👋 About Me

I am a second-year Ph.D. student in the LEVIR lab. at Beihang University, advised by Prof. Zhenwei Shi (史振威,国家杰青) and Prof. Zhengxia Zou (邹征夏,国家优青海外). Before that, I received my B.S. and M.S. degree in Image Processing Center at Beihang University.

My research interests lie in the deep learning, computer vision, remote sensing image processing, and multimodal.

🔥 News

  • 2024.01: Two papers are accepted by TGRS (IF=8.2) in 2024.
  • 2023.12: My google scholar citations have exceeded 2000! 🎉🎉
  • 2023.12: Two papers are accepted by TGRS (IF=8.2) in 2023.
  • 2023.10: A paper is accepted by TCSVT (IF=8.4).
  • 2023.02: A paper is accepted by CVPR.
  • 2023.01: My google scholar citations have exceeded 1000! 🎉🎉
  • 2023.01: Our survey on object detection is accepted by Proceedings of the IEEE (IF=20.6).
  • 2022.12: Three papers are accepted by TGRS (IF=8.2) in 2022.

📝 Publications

My full paper list can be found at .

sym

RSMamba: Remote Sensing Image Classification with State Space Model
GRSL, 2024
Keyan Chen, Bowen Chen, Chenyang Liu, Wenyuan Li, Zhengxia Zou, and Zhenwei Shi
[Arxiv] [Github]

We introduce RSMamba, a novel architecture for remote sensing image classification. RSMamba is based on the State Space Model (SSM) and incorporates an efficient, hardware-aware design known as the Mamba. To overcome the limitation of the vanilla Mamba, which can only model causal sequences and is not adaptable to two-dimensional image data, we propose a dynamic multi-path activation mechanism to augment Mamba's capacity to model non-causal data.
sym

Time Travelling Pixels: Bitemporal Features Integration with Foundation Model for Remote Sensing Image Change Detection
IGARSS, 2024
Keyan Chen, Chenyang Liu, Wenyuan Li, Zili Liu, Hao Chen, Haotian Zhang, Zhengxia Zou, and Zhenwei Shi
[Arxiv] [Github] [Page] [Demo]

We integrate the latent knowledge of the SAM foundation model into change detection, effectively addressing the domain shift in general knowledge transfer and the challenge of expressing homogeneous and heterogeneous characteristics of multi-temporal images.
sym

RSPrompter: Learning to Prompt for Remote Sensing Instance Segmentation based on Visual Foundation Model
IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2024
Keyan Chen, Chenyang Liu, Hao Chen, Haotian Zhang, Wenyuan Li, Zhengxia Zou, and Zhenwei Shi
[Arxiv] [Github] [Page] [Demo]

We consider designing an automated instance segmentation approach for remote sensing images based on the SAM foundation model, incorporating semantic category information with prompt learning.
sym

Continuous Remote Sensing Image Super-Resolution based on Context Interaction in Implicit Function Space
IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2023
Keyan Chen, Wenyuan Li, Sen Lei, Jianqi Chen, Xiaolong Jiang, Zhengxia Zou, and Zhenwei Shi
[Arxiv] [Github] [Page] [Demo]

We propose a new super-resolution framework based on context interaction in implicit function space for learning continuous representations of remote sensing images, called FunSR, which consists of three main components: a functional representor, a functional interactor, and a functional parser.
sym

Object Detection in 20 Years: A Survey
Proceedings of the IEEE (P IEEE), 2023
Zhengxia Zou, Keyan Chen, Zhenwei Shi, Yuhong Guo and Jieping Ye
🏆️ ESI Highly Cited Paper & ESI Hot Paper
[PDF] [Github] [Page]

This paper extensively reviews the fast-moving research field in the light of technical evolution, spanning over a quarter-century's time (from the 1990s to 2022). A number of topics have been covered, including the milestone detectors in history, detection datasets, metrics, fundamental building blocks of the detection system, speed-up techniques, and the recent state-of-the-art detection methods.
sym

Resolution-agnostic Remote Sensing Scene Classification with Implicit Neural Representations
IEEE Geoscience and Remote Sensing Letters (GRSL), 2022
Keyan Chen, Wenyuan Li, Jianqi Chen, Zhengxia Zou and Zhenwei Shi
[PDF] [Github] [Page]

We propose a novel scene classification method with scale and resolution adaptation ability. Unlike previous CNNbased methods that make predictions based on rasterized image inputs, the proposed method converts the images as continuous functions with INRs optimization and then performs classification within the function space.
sym

Building Extraction from Remote Sensing Images with Sparse Token Transformers
Remote Sensing, 2021
Keyan Chen, Zhengxia Zou and Zhenwei Shi
[PDF] [Github] [Page] [Demo]

We propose STT to explore the potential of using transformers for efficient building extraction. STT conducts an efficient dual-pathway transformer that learns the global semantic information in both their spatial and channel dimensions and achieves state-of-the-art accuracy on two building extraction benchmarks.

🎖 Honors and Awards

  • 2024 “Youth Basic Research Project (Ph.D.)”, National Natural Science Foundation of China (NSFC) (500 in China)
  • 2023 “National Scholarship of Ph.D.”, Beihang University (Top 1%)
  • 2022 “Outstanding M.S. Dissertation Award” of Beihang, Beihang University (Top 0.5%)
  • 2022 “Outstanding Graduates” of Beijing, Beijing (Top 1%)
  • 2019 “Outstanding B.S. Dissertation Award” of Beihang and Beijing, Beijing, (Top 0.5%)
  • 2019 “Outstanding Graduates” of Beihang, Beihang University (Top 5%)
  • 2019 “Excellent Student” of Beihang, Beihang University (Top 1%)
  • 2018 First Prize of the “Mathematics Modeling Competition”, Beihang University (Top 3%)
  • 2018 First Prize of “Lee Kum Kee Astronautics Scholarship”, Beihang University (Top 0.5%)
  • 2018 Special Scholarship of “Outstanding Academic Performance”, Beihang University (Top 2%)

📖 Educations

  • 2022.09 - Present, Ph.D. in Pattern Recognition and Intelligent System, Beihang University, China.
  • 2019.09 - 2022.01, M.S. in Pattern Recognition and Intelligent System, Beihang University, China.
  • 2015.09 - 2019.06, B.S. in Image Processing, Beihang University, China.

💬 Invited Talks

  • 2023.07, “Instance Segmentation Research Based on SAM Foundation Model”, OpenMMLab | [Video]
  • 2023.07, “Lightweight Interactive Instance Segmentation and Applications”, OPPO Research Institute internal talk
  • 2023.03, “Open-vocabulary Object Understanding”, Xiaohongshu Multimodal Team internal talk

💻 Academic Service

  • Reviewer for International Journal of Computer Vision (IJCV)
  • Reviewer for IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)
  • Reviewer for ISPRS Journal of Photogrammetry and Remote Sensing (ISPRS)
  • Reviewer for IEEE Transactions on Geoscience and Remote Sensing (TGRS)
  • Reviewer for IEEE International Journal of Digital Earth (IJDE)
  • Reviewer for IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (JSTARS)
  • Reviewer for Remote Sensing (RS)