Gengwei Zhang

ORCID: 0000-0003-1823-502X
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Advanced Neural Network Applications
  • Extraction and Separation Processes
  • Metal Extraction and Bioleaching
  • Domain Adaptation and Few-Shot Learning
  • Electrocatalysts for Energy Conversion
  • Multimodal Machine Learning Applications
  • Adsorption and biosorption for pollutant removal
  • Advanced battery technologies research
  • Advanced Image and Video Retrieval Techniques
  • Visual Attention and Saliency Detection
  • Nanomaterials for catalytic reactions
  • Fuel Cells and Related Materials
  • Minerals Flotation and Separation Techniques
  • Recycling and Waste Management Techniques
  • Electrochemical Analysis and Applications
  • Human Pose and Action Recognition
  • Nonlinear Dynamics and Pattern Formation
  • Chaos control and synchronization
  • Metal-Organic Frameworks: Synthesis and Applications
  • Optical Network Technologies
  • IoT and Edge/Fog Computing
  • Advanced Welding Techniques Analysis
  • Tuberculosis Research and Epidemiology
  • Ocular Oncology and Treatments
  • Anomaly Detection Techniques and Applications

Kunming University of Science and Technology
2015-2025

Shenzhen Third People’s Hospital
2020-2025

Southern University of Science and Technology
2021-2025

University of Technology Sydney
2021-2023

Sun Yat-sen University
2018-2022

Xi'an Jiaotong University
2019-2021

Baidu (China)
2021

China XD Group (China)
2019

First Automotive Works (China)
2019

Equilibrium Research
2019

Abstract Designing the electrocatalysts that are stable and active for extensively adaptable water splitting is highly desirable developing hydrogen based energy. IrO 2 a promising widely used catalyst oxygen evolution reaction in commercial applications, but rarely (HER), due to high Gibbs free energy adsorption (Δ G H* ). Herein, an approach modify electronic structure of via cyclic voltammetry proposed. In this process, Ir(+4) partially reduced trace Pt simultaneously deposited on , which...

10.1002/aenm.202001600 article EN Advanced Energy Materials 2020-06-18

Exploring highly efficient non-noble-metal bifunctional catalysts for water electrolysis is vital the commercial production of hydrogen.

10.1039/c9ta00530g article EN Journal of Materials Chemistry A 2019-01-01

Abstract It is a great challenge to design active and durable oxygen evolution reaction (OER) electrocatalysts for proton exchange membrane (PEM) electrolyzer due the high dissolution of in acidic solution. Herein, Nd‐doped RuO 2 (Nd 0.1 x ) developed enhanced 0.5 m H SO 4 solution with an overpotential 211 mV achieve 10 mA cm −2 . The theoretical calculation reveals that improved activity Nd moderate decrease d‐band center energy, which balances adsorption desorption intermediates....

10.1002/adfm.202213304 article EN Advanced Functional Materials 2023-01-05

Abstract Precisely modulating the Ru-O covalency in RuO x for enhanced stability proton exchange membrane water electrolysis is highly desired. However, transition metals with d -valence electrons, which were doped into or alloyed , are inherently susceptible to influence of coordination environment, making it challenging modulate a precise and continuous manner. Here, we first deduce that introduction lanthanide gradually changing electronic configurations can continuously owing shielding...

10.1038/s41467-024-49281-2 article EN cc-by Nature Communications 2024-06-11

Few-shot segmentation aims to train a model that can fast adapt novel classes with few exemplars. The conventional training paradigm is learn make predictions on query images conditioned the features from support images. Previous methods only utilized semantic-level prototypes of as conditional information. These cannot utilize all pixel-wise information for predictions, which however critical task. In this paper, we focus utilizing relationships between and facilitate few-shot We design...

10.48550/arxiv.2106.02320 preprint EN other-oa arXiv (Cornell University) 2021-01-01

The goal of continual learning is to improve the performance recognition models in sequentially arrived data. Although most existing works are established on premise from scratch, growing efforts have been devoted incorporating benefits pre-training. However, how adaptively exploit pre-trained knowledge for each incremental task while maintaining its generalizability remains an open question. In this work, we present extensive analysis a model (CLPM), and attribute key challenge progressive...

10.1109/iccv51070.2023.01754 article EN 2021 IEEE/CVF International Conference on Computer Vision (ICCV) 2023-10-01

The re-emergence of the mpox pandemic poses considerable challenges to human health and societal development. There is an urgent need for effective prevention treatment strategies against virus (MPXV). In this study, we focused on A35R protein created a chimeric A35R-Fc by fusing Fc region IgG its C-terminal. We then assessed reactivity with A35R-specific antibodies convalescent plasma, as well immunogenicity. Our findings indicate that significantly enhances affinity compared commercially...

10.3390/v17010116 article EN cc-by Viruses 2025-01-16

Recent researches on panoptic segmentation resort to a single end-to-end network combine the tasks of instance and semantic segmentation. However, prior models only unified two related at architectural level via multi-branch scheme or revealed underlying correlation between them by unidirectional feature fusion, which disregards explicit co-occurrence relations among objects background. Inspired fact that context information is critical recognize localize objects, inclusive object details...

10.1109/cvpr42600.2020.00910 article EN 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2020-06-01

Rational design and synthesis of bifunctional electrocatalysts with high efficiency low-cost for overall water splitting is still a challenge. A simple approach reported to prepare phosphorus yttrium codoped cobalt hydroxyfluoride (YP-Co(OH)F) nanoarray on nickel foam, which displays high-performance both hydrogen evolution reaction (HER) oxygen in alkaline solution. The codoping into Co(OH)F leads tuned electronic environment favorable electron transfer, thus resulting superior activity....

10.1002/smll.201904105 article EN Small 2019-08-30

Hydrogen production from the electrolysis of seawater and domestic sewage is more attractive than that pure water, especially in regions where freshwater resources are scarce. However, under such harsh conditions, higher requirements put forward for catalytic activity adaptability a electrode. Herein, we advance an ultrasimple dipping-and-heating method to engineer surface Ni foam (NF) into interface-rich FeNi oxide layer realize exceptional oxygen evolution reaction (OER) performance. It...

10.1021/acsami.1c09274 article EN ACS Applied Materials & Interfaces 2021-07-28

The synergistic effect between Pt and Ru in PtRu/CC<sub>1500</sub> catalyst results a promising catalytic activity toward HER, which displays performance dozens of times higher than that commercial Pt/C.

10.1039/c9ta12300h article EN Journal of Materials Chemistry A 2019-12-13

RuO<sub>2</sub>–NiO/NF, obtained through a dip-heating method, achieved continuous electrolysis for 2000 h at an ultrahigh current density. The RuO<sub>2</sub>–NiO interface promoted H<sub>2</sub>O dissociation and modulated chemisorbed O-containing intermediates, improving HER OER activity.

10.1039/d0ta06565j article EN Journal of Materials Chemistry A 2020-01-01

Continual learning is a challenging real-world problem for constructing mature AI system when data are provided in streaming fashion. Despite recent progress continual classification, the researches of object detection impeded by diverse sizes and numbers objects each image. Different from previous works that tune whole network all tasks, this work, we present simple flexible framework via pRotOtypical taSk corrElaTion guided gaTing mechAnism (ROSETTA). Concretely, unified shared tasks while...

10.1109/cvpr52688.2022.00904 article EN 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2022-06-01

Human parsing, which segments a human-centric image into pixel-wise categorization, has wide range of applications. However, none the existing methods can productively solve issue label parsing fragmentation due to confused and complicated annotations. In this paper, we propose novel Trusted Guidance Pyramid Network (TGPNet) address limitation. Based on pyramid architecture, design Residual Pooling (PRP) module setting at end bottom-up approach capture both global local level context....

10.1145/3240508.3240634 article EN Proceedings of the 30th ACM International Conference on Multimedia 2018-10-15
Coming Soon ...