaranteed. In this paper, we propose estimating extrinsic parameters system aided an inclinometer range sensor. Through given sensors, initial rotation angle translation vector are pre-established solving set linear equations. The metric scale determined...