Eason Zhang

Yesheng Zhang Eason Zhang

I am a PhD student in the Computer Vision Lab at Shanghai Jiao Tong university, under the supervision of Prof. Xu Zhao.

Before that I received my Master's (2023) and Bachelor's (2020) degree from School of Biomedical Engineering and Institute of Medical Robotics, Shanghai Jiao Tong University, advised by Prof. Dahong Qian and Prof. Xu Zhao.

Email / Google Scholar / Github

Research

I am passionate about 3D computer vision, including scene reconstruction and understanding.

Publications

	Searching from Area to Point: A Hierarchical Framework for Semantic-Geometric Combined Feature Matching Yesheng Zhang, Xu Zhao Pattern Recognition (PR), 2026 paper \| page \| code Semantic-geometric combined feature matching in a unified searching perspective.
	MESA: Effective Matching Redundancy Reduction by Semantic Area Segmentation Yesheng Zhang, Shuhan Shen, Xu Zhao IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2026 paper \| page \| invited talk \| code Densely or Sparsely Matching every areas and points between images for many 3D downstream tasks, utilizing SAM.
	Semantic-guided Camera Ray Regression for Visual Localization Yesheng Zhang, Xu Zhao International Conference on Computer Vision (ICCV), 2025 paper \| page Visual Localization by regressing camera rays with semantic guidance.
	MESA: Matching Everything by Segmenting Anything Yesheng Zhang, Xu Zhao IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024 paper \| page \| video \| code Matching every areas and points between images, utilizing SAM.
	Learning-Based Distortion Correction and Feature Detection for High Precision and Robust Camera Calibration Yesheng Zhang, Xu Zhao, Dahong Qian IEEE Robotics and Automation Letters (RA-L), 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2022 paper \| code \| video We proposed a learning-based camera calibration system, which combines deep learning methods and traditional methods.

Preprints

	Structuring GUI Elements through Vision Language Models: Towards Action Space Generation Yi Xu, Yesheng Zhang, Jiajia Liu, Jingdong Chen arXiv preprint, 2025 paper An IoU-augmented maximum likelihood training paradigm for improved GUI element coordinate prediction in MLLMs.
	Trajectory Entropy: Modeling Game State Stability from Multimodality Trajectory Prediction Yesheng Zhang, Wenjian Sun, Yuheng Chen, Qingwei Liu, Qi Lin, Rui Zhang, Xu Zhao arXiv preprint, 2025 paper Trajectory Entropy quantifies game-state stability from multimodal trajectory predictions to improve level-k planning accuracy and efficiency.

Thank Jon Barron for sharing his website's source code.