A5101650221- Advanced Numerical Analysis Techniques
- Manufacturing Process and Optimization
- 3D Shape Modeling and Analysis
- Computational Geometry and Mesh Generation
- Computer Graphics and Visualization Techniques
- Advanced Vision and Imaging
- Membrane Separation Technologies
- Advanced Measurement and Metrology Techniques
- Topology Optimization in Engineering
- Finite Group Theory Research
- Video Surveillance and Tracking Methods
- Advanced Neural Network Applications
- Innovations in Concrete and Construction Materials
- Aerosol Filtration and Electrostatic Precipitation
- Membrane Separation and Gas Transport
- Radiomics and Machine Learning in Medical Imaging
- Image Processing and 3D Reconstruction
- Image Retrieval and Classification Techniques
- Optical measurement and interference techniques
- Lattice Boltzmann Simulation Studies
- Advanced Image Processing Techniques
- Advanced Image and Video Retrieval Techniques
- Image and Object Detection Techniques
- Coding theory and cryptography
- Color perception and design
Dalian University of Technology
2015-2024
Dezhou University
2023
Nantong University
2023
Dalian University
2009-2019
Dalian Maritime University
2018
Guangdong University of Technology
2016-2018
University of Western Macedonia
2017
Chengdu University of Information Technology
2015
Northwestern Polytechnical University
2011
Xi'an University of Science and Technology
2010
In this approach, we present an efficient topology and geometry optimization of triply periodic minimal surfaces (TPMS) based porous shell structures, which can be represented, analyzed, optimized stored directly using functions. The proposed framework is executed on functions instead remeshing (tetrahedral/hexahedral), substantially improves the controllability efficiency. Specifically, a valid TPMS-based structure first constructed by function expressions. permits continuous smooth changes...
This paper introduces a novel hierarchical structured representation for leaf modeling and proposes corresponding multi-resolution remeshing method large-scale visual computation. Leaf is very difficult challenging problem due to the wide variations in shape structures among different species of plants. Firstly, we introduce Hierarchical Parametric Veins Margin (HPVM) approach, which describes biological exact geometry via interpolation parametric curves from extracted vein features...
In this paper, we proposed a Highly Coupled Network (HCNet) for joint objection detection and semantic segmentation. It follows that our method is faster performs better than the previous approaches whose decoder networks of different tasks are independent. Besides, present multi-scale loss architecture to learn representation scale objects, but without extra time in inference phase. Experiment results show achieves state-of-the-art on KITTI datasets. Moreover, it can run at 35 FPS GPU thus...
Purpose The strength of printed parts by application fused deposition modeling (FDM) has been broadly studied through experimental methods. However, constitutive behaviors the in theory are still unclear. Therefore, this paper aims to focus on building an elasto-plastic model reveal behavior. Design/methodology/approach An that considers anisotropic characteristics is proposed. Tensile tests performed for parameter identification using different samples with varying printing angles. Finally,...
We propose a robust method for detecting features on triangular meshes by combining normal tensor voting with neighbor supporting. Our contains two stages: feature detection and refinement. First, the is modified to detect initial features, which may include some pseudo features. Then, at refinement stage, novel salient measure deriving from idea of supporting developed. Benefiting integrated reliable feature, can be effectively discriminated initially detected removed. Compared previous...
In this paper, a novel hole-filling algorithm for triangular meshes is proposed. Firstly, the hole triangulated into set of new triangles using modified principle minimum angle. Then initial patching mesh refined according to density vertices on boundary edges. Finally, optimized via bilateral filter recover missed features. Experimental results demonstrate that proposed fills complex holes robustly, and preserves geometric features certain extent as well. The resulted are good quality engineering.
To avoid the requirement of expert knowledge in conventional methods for car styling analysis, this article proposes a machine learning–based method which requires no expert-engineered features frontal analysis. In article, we aim to identify group behaviors such as degree brand consistency among different automakers and patterns. The is considered behavior formulated classification problem. This problem then solved by learning based on PCANet automatic feature encoding support vector...
This paper discusses the results for second edition of Monocular Depth Estimation Challenge (MDEC). was open to methods using any form supervision, including fully-supervised, self-supervised, multi-task or proxy depth. The challenge based around SYNS-Patches dataset, which features a wide diversity environments with high-quality dense ground-truth. includes complex natural environments, e.g. forests fields, are greatly underrepresented in current benchmarks.The received eight unique...
We describe a novel method to build 3D statistical shape models for anatomic objects in tomographic images, and demonstrate the use of model guide image segmentation. Our consists two main steps. In first step, is built collection training images. Boundary similarities between adjacent transverse slices are matched inter-slice interpolation. Slice-by-slice correspondences established images sets by matching mean boundary curvatures. A shep then obtained principal components analysis. During...