Wenxuan Liu

Dr. Wenxuan Liu

orcid

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

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

Representative Publications

View All Publications
27 December 2016 in Remote Sensing
Verified
Journal Article
Open access
PDF file
Kunlun QiKunlun Qi
Kunlun Qi
WLWenxuan Liu
Wenxuan Liu
CYChao Yang
Chao Yang
Qingfeng GuanQingfeng Guan
Qingfeng Guan
Huayi WuHuayi Wu
Huayi Wu
Scene classification plays an important role in the intelligent processing of High-Resolution Satellite (HRS) remotely sensed images. In HRS image classification, multiple features, e.g., shape, color, and texture features, are employed to represent scenes from different perspectives. Accordingly, effective integration of multiple features always results in better performance compared to methods based on a single feature in the interpretation of HRS images. In this paper, we introduce a multi-task joint sparse and low-rank representation model to combine the strength of multiple features for HRS image interpretation. Specifically, a multi-task learning formulation is applied to simultaneously consider sparse and low-rank structures across multiple tasks. The proposed model is optimized as a non-smooth convex optimization problem using an accelerated proximal gradient method. Experiments on two public scene classification datasets demonstrate that the proposed method achieves remarkable performance and improves upon the state-of-art methods in respective applications.

15 March 2024 in Engineering Applications of Artificial Intelligence
Verified
Journal Article
Closed access
WLWenxuan Liu
Wenxuan Liu
ZZZizhuo Zhang
Zizhuo Zhang
Bang WangBang Wang
Bang Wang

Verified
Journal Article
Open access
PDF file
QLQing Luo
Qing Luo
Kai HuKai Hu
Kai Hu
WLWenxuan Liu
Wenxuan Liu
Huayi WuHuayi Wu
Huayi Wu
Spatial autocorrelation describes the interdependent relationship between the realizations or observations of a variable that is distributed across a geographical landscape, which may be divided into different units/areas according to natural or political boundaries. Researchers of Geographical Information Science (GIS) always consider spatial autocorrelation. However, spatial autocorrelation research covers a wide range of disciplines, not only GIS, but spatial econometrics, ecology, biology, etc. Since spatial autocorrelation relates to multiple disciplines, it is difficult gain a wide breadth of knowledge on all its applications, which is very important for beginners to start their research as well as for experienced scholars to consider new perspectives in their works. Scientometric analyses are conducted in this paper to achieve this end. Specifically, we employ scientometrc indicators and scientometric network mapping techniques to discover influential journals, countries, institutions, and research communities; key topics and papers; and research development and trends. The conclusions are: (1) journals categorized into ecological and biological domains constitute the majority of TOP journals;(2) northern American countries, European countries, Australia, Brazil, and China contribute the most to spatial autocorrelation-related research; (3) eleven research communities consisting of three geographical communities and eight communities of other domains were detected; (4) hot topics include spatial autocorrelation analysis for molecular data, biodiversity, spatial heterogeneity, and variability, and problems that have emerged in the rapid development of China; and (5) spatial statistics-based approaches and more intensive problem-oriented applications are, and still will be, the trend of spatial autocorrelation-related research. We also refine the results from a geographer’s perspective at the end of this paper.

22 November 2021 in IEEE Access
Verified
Journal Article
Open access
WLWenxuan Liu
Wenxuan Liu
Huayi WuHuayi Wu
Huayi Wu
Kai HuKai Hu
Kai Hu
QLQing Luo
Qing Luo
XCXiaoqiang Cheng
Xiaoqiang Cheng
Image captioning has gradually gained attention in the field of artificial intelligence and become an interesting and challenging task for image understanding. It needs to identify important objects in images, extract attributes, tell relationships, and help the machine generate human-like descriptions. Recent works in deep neural networks have greatly improved the performance of image caption models. However, machines are still unable to imitate the way humans think, talk and communicate, so image captioning remains an ongoing task. It is thus very important to keep up with the latest research and results in the field of image captioning whereas publications on this topic are numerous. Our work aims to help researchers to have a macro-level understanding of image captioning from four aspects: spatial-temporal distribution characteristics, collaborative networks, trends in subject research, and historical evolutionary path. We employ scientometric visualization methods to achieve this goal. The results show that China has published the largest amount of publications in image captioning, but the United States has the greatest impact on research in this area. Besides, thirteen academic groups are identified in the field of image description, with institutions such as Microsoft, Google, Australian National University, and Georgia Institute of Technology being the most prominent research institutions. Meanwhile, we find that evaluation methods, datasets, novel image captioning models based on generative adversarial networks, reinforcement learning, and Transformer, as well as remote sensing image captioning, are the new research trends. Lastly, we conclude that image captioning research has gone through three major development stages from 2010 to 2020, and on this basis, we propose a more comprehensive taxonomy of image captioning.

7 November 2024 in Remote Sensing
Verified
Journal Article
Open access
PDF file
BLBeining Liu
Beining Liu
WLWenxuan Liu
Wenxuan Liu
ZLZhen Lei
Zhen Lei
FZFan Zhang
Fan Zhang
XHXianfeng Huang
Xianfeng Huang
TATarek M. Awwad
Tarek M. Awwad
Oblique photogrammetry-based 3D modeling is widely used for large-scale urban reconstruction. However, textures generated with photogrammetric techniques often exhibit scattered and irregular characteristics, which lead to significant challenges with texture seams and UV map discontinuities, increasing storage requirements and affecting rendering quality. In this paper, we propose a planar feature-preserving texture defragmentation method designed specifically for urban building models. Our approach leverages the multi-planar topology of buildings to optimize texture merging and reduce fragmentation. The proposed approach is composed of three main stages: the extraction of planar features from texture fragments; the employment of these planar elements as guiding constraints for merging adjacent texture fragments within a two-dimensional texture space; and an enhanced texture-packing algorithm designed for more regular texture charts to systematically generate refined texture atlas. Experiments on various urban building models demonstrate that our method significantly improves texture continuity and storage efficiency compared to traditional approaches, with both quantitative and qualitative validation.

7 November 2016 in Preprints.org
Verified
Preprint
Open access
PDF file
Kunlun QiKunlun Qi
Kunlun Qi
WLWenxuan Liu
Wenxuan Liu
CYChao Yang
Chao Yang
Qingfeng GuanQingfeng Guan
Qingfeng Guan
Huayi WuHuayi Wu
Huayi Wu
Scene classification plays an important role in the intelligent processing of high-resolution satellite (HRS) remotely sensed image. In HRS image classification, multiple features, e.g. shape, color, and texture features, are employed to represent scenes from different perspectives. Accordingly, effective integration of multiple features always results in better performance compared to methods based on a single feature in the interpretation of HRS image. In this paper, we introduce a multi-task joint sparse and low-rank representation model to combine the strength of multiple features for HRS image interpretation. Specifically, a multi-task learning formulation is applied to simultaneously consider sparse and low-rank structure across multiple tasks. The proposed model is optimized as a non-smooth convex optimization problem using an accelerated proximal gradient method. Experiments on two public scene classification datasets demonstrate that the proposed method achieves remarkable performance and improves upon the state-of-art methods in respective applications.