A5100458860- 3D Shape Modeling and Analysis
- Computer Graphics and Visualization Techniques
- 3D Surveying and Cultural Heritage
- Advanced Vision and Imaging
- Multilevel Inverters and Converters
- Advanced DC-DC Converters
- Remote Sensing and LiDAR Applications
- Advanced machining processes and optimization
- Microgrid Control and Optimization
- Advanced Numerical Analysis Techniques
- Advanced Surface Polishing Techniques
- Coal and Its By-products
- Manufacturing Process and Optimization
- Particle Detector Development and Performance
- Recommender Systems and Techniques
- Image Processing and 3D Reconstruction
- Advanced Optical Imaging Technologies
- Thermochemical Biomass Conversion Processes
- Sparse and Compressive Sensing Techniques
- Advanced Data Compression Techniques
- Recycling and utilization of industrial and municipal waste in materials production
- Energy Load and Power Forecasting
- Optical Coherence Tomography Applications
- Solar Radiation and Photovoltaics
- Photoacoustic and Ultrasonic Imaging
Yantai University
2023-2025
Shandong University
2018-2024
Chengdu Jincheng College
2023
Chengdu Medical College
2023
Purple Mountain Observatory
2023
Chinese Academy of Sciences
2023
Xi'an Jiaotong University
2022
Georgia Institute of Technology
2020
We propose a generative adversarial network for point cloud upsampling, which can not only make the upsampled points evenly distributed on underlying surface but also efficiently generate clean high frequency regions. The generator of our includes dynamic graph hierarchical residual aggregation unit and feature extraction respectively. former extracts multiscale point-wise descriptive features, while latter captures rich details with residuals. To neat edges, discriminator uses filter to...
In recent years, point clouds have become increasingly popular for representing three-dimensional (3D) visual objects and scenes. To efficiently store transmit clouds, compression methods been developed, but they often result in a degradation of quality. reduce color distortion we propose graph-based quality enhancement network (GQE-Net) that uses geometry information as an auxiliary input graph convolution blocks to extract local features efficiently. Specifically, use parallel-serial...
Point cloud based 3D visual representation is becoming popular due to its ability exhibit the real world in a more comprehensive and immersive way. However, under limited network bandwidth, it very challenging communicate this kind of media huge data volume. Therefore, MPEG have launched standardization for point compression (PCC), proposed three model categories, i.e., TMC1, TMC2, TMC3. Because geometry methods TMC1 TMC3 are similar, further merged into new platform namely TMC13. In paper,...
Recently, 3D point cloud is becoming popular due to its capability represent the real world for advanced content modality in modern communication systems. In view of wide applications, especially immersive towards human perception, quality metrics clouds are essential. Existing evaluations rely on a full or certain portion original cloud, which severely limits their applications. To overcome this problem, we propose novel deep learning-based no reference assessment method, namely PQA-Net....
The emergence of 3D point clouds (3DPCs) is promoting the rapid development immersive communication, autonomous driving, and so on. Due to huge data volume, compression 3DPCs becoming more attractive. We propose a novel efficient color attribute method for static 3DPCs. First, 3DPC partitioned into several sub-point by distribution analysis. Each cloud then decomposed lot blocks an improved k-d tree-based decomposition algorithm. Afterwards, virtual adaptive sampling-based sparse...
In recent years, the abundance of information in 3D data has made semantic segmentation point clouds a topic great interest. However, current methods often rely solely on original three-dimensional coordinates cloud as input geometric features, leading to poor generalization performance. Additionally, occlusion can negatively impact accuracy when only local is considered. To address these issues, this paper proposes network named LGFF-Net. fully utilize clouds, we designed Local Feature...
We present PU-Refiner, a generative adversarial network for point cloud upsampling. The generator of our includes coarse feature expansion module to create upsampled features, geometry generation regress from the and progressive refinement restore dense in coarse-to-fine fashion based on cloud. discriminator helps produce clouds closer target distribution. It makes full use multi-level features improve its classification performance. Extensive experimental results show that PU-Refiner is...
We present PU-Mask, a virtual mask-based network for 3D point cloud upsampling. Unlike existing upsampling methods, which treat as an "unconstrained generative" problem, we propose to address it from the perspective of "local filling", i.e., assume that sparse input (i.e., unmasked set) is obtained by locally masking original dense with masks. Therefore, given set and masks, our goal fill hidden Specifically, because masks do not actually exist, first locate form each mask generation module....
Rate-distortion optimal 3D point cloud compression is very challenging due to the irregular structure of clouds. For a popular codec that uses octrees for geometry and JPEG color compression, we first find analytical models describe relationship between encoding parameters bitrate distortion, respectively. We then use our formulate rate-distortion optimization problem as constrained convex apply an interior method solve it. Experimental results six clouds show technique gives similar...
The DArk Matter Particle Explorer (DAMPE) is a high-energy cosmic ray and gamma-ray detector located in space. Over period of seven years since its launch on December 17, 2015, DAMPE has surveyed the entire sky collected an extensive dataset more than 300,000 photons with energies above 2 GeV. To analyze data obtained by DAMPE, instrument response functions (IRFs) have been derived, specialized software called DmpST developed. In this context, we present results point-like source catalog....
This article presents a reduced switch hybrid multilevel converter (RSHMC). The proposed RSHMC is attractive for medium voltage industrial applications as it can work in wide range of voltages, and lower count compared with other topologies. Nevertheless, the faces challenges neutral-point (NP) voltage, flying capacitor (FC) balancing input current distortion mitigation. Therefore, four-layer coordinated control (FLCC) method presented this article. Firstly, relationship between vector...
3D object detection is a crucial and complex undertaking in the realm of scene comprehension. Monocular-based detectors, comparison to LiDAR detectors that utilize point clouds as input, often exhibit significant performance gap. Incorporating guidance from LiDAR-based has led notable advancements monocular detection. Nevertheless, some current approaches focus solely on transferring feature or response knowledge data, neglecting valuable geometric structural information present data. As...
We consider constructing a surface from given set of point cloud data. explore two fast algorithms to minimize the weighted minimum energy in [Zhao, Osher, Merriman and Kang, Comp.Vision Image Under., 80(3):295-319, 2000]. An approach using Semi-Implicit Method (SIM) improves computational efficiency through relaxation on time-step constraint. based Augmented Lagrangian (ALM) reduces run-time via an Alternating Direction Multipliers-type algorithm, where each sub-problem is solved...
Point cloud sampling can reduce storage requirements and computation costs for various vision tasks. Traditional methods, such as farthest point sampling, are not geared towards downstream tasks may fail on In this paper, we propose a cascade attention-based network (CAS-Net), which is end-to-end trainable. Specifically, an module (ASM) to capture the semantic features preserve geometry of original cloud. Experimental results ModelNet40 dataset show that CAS-Net outperforms state-of-the-art...
In photovoltaic applications, the zero-magnitude common mode voltage (CMV) is required to suppress leakage current for hybrid seven-level converter cascaded by three-level T-type and H-bridge (T2C-HB). However, ripples are enlarged due abandoning of no-zero magnitude CMV vectors. Moreover, neutral-point (NP) flying capacitor balance highly coupled with reduction. To overcome above issues, an improved model predictive control (IMPC) method proposed. Firstly, restrict into zero, only 37...
Abstract Assembly interference matrix is one of foundation information model for assembly process planning like sequence and path planning, which supports the digital simulation, intelligent assembly, twins based etc. Assemble represents each two parts are collision or not when they moving along specific directions. Traditional construction adopts geometric detections on approximate swept volume part composed by step step. This manner only leads to heavy computational loading since small but...