👋 Hi there

I am currently a Postdoctoral Fellow in the Department of Land Surveying and Geo-Informatics (LSGI) at The Hong Kong Polytechnic University (PolyU), affiliated with the JC STEM Lab of Earth Observations and the Research Centre for Artificial Intelligence in Geomatics (Lab Head: Prof. Qihao Weng). I received my Ph.D. and M.E. degrees from Wuhan University in 2022 and 2019, respectively, under the supervision of Prof. Yanfei Zhong and Prof. Liangpei Zhang.

My research started in the interdisciplinary field of remote sensing science, medical science, and artificial intelligence, focusing on intelligent interpretation of linear objects (eg: road extraction, endoscopic extraction of common bile duct stones). Currently, my research focus is on multimodal remote sensing data fusion, analysis, and application, serving disaster emergency response and sustainable development.

Email: xiaoyan07.lu@polyu.edu.hk; luxiaoyan@whu.edu.cn

🎉 News

📝 Publications

(*Corresponding Author)

TGRS 2024
sym

Multi-LoRA Fine-Tuned Segment Anything Model for Urban Man-Made Object Extraction

Xiaoyan Lu, Qihao Weng*

Code and data

IEEE Transactions on Geoscience and Remote Sensing (TGRS), SCI Q1 TOP, 2024

Highlight: Fine-tuning the SAM foundation model, Multi-LoRA alleviates overfitting; Supervised and unsupervised fine-tuning strategies

GSIS 2024
sym

Global road extraction using a pseudo-label guided framework: from benchmark dataset to cross-region semi-supervised learning

Xiaoyan Lu, Yanfei Zhong*, Zhuo Zheng, Junjue Wang, Dingyuan Chen, Yu Su

Code and data

Geo-spatial Information Science (GSIS), SCI Q1, 2024

Highlight: Global-scale road benchmark dataset: GRSet and GSRV; Semi-supervised learning adapts the model from GRSet to new imagery across diverse geographical regions worldwide.

TIP 2022
sym

Open-Source Data-Driven Cross-Domain Road Detection From Very High Resolution Remote Sensing Imagery

Xiaoyan Lu, Yanfei Zhong*, Liangpei Zhang

IEEE Transactions on Image Processing (TIP), SCI Q1 TOP, CCF-A, 2022

Highlight: Utilizing OSM road centerlines to generate the supervised information of the target domain; Domain-specific representation framework designed to learn domain-invariant structure features and domain-specific texture information.

ISPRS 2021
sym

Cross-domain road detection based on global-local adversarial learning framework from very high resolution satellite imagery

Xiaoyan Lu, Yanfei Zhong*, Zhuo Zheng, Junjue Wang

Data

ISPRS Journal of Photogrammetry and Remote Sensing (ISPRS), SCI Q1 TOP, 2021

Highlight: Unsupervised domain adaptation requires only target domain images; Global-local adversarial learning adaptively reweights the adversarial loss according to the recognition difficulty of each pixel.

  • Xiaoyan Lu, Yanfei Zhong*, Zhuo Zheng, Dingyuan Chen, Yu Su, Ailong Ma, Liangpei Zhang. Cascaded multi-task road extraction network for road surface, centerline, and edge extraction. IEEE Transactions on Geoscience and Remote Sensing (TGRS), SCI Q1 TOP, 2022. paper
  • Xiaoyan Lu, Yanfei Zhong*, Zhuo Zheng, Liangpei Zhang. GAMSNet: Globally aware road detection network with multi-scale residual learning. ISPRS Journal of Photogrammetry and Remote Sensing (ISPRS). SCI Q1 TOP, 2021. paper
  • Li Huang, Xiaoyan Lu (co-first author), Xu Huang, Xiaoping Zou, Lianlian Wu, Zhongyin Zhou, Deqing Wu, Dehua Tang, Dingyuan Chen, Xinyue Wan, Zhongchao Zhu, Tao Deng, Lei Shen, Jun Liu, Yijie Zhu, Dexin Gong, Di Chen, Yanfei Zhong, Feng Liu, Honggang Yu*. Intelligent difficulty scoring and assistance system for endoscopic extraction of common bile duct stones based on deep learning: multicenter study. Endoscopy, SCI Q1 TOP, 2021. paper, video
  • Xiaoyan Lu, Yanfei Zhong*, Zhuo Zheng, Ji Zhao, Liangpei Zhang. Edge-reinforced convolutional neural network for road detection in very-high-resolution remote sensing imagery. Photogrammetric Engineering & Remote Sensing (PE&RS), SCI, 2020. paper, John I. Davidson President’ s Award
  • Xiaoyan Lu, Yanfei Zhong*, Zhuo Zheng, Yanfei Liu, Ji Zhao, Ailong Ma, Jie Yang. Multi-scale and multi-task deep learning framework for automatic road extraction. IEEE Transactions on Geoscience and Remote Sensing (TGRS), SCI Q1 TOP, 2019. paper

🎖 Selected Honors and Awards

📖 Educations

🙆‍♀️ Services

  • Journal Reviewer: IEEE Transactions on Image Processing (TIP); ISPRS Journal of Photogrammetry and Remote Sensing (ISPRS); IEEE Transactions on Geoscience and Remote Sensing (TGRS); International Journal of Digital Earth (IJDE); IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (JSTARS); European Journal of Remote Sensing (EuJRS); Remote Sensing; Conference on Neural Information Processing Systems (NeurIPS);
  • Session Chair (Infrastructure Detection) of IGARSS 2021.
  • Guest Editor, Remote Sensing.