Guowen Kuang

Dr. Guowen Kuang

orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
23 December 2022
Verified
Conference Paper
Open access
YLYaqiao Liao
Yaqiao Liao
XWXu Wang
Xu Wang
GKGuowen Kuang
Guowen Kuang
ZLZhigang LI
Zhigang LI
TLTingting Li
Tingting Li
Person re-identification has great potential in applications of public security. However, non-discriminative information like background noise still hinders further improvement of current person re-identification algorithms. This paper proposes an information bottleneck-enhanced video-based person re-identification algorithm concentrating on discriminative information. Specifically, the spatial-temporal information purification (STIP) module is designed to filter out irrelevant information and the variance of information is estimated to weigh the importance of each feature dimension. Our method is evaluated on MARS and results show that it pays more attention to discriminative information and suppresses the irrelevant feature.

21 November 2024 in Nanomaterials
Verified
Journal Article
Open access
PDF file
XWXiaobo Wang
Xiaobo Wang
GKGuowen Kuang
Guowen Kuang
HTHongmei Tian
Hongmei Tian
ZSZhibin Shao
Zhibin Shao
NDNing Dong
Ning Dong
LHLi UMAR Huang
Li UMAR Huang
Carbon nanorings (CNRs) serve as an ideal quantum system for novel electronic and magnetic properties. Although extensive theoretical studies utilizing molecular dynamics (MD) simulations have investigated the formation and structural characteristics of CNRs, systematically analyzing their properties across various toric sizes remains challenging due to the inherent complexity of this system. In this study, we introduce a novel finite element method, the Chebyshev–Ritz method, as an alternative approach to investigating the structural properties of CNRs. Previous MD simulations demonstrated that stable CNRs adopt a regular buckled shape at specific toric sizes. By meticulously selecting mechanical parameters, we observe that the critical deformation of a CNR with 50 repeated units, as determined by the Chebyshev–Ritz method, aligns with an MD simulation presenting a buckling number of 14. Additionally, the implementation of the Chebyshev–Ritz method with a constant mechanical parameter for 50 repeated units reveals a structural transition at varying toric sizes, leading to the stabilization of buckling numbers 13, 14, and 15. This structural transition across different buckling modes has also been corroborated by MD simulations. Our approach offers a reliable and accurate means of examining the structural properties of large-scale nanomaterials and paves the way for further exploration in nanoscale mechanics.

1 April 2025 in CAAI Artificial Intelligence Research
Verified
Journal Article
Open access
PDF file
XWXu Wang
Xu Wang
YLYaqiao Liao
Yaqiao Liao
GKGuowen Kuang
Guowen Kuang
JYJinfeng Yang
Jinfeng Yang
Person re-identification (ReID) is a desirable yet challenging issue in computer vision, which has numerous potential applications in the public security area. In this paper, the information bottleneck (IB) theory is introduced to filter out the undiscriminating information and noise, and keep the distinctive information for persons, to identify the assigned person from the candidates. To meet the requirement of the IB theory, a dual-stream IB-enhanced (DSIB) ReID framework is proposed to make unentangled estimations of the expectation and variance of the feature map. Moreover, the unconditional feature distribution is relaxed from standard normal distribution to Gaussian distribution, to enrich the information contained in each feature channel. Comprehensive experiments are conducted with several baseline models on datasets MARS, LS-VID, iLiDS-VID, and PRID-2011. The results reveal that our proposed DSIB framework has improved the performance of all baseline models, and it outperforms most state-of-the-art methods in terms of mean average precision (mAP) and Rank-1 precision.