A5100694198- Computer Graphics and Visualization Techniques
- Advanced Vision and Imaging
- 3D Shape Modeling and Analysis
- Video Surveillance and Tracking Methods
- Remote Sensing and LiDAR Applications
- Optical measurement and interference techniques
- Robotics and Sensor-Based Localization
- Modular Robots and Swarm Intelligence
- Domain Adaptation and Few-Shot Learning
- Advanced Image and Video Retrieval Techniques
- Software-Defined Networks and 5G
- COVID-19 diagnosis using AI
- Advanced Numerical Analysis Techniques
- Antenna Design and Optimization
- Biofuel production and bioconversion
- Evaluation Methods in Various Fields
- Image Processing and 3D Reconstruction
- Advanced Measurement and Detection Methods
- Image Retrieval and Classification Techniques
- Landslides and related hazards
- Corrosion Behavior and Inhibition
- Obstructive Sleep Apnea Research
- Human Pose and Action Recognition
- Advanced Antenna and Metasurface Technologies
- Antenna Design and Analysis
Northwest A&F University
2009-2024
Chongqing University of Posts and Telecommunications
2021-2024
Durham University
2024
Shandong Agricultural University
2024
Shanghai Jiao Tong University
2024
Shenzhen Second People's Hospital
2024
Peking University
2023-2024
China University of Geosciences
2020-2023
Sichuan University of Arts and Science
2023
Northwest Institute of Mechanical and Electrical Engineering
2023
Acquiring realistic texture details for 3D models is important in reconstruction. However, the existence of geometric errors, caused by noisy RGB-D sensor data, always makes color images cannot be accurately aligned onto reconstructed models. In this paper, we propose a global-to-local correction strategy to obtain more desired mapping results. Our algorithm first adaptively selects an optimal image each face model, which can effectively remove blurring and ghost artifacts produced multiple...
Pollen plays a vital role in plant reproduction, acting as carrier of male genetic material for fertilization and ensuring species propagation the maintenance biodiversity. Ultra-low temperature preservation pollen provides reliable method long-term storage while preserving its viability, thereby facilitating crop breeding, resource conservation, ecological restoration. This study aimed to establish pepper under ultra-low conditions. was collected from unopened flowers during peak flowering...
Targeting in-memory one-dimensional search keys, we propose a novel DIstribution-driven Learned Index tree ( DILI ), where concise and computation-efficient linear regression model is used for each node. An internal node's key range equally divided by its child nodes such that enjoys perfect prediction accuracy to find the relevant leaf A node uses machine learning models generate searchable data layout thus accurately predicts record position key. To construct DILI, first build bottom-up...
In this article, a millimeter-wave LHCP <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$1\times32$ </tex-math></inline-formula> with dummy on both sides phased array antenna is proposed. Multilayer structure and single feed are employed. The patch adopts notation="LaTeX">$3\times $ 3 metasurface. measured impedance bandwidth 20%, ranging from 22.5 to 27.5 GHz. axial ratio (AR) an AR level < dB achieves...
A perennial problem in structure from motion (SfM) is visual ambiguity posed by repetitive structures. Recent disambiguating algorithms infer ambiguities mainly via explicit background context, thus face limitations highly ambiguous scenes which are visually indistinguishable. Instead of analyzing local information, we propose a novel algorithm for SfM disambiguation that explores the global topology as encoded photo collections. An important adaptation this work to approximate available...
Handheld scanning using commodity depth cameras provides a flexible and low-cost manner to get 3D models. The existing methods scan target by densely fusing all the captured images, yet most frames are redundant. jittering inevitably embedded in handheld process will cause feature blurring on reconstructed model even trigger failure (i.e., camera tracking losing). To address these problems, this paper, we propose novel sparse-sequence fusion (SSF) algorithm for cameras. It first extracts...
The quality of fermented plant leaves is closely related to the interleaf microorganisms and their metabolic activities. In this experiment, a multi-omics analysis was applied investigate link between structural composition phyllosphere microbial community main metabolites during fermentation process. It found that whole process cigar could be divided into three stages, in which Mid-Stage most active period activities occupied an important position.
Classifying tree species from point clouds acquired by light detection and ranging (LiDAR) scanning systems is important in many applications, including remote sensing, virtual reality, forestry inventory. Compared with terrestrial laser systems, airborne (ALS) can acquire large-scale only a single scan. However, ALS have the disadvantages of low density, uneven distribution, unclear branch structure, making classification challenging task. Recently, deep learning-based approaches, such as...
Multi-lane detection is one of the key technologies for advanced intelligent assisted driving. In this paper, we propose a multi-lane algorithm based on constraints lane information. Unlike methods detect multiple lanes directly, proposed method detects ego-lane firstly and then side lane, which enables lanes. Specifically, in paper accomplishes three steps. Firstly, results historical frames are used to extract ego-lane's region interest(ROI), then, Hough transform sliding window search...
Virtual network embedding (VNE) is an essential resource allocation task in virtualization, aiming to map virtual requests (VNRs) onto physical infrastructure. Reinforcement learning (RL) has recently emerged as a promising solution this problem. However, existing RL-based VNE methods are limited by the unidirectional action design and one-size-fits-all training strategy, resulting restricted searchability generalizability. In paper, we propose FLexible And Generalizable RL framework for...
Virtual network embedding (VNE) is an essential resource allocation task in virtualization, aiming to map virtual requests (VNRs) onto physical infrastructure. Reinforcement learning (RL) has recently emerged as a promising solution this problem. However, existing RL-based VNE methods are limited by the unidirectional action design and one-size-fits-all training strategy, resulting restricted searchability generalizability. In paper, we propose flexible generalizable RL framework for VNE,...
The reactivation mechanism of old landslides has been extensively studied from building load, erosion the slope toe, heavy rainfall, and cutting for existing research. However, previous research on pays little attention to combined action engineering disturbance rainfall is rarely studied. This paper describes an landslide in Wushan County, Chongqing, China, that was reactivated August 2019 due rainfall. deformation first observed 2007 2008, resulting excavation respectively, remained stable...
Current image processing methods usually operate on the finest-granularity unit; that is, pixel, which leads to challenges in terms of efficiency, robustness, and understandability deep learning models. We present an improved granular-ball computing method represent as a graph, each node expresses structural block edge represents association between two nodes. Specifically:(1) design gradient-based strategy for adaptive reorganization all pixels into numerous rectangular regions, can be...