曹思源，2016年本科毕业于天津大学电子信息工程学院，2022年本科博士于浙江大学信息与电子工程学院，现任浙江大学宁波校区信电分院助理研究员。博士期间获浙江大学优秀研究生、本科连续3年获国家奖学金。主持或参与国家自然科学基金青年基金，宁波市青年博士创新研究项目，航空科学基金，国家重点研发计划，浙江省自然科学基金重大项目等项目。长期从事多模态图像匹配、基于图像匹配的位姿估计、多模态图像处理、神经网络架构设计等研究，发表相关SCI论文22篇，其中CCF-A类/中科院1区论文18篇，一作/通讯论文11篇，累计受理/授权专利12篇。承担基于图像防抖去模糊的“超级夜景”技术研究，算法成果应用于vivo公司手机产品；承担针对目标检测准确率提升的短波近红外成像系统，形成相关硬件系统与软件算法交付华为公司；承担边缘端XXX检测跟踪相关课题，已成功应用于XXX研究院相关产业化产品。

科研情况

1. 项目研究

[1] 多模态（光谱）图像的无监督高精度可靠配准方法研究，科技部，国家自然科学基金青年基金，2024.01-2026.12 主持

[2] 多模态图像匹配自监督学习框架与高泛化设计应用，宁波市，青年博士创新研究项目，2024.06-2027.06 主持

[3] XXXX技术研究，中国航空研究院，航空科学基金，2024.05-2025.10 主持

[4] XXXX轻量化部署与联调技术，研究院，2024.09-2024.10 主持

[5] 边缘端轻量化XXXX算法，XXXX研究院，2023.01-2024.01 参与

[6] 极低照度多光谱成像降质机理建模与高质量复原技术研究，浙江省科技厅，省重大项目，2024.01-2026.12 参与

[7] 4D成像毫米波雷达传感器关键技术研究与产业化，科技部，国家重点研发计划，2023.12-2026.11 参与

[8] 基于RGB-IR Sensor的视频图像处理，瑞芯微电子股份有限公司，校企合作，2021.07-2023.06 参与

[9] 多源图像融合画质增强技术研究，维沃移动通信有限公司（vivo），校企合作，2020.11-2022.12 参与

[10] 短波红外波段彩色成像系统，华为技术有限公司，校企合作，2018.11-2020.05 参与

[11] 手机照片拍摄防抖技术研究，维沃移动通信有限公司（vivo），校企合作，2017.11-2019.11 参与

2. 论文、著作

[1] Si-Yuan Cao, Hui-Liang Shen, Shu-Jie Chen, and Chunguang Li. Boosting structure consistency for multispectral and multimodal image registration. lEEE Transactions on Image Processing. 29:5147-5162.2020.1 (CCF-A)

[2] Si-Yuan Cao, Jianxin Hu, Zehua Sheng, and Hui-Liang Shen. Iterative deep homography estimation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 1879–1888, 2022. (CCF-A)

[3] Beinan Yu, Yifan Chen, Si-Yuan Cao*, Hui-Liang Shen, and Junwei Li. Three-channel infrared imaging for object detection in haze. IEEE Transactions on Instrumentation and Measurement, 71:1–13, 2022.

[4] Si-Yuan Cao, Beinan Yu, Lun Luo, Runmin Zhang, Shu-Jie Chen, Chunguang Li, and Hui-Liang Shen. PCNet: A structure similarity enhancement method for multispectral and multimodal image registration. Information Fusion, 94:200–214, 2023. (中科院1区)

[5] Si-Yuan Cao, Runmin Zhang, Lun Luo, Beinan Yu, Zehua Sheng, Junwei Li, and Hui-Liang Shen. Recurrent homography estimation using homography-guided image warping and focus transformer. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 9833–9842, 2023. (CCF-A)

[6] Runmin Zhang, Jun Ma, Si-Yuan Cao*, Lun Luo, Beinan Yu, Shu-Jie Chen, Junwei Li, and Hui-Liang Shen. SCPNet: Unsupervised cross-modal homography estimation via intra-modal self-supervised learning. In Proceedings of the European Conference on Computer Vision (ECCV), 2024. (CAAI-A)

[7] Haokai Zhu, Si-Yuan Cao*, Jianxin Hu, Sitong Zuo, Beinan Yu, Jiacheng Ying, Junwei Li, and Hui-Liang Shen. MCNet: Rethinking the core ingredients for accurate and efficient homography estimation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 25932–25941, 2024. (CCF-A)

[8] Xue Zhang, Si-Yuan Cao*, Fang Wang, Runmin Zhang, ZheWu, Xiaohan Zhang, Xiaokai Bai, and Hui-Liang Shen*. Rethinking early-fusion strategies for improved multispectral object detection. IEEE Transactions on Intelligent Vehicles, 2024. (中科院1区)

[9] Lun Luo, Shuhang Zheng, Yixuan Li, Yongzhi Fan, Beinan Yu, Si-Yuan Cao*, Junwei Li, and Hui-Liang Shen*. BEVPlace: Learning LiDAR-based place recognition using bird’s-eye view images. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 8700–8709, 2023. (CCF-A)

[10] Zhu Yu, Zehua Sheng, Zili Zhou, Lun Luo, Si-Yuan Cao*, Hong Gu, Huaqi Zhang, and Hui-Liang Shen*. Aggregating feature point cloud for depth completion. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 8732–8743, 2023. (CCF-A)

[11] Zhu Yu, Runmin Zhang, Jiacheng Ying, Junchen Yu, Xiaohai Hu, Lun Luo, Si-Yuan Cao, and Hui-Liang Shen*. Context and geometry aware voxel transformer for semantic scene completion. Advances in Neural Information Processing Systems,2024. (CCF-A)

[12] Lun Luo, Si-Yuan Cao, Bin Han, Hui-Liang Shen, and Junwei Li. BVmatch: Lidar-based place recognition using bird's-eye view images. lEEE Robotics and Automation Letters. 6(3):6076-6083.2021.

[13] Jiacheng Ying, Hui-Liang Shen, and Si-Yuan Cao. Unaligned hyperspectral image fusion via registration and interpolation modeling. lEEE Transactions on Geoscience and Remote Sensing, 60:1-14.2021. (中科院1区)

[14] Lun Luo, Si-Yuan Cao, Zehua Sheng, and Hui-Liang Shen. LiDAR-based global localization using histogram of orientations of principal normals. IEEE Transactions on Intelligent Vehicles, 7(3):771–782, 2022. (中科院1区)

[15] Zehua Sheng, Xiongwei Liu, Si-Yuan Cao, Hui-Liang Shen, and Huaqi Zhang. Frequency-domain deep guided image denoising. IEEE Transactions on Multimedia, 25:6767–6781, 2022. (中科院1区)

[16] Zehua Sheng, Zhu Yu, Xiongwei Liu, Si-Yuan Cao, Yuqi Liu, Hui-Liang Shen, and Huaqi Zhang. Structure aggregation for cross-spectral stereo image guided denoising. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 13997–14006, 2023. (CCF-A)

[17] Jiacheng Ying, Can Tong, Zehua Sheng, Bowen Yao, Si-Yuan Cao, Heng Yu, and Hui-Liang Shen. Region-aware RGB and near-infrared image fusion. Pattern Recognition, 142:109717, 2023. (中科院1区)

[18] Beinan Yu, Jiacheng Ying, Lun Luo, Si-Yuan Cao, Xiansong Bao, and Hui-Liang Shen. Vignetting correction using an optical model and constant chromaticity prior. IEEE Transactions on Computational Imaging, 9:1071–1083, 2023.

[19] Shuhang Zheng, Yixuan Li, Zhu Yu, Beinan Yu, Si-Yuan Cao, Minhang Wang, Jintao Xu, Rui Ai, Weihao Gu, Lun Luo, et al. I2P-Rec: Recognizing images on large-scale point cloud maps through bird’s eye view projections. In 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 1395–1400, 2023.

[20] Xiaohan Zhang, Xue Zhang, Si-Yuan Cao, Beinan Yu, Chenghao Zhang, and Hui-Liang Shen. MRF3Net: Infrared small target detection using multi-receptive field perception and effective feature fusion. IEEE Transactions on Geoscience and Remote Sensing, 2024. (中科院1区)

[21] Runmin Zhang, Zhu Yu, Zehua Sheng, Jiacheng Ying, Si-Yuan Cao, Shu-Jie Chen, Bailin Yang, Junwei Li, and Hui-Liang Shen. SGDFormer: One-stage transformer-based architecture for cross-spectral stereo image guided denoising. Information Fusion, 113:102603, 2025. (中科院1区)

[22] Zhe Wu, Zehua Sheng, Xue Zhang, Si-Yuan Cao, Runmin Zhang, Beinan Yu, Chenghao Zhang, Bailin Yang, and Hui-Liang Shen. STARNet: Low-light video enhancement using spatio-temporal consistency aggregation. Pattern Recognition, 160:111180, 2025. (中科院1区)

3. 成果奖项

无

4.发明专利

[1] 基于多光谱与多模态图像一致性增强网络的配准方法专利号：2021108906387 申请日：2021.08.04 授权日：2023.10.27

[2] 基于运动场景的热红外非均匀噪声校正方法专利号：2022112942291 申请日：2022.10.21 授权日：/

[3] 一种基于卷积神经网络的烟叶分类方法专利号：2023102748069 申请日：2023.03.15 授权日：/

[4] 一种黑白图像引导的彩色RAW图像联合去噪去马赛克方法 专利号： 2023102775812 申请日：2023.03.21 授权日：/

[5] 一种基于自适应卡尔曼滤波的实时视频稳像方法专利号：2023102888647 申请日：2023.03.22 授权日：/

[6] 一种基于多尺度深度特征图融合的多光谱图像配准方法专利号：2023112454907 申请日：2023.09.25 授权日：/

[7] 基于预测校正和汇聚注意力transformer的多模态图像配准方法 专利号： 2023112455702 申请日：2023.09.25 授权日：/

[8] 一种基于强化学习的相机自动对焦方法专利号：2023114630547申请日：2023.11.06 授权日：/

[9] 一种基于特征增强及多尺度相关的多模态图像配准方法专利号：2023115049838 申请日：2023.11.13 授权日：/

[10] 基于多模态光谱图像配准的无监督学习方法专利号：202311839902X 申请日：2023.12.27 授权日：/

[11] 基于双层特征提取网络的可见光与热红外融合方法专利号：2024100986283 申请日：2024.01.24 授权日：/

[12] 基于深度多亮度映射无监督融合网络的红外图像增强方法专利号：2024101251797 申请日：2024.01.30 授权日：/