Yuhua Xu

ORCID: 0000-0002-7427-0156
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Advanced Vision and Imaging
  • Robotics and Sensor-Based Localization
  • Robotic Path Planning Algorithms
  • Optical measurement and interference techniques
  • Image Enhancement Techniques
  • Advanced Image Processing Techniques
  • Image Processing Techniques and Applications
  • Control and Dynamics of Mobile Robots
  • Image and Object Detection Techniques
  • Advanced Image and Video Retrieval Techniques
  • Image and Video Stabilization
  • 3D Surveying and Cultural Heritage
  • Medical Image Segmentation Techniques
  • Remote Sensing and LiDAR Applications
  • Human Pose and Action Recognition
  • Adaptive Control of Nonlinear Systems
  • Simulation and Modeling Applications
  • Human Motion and Animation
  • Infrared Thermography in Medicine
  • Advanced Measurement and Detection Methods
  • Infrared Target Detection Methodologies

Hefei University of Technology
2009-2022

National University of Defense Technology
2013-2018

Hunan University
2014

Hefei University
2008

Real-time performance of stereo matching networks is important for many applications, such as automatic driving, robot navigation and augmented reality (AR). Although significant progress has been made in recent years, it still challenging to balance real-time accuracy. In this paper, we present a novel edge-preserving cost volume upsampling module based on the slicing operation learned bilateral grid. The layer parameter-free, which allows us obtain high quality resolution from...

10.1109/cvpr46437.2021.01231 article EN 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2021-06-01

The mosaicking of Unmanned Aerial Vehicle (UAV) imagery usually requires information from additional sensors, such as Global Position System (GPS) and Inertial Measurement Unit (IMU), to facilitate direct orientation, or 3D reconstruction approaches (e.g., structure-from-motion) recover the camera poses. In this paper, we propose a novel method for UAV in which neither nor indirect orientation procedures are required. Inspired by embedded deformation model, widely used non-rigid mesh present...

10.3390/rs8030204 article EN cc-by Remote Sensing 2016-03-02

Abstract. This paper discusses the automatic coregistration and fusion of 3d point clouds generated from aerial image sequences corresponding thermal infrared (TIR) images. Both RGB TIR images have been taken a RPAS platform with predefined flight path where every has same position orientation respect to accuracy system inertial measurement unit. To remove remaining differences in exterior orientation, different strategies for coregistering are discussed: (i) based on 2D line segments single...

10.5194/isprs-archives-xli-b3-241-2016 article EN cc-by ˜The œinternational archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences 2016-06-09

In this paper, a vision-guided mobile robot's path following is studied. A dynamic sliding mode controller (DSMC) designed based on the nonlinear kinematic model of robot without linearization system and stability proved through Lyapunov function. Particle swarm optimization (PSO) algorithm utilized to optimize parameters controller. Simulation analysis experimental results demonstrate that proposed can steer track guide line accurately, which also show eliminates chattering problem...

10.1109/ifita.2009.173 article EN International Forum on Information Technology and Applications 2009-05-01

It is well known that the passive stereo system cannot adapt to weak texture objects, e.g., white walls. However, these targets are very common in indoor environments. In this paper, we present a novel system, which consists of two cameras (an RGB camera and an IR camera) speckle projector. The used both for depth estimation acquisition. projector can form monocular structured-light (MSL) subsystem, while binocular subsystem. map generated by MSL subsystem provide external guidance matching...

10.1109/cvpr52688.2022.00179 article EN 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2022-06-01

In this study, the authors present a novel human body contour tracking method which adaptively combines colour and depth cues of RGB‐D images in level set framework. The model object by active contour. When is far away from background objects, it relatively easy to separate using cue. case, cue should dominate evolution some part very close discriminability decreases rapidly. local region evolution, whereas still play an important role other regions. To achieve these objectives, propose use...

10.1049/iet-cvi.2013.0164 article EN IET Computer Vision 2014-01-15

A novel real-time full-field phase unwrapping framework is proposed for the one-projector and one-camera structured light system. In this framework, only four patterns (including three fringe a binary speckle pattern) are required to measure absolute 3D shape of targets. We use system capture images nearly planar target (e.g. wall), which image taken as reference image, corresponding map computed stored, before measuring. So each pixel in can be mapped an value. way, if we create...

10.1364/oe.26.029588 article EN cc-by Optics Express 2018-10-26

Real-time performance of stereo matching networks is important for many applications, such as automatic driving, robot navigation and augmented reality (AR). Although significant progress has been made in recent years, it still challenging to balance real-time accuracy. In this paper, we present a novel edge-preserving cost volume upsampling module based on the slicing operation learned bilateral grid. The layer parameter-free, which allows us obtain high quality resolution from...

10.48550/arxiv.2101.01601 preprint EN other-oa arXiv (Cornell University) 2021-01-01

The high portability of small Unmanned Aircraft Vehicles (UAVs) makes them play an important role in surveillance and reconnaissance tasks, so the military civilian desires for UAVs are constantly growing. Recently, we have developed a real-time video exploitation system our UAV which is mainly used forest patrol tasks. Our consists six key models, including image contrast enhancement, stabilization, mosaicing, salient target indication, moving display footprint flight path on map. Extensive...

10.1117/12.2034468 article EN Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE 2013-08-21

In this paper we propose a method named spacetime stereo phase-shift for fast 3D shape measurement. It combines the advantages of approach and approach. We first generate two different binary random patterns. each projection cycle, four patterns including three one pattern are projected. adjacent cycles, switched alternatively, which is main feature distinguishing proposed with other approaches. As result, can use five image pairs reconstruction without increasing number Detailedly, (one...

10.1109/yac.2017.7967571 article EN 2017-05-01
Coming Soon ...