A5089102115- Advanced Vision and Imaging
- Robot Manipulation and Learning
- Image Enhancement Techniques
- Robotics and Sensor-Based Localization
- Advanced Image Processing Techniques
- Image and Signal Denoising Methods
- Robotic Path Planning Algorithms
- Advanced Neural Network Applications
- Robotic Mechanisms and Dynamics
- Human Pose and Action Recognition
- Urban Transport and Accessibility
- Land Use and Ecosystem Services
- Water Resources and Sustainability
- Advanced Image Fusion Techniques
- Image Processing Techniques and Applications
- 3D Shape Modeling and Analysis
- Video Surveillance and Tracking Methods
- Image and Object Detection Techniques
- Metaheuristic Optimization Algorithms Research
Hunan University
2021-2024
Central South University
2019
Tracking the 6-degree-of-freedom (6D) object pose in video sequences is gaining attention because it has a wide application multimedia and robotic manipulation. However, current methods often perform poorly challenging scenes, such as incorrect initial pose, sudden re-orientation, severe occlusion. In contrast, we present robust 6D tracking method with novel hierarchical feature fusion network, refer HFF6D, which aims to predict object's relative between adjacent frames. Instead of...
A new improved differential evolution constrained optimization algorithm is proposed to determine the optimum path generation of a rock-drilling manipulator with nine degrees freedom. This developed minimize total joint displacement without compromising pose accuracy end-effector. Considering rule for optimal operation time and smooth motion, minimization end-effector error are respectively taken as objective constraints. In algorithm, inverse kinematics solution computed by self-adaptive...
Monocular 6D object pose estimation aims to estimate 6 degrees of freedom known objects, gaining attention. Correspondence-based methods are the mainstream methods. They analyze geometric information in 2D RGB images and establish 2D-3D correspondences calculate pose. However, accuracy suffers from that can not provide enough information. To solve this problem, We propose a novel prior geometry guided direct regression network (PGDRN), which fully uses knowledge contained given models....
Binocular stereo matching is in an effort to obtain disparity maps from binocular images taken by the same camera. Existing methods resulted mismatches discontinuous regions. In this study, we put forward a innovative correspondence method based on improved census transformation. After that, semiglobal algorithm was performed cost aggregation stage. Finally, further conducted Left and Right Consistency Check median filter generate better disparities. The results indicated that compared with...
Estimating the 6D pose of known objects has attracted a lot research attention since it is important for intelligent robot manipulation and virtual reality. In this paper, in order to improve performance estimation using RGB image, we propose novel object framework based on monocular regression depth. To get depth map use U-Net regress information effectively. For estimation, our approach two steps, which first uses Convolutional Neural Network (CNN) extract feature data regressed data,...
At present, the learning-based stereo matching networks focus more on effective extraction of features and less selection at same time, weak texture areas are still a difficult research point. To tackle problem above, self-attention multi-scale pyramid network (SAMPSM-Net) is proposed. The channel spatial feature fusion module used to effectively select features, strengthens expression channels suppresses obstruction redundant features. Moreover, aggregation improve accuracy for regions....