紫藤学者

陈圣梅，中共党员，博士，西北大学电子信息学院（人工智能学院）博士后。2026年获得西北大学工学博士学位。主要研究领域为人工智能，深度学习。研究工作主要发表于IEEE TGRS、IEEE TASE、NeurIPS和ICASSP等领域内主流期刊和会议，授权/受理发明专利5件。研究方向：智能机器人环境感知，三维点云处理。

代表性研究成果：

Shengmei Chen, Hao Deng, Cheng Liu, Jinye Peng, Lin Wang*. Full-DOF calibration method for 3D point cloud acquisition device based on BiK loss function[J]. IEEE Transactions on Automation Science and Engineering, 2026.

Shengmei Chen, Chenyu Hou, Xiang Zhang, Junhao Jia, Chenji Wang, Jinye Peng, Lin Wang*. REKFReg: rotation equivariant and keypoint focused learning for RGB-D point cloud registration[J]. Applied Soft Computing, 2026.

Shengmei Chen, Chenyu Hou, Hao Deng, Jinye Peng, Lin Wang*. GIREG: geometirc-image collaborative point cloud registration[C]. IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2026.

Shengmei Chen, Lu Ren, Jinye Peng, Gang Wei*, Zhe Zhang, Cheng Liu, Lin Wang*. A calibration algorithm of 3D laser scanner based on KMPE loss function[J]. IEEE Sensors Journal, 2023.

Shengmei Chen, Jingyi Han, Lu Ren, Tianzhang Xing*, Lin Wang*. KMPE loss function-based clustering multi-view point cloud registration algorithm[C]. IEEE International Conference on Parallel and Distributed Systems (ICPADS), 2025.

Shengmei Chen, Hao Deng, Jingyi Han, Cheng Liu, Jinye Peng, Lin Wang*. MSKICP: multiscale descriptor and KMPE kernel function-improved iterative closest point for point cloud registration[J]. Multimedia Systems, 2026.

Hao Deng, Kunlei Jing*, Shengmei Chen, Cheng Liu, Jiawei Ru, Bo Jiang, Lin Wang*. LinNet: linear network for efficient point cloud representation learning[C]. Annual Conference on Neural Information Processing Systems (NeurIPS), 2024.