Matching images with significant scale differences remains a persistent challenge in photogrammetry and remote sensing. The scale discrepancy often degrades appearance consistency and introduces uncertainty in keypoint localization. While existing methods address scale variation through scale pyramids or scale-aware training, matching under significant scale differences remains an open challenge. To overcome this, we address the scale difference issue by detecting co-visible regions between image pairs and propose SCoDe (Scale-aware Co-visible region Detector), which both identifies co-visible regions and aligns their scales for highly robust, hierarchical point correspondence matching. Specifically, SCoDe employs a novel Scale Head Attention mechanism to map and correlate features across multiple scale subspaces, and uses a learnable query to aggregate scale-aware information of both images for co-visible region detection. In this way, correspondences can be established in a coarse-to-fine hierarchy, thereby mitigating semantic and localization uncertainties. Extensive experiments on three challenging datasets demonstrate that SCoDe outperforms state-of-the-art methods, improving the precision of a modern local feature matcher by 8.41%. Notably, SCoDe shows a clear advantage when handling images with drastic scale variations.
Hello World!
Hi, I’m currently a Master's student in the
State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing
(LIESMARS)
at
Wuhan University, under the
guidance of
Prof. Xianwei Zheng. I received my B.Eng. in Remote Sensing Science and Technology from Wuhan University in
2023. I have previously researched GenAI applications in image and video generation under the
supervision of
Dr. Yan Zhang
during my internship at
Baidu, Inc. (International Technology R&D Department)
Currently, I am also a remote research intern at
Centre for Frontier AI Research (CFAR), Agency for Science, Technology and Research
(A*STAR), supervised by
Dr. Xingrui Yu, where I work on
reinforcement learning and embodied intelligence, with a focus on generalizable, agent-centric
policy learning.
My research interests lie in computer vision and generative AI, with a focus on unifying 2D and 3D representations through image correspondence, cross-view understanding, and structure-aware generation. I aim to develop general spatial intelligence models that bridge perception, geometry, and trustworthy generation at scale, advancing the next generation of spatially grounded, intelligent visual systems. Beyond foundational research, I am particularly interested in enabling embodied intelligence and autonomous agents, where spatial reasoning and generative modeling empower real-world interaction, decision-making, and large-scale multi-agent collaboration.
Experiences
Publications
SAMatcher: Segment Anything Co-visible for Robust Feature Matching
Xu Pan, Qiyuan Ma, Jintao Zhang, Xianwei Zheng*
(In Preparation) 2026
Flow-Matching RL with Streaming Spatial Perception for Dynamic Manipulation
Xu Pan, Zhenglin Wan, Xingrui Yu*
(In Preparation) 2026