Skip to main content
Document Analysis and Pattern Recognition Laboratory

Document Analysis and Pattern Recognition Laboratory

Virtue·Conduct·Erudition·Competence

中文
  • Home
  • Research
  • People
  • Publications
  • News
  • Positions
  • Contact
Home > People > Haotian Liu

Haotian Liu

📞 18764387993 📧 [email protected] Github ORCID
Haotian Liu
supervisor: Yuchen Zheng

2026-09-01 — 2029-07-01

👤 Personal Information

  • Class of 2026, Master candidate in Electronic Information
  • Document Analysis and Pattern Recognition Laboratory, School of Information Science and Technology, Shihezi University

🔬 Research Interests

  • Remote‑sensing Change Detection Based on Large‑scale Model and World Model

This work focuses on the problems of insufficient accuracy, poor generalization and weak scene adaptability in remote‑sensing image change detection. Traditional remote‑sensing change detection methods rely on manually‑designed features and shallow model architectures, which can hardly capture subtle changes, spatio‑temporal correlation features and global semantic information of complex land‑surface scenes. Therefore, a brand‑new remote‑sensing change detection framework is constructed relying on the powerful semantic comprehension capability of large‑scale models as well as the scene‑modeling and spatio‑temporal‑reasoning capability of world models. Specifically, large‑scale models are adopted to extract deep global semantic features from remote‑sensing images. Combined with world models to model dynamic evolution rules of geographic scenes, fine‑grained features and global features from bi‑temporal images are fused to realize pixel‑level and scene‑level identification and classification of changed regions. It effectively improves the accuracy and robustness of remote‑sensing change detection under complex landforms, heavy cloud‑fog interference and mixed ground‑object scenarios.

Keywords: remote‑sensing image, change detection, large‑scale model, world model, spatio‑temporal feature modeling

📄 Academic Outputs

None

🏆 Honors & Awards

None

Contact

Lvyuan Building No.2, College of Information Science and Technology, Shihezi University, Shihezi 832003, China

Email: [email protected]

Phone: 15610007856

Links

  • 石河子大学
  • 模板来源:SynELab(上科大纪清清团队)
  • 石河子大学信息科学与技术学院
  • 上科大物质科学与技术学院合成电子学实验室
  • 报考石大
Visits:

© 2021–2026 Document Analysis and Pattern Recognition Laboratory, Shihezi University

苏公网安备32072102010392号 苏ICP备2025166034号-1