蔡棽
男、副教授
个人简介
蔡棽,现为东华大学计算机学院应用系副教授。分别于2004年和2007在山东大学取得本科和硕士学位,2012年毕业于上海交通大学电院教育部重点实验室“系统控制与信息处理”模式识别与智能系统方向,2013年加入东华大学,并成立视觉与几何感知实验室。目前主要从事计算机视觉和图形学领域的研究工作,具体研究方向包括:摄像机标定、姿态估计、三维物体表示、识别与重建等。以一作或通信作者身份发表了10余篇较高质量论文,并获得国家发明专利授权三项,曾担任多个期刊会议的审稿人和sci期刊KSII的编辑,同时加入了CSIG视觉大数据专委会、三维视觉专委会和CAAI自然计算与数字智能城市专委会,并被聘为北航青岛研究院外聘专家。
研究方向:
相机标定
姿态估计
特征检测与对应
三维物体表示、识别与重建
基于深度学习的上述任务
讲授课程:
计算机图形学(2013-2022)
图像处理(2013-2018)
计算机视觉(2020-2022)
计算机组成原理(2013-2021)
工程实训(2013-2022)
主要论文研究成果:
[1] Shen Cai*, Longxiang Huang, Yuncai Liu. Automatically obtaining the correspondences of 4 coplanar points for an un-calibrated camera [J]. Applied Optics, 51(22), 5369-5376, 2012. (SCI)
[2] Shen Cai*, Zijian Zhao, Longxiang Huang, Yuncai Liu. Camera Calibration with Enclosing Ellipses by an Extended Application of Generalized Eigenvalue Decomposition [J]. Machine Vision and Applications, 24(3), 513-520, 2013. (SCI)
[3] Jiachun Wang, Hao Xu, Shen Cai*, Juan Guo. A rapid method for detecting objects with rectangular structures based on line correspondences[J]. Multimedia Tools & Applications, 75(15), 9409-9426, 2016. (SCI)
[4] Juan Guo, Shen Cai*, Zhanhao Wu, Yuncai Liu.A Versatile Homography Computation Method Based on Two Real Points [J]. Image and Vision Computing, 64C, 23-33, 2017. (SCI)
[5] Longxiang Huang, Xu Zhao*, Shen Cai and Yuncai Liu. Plate refractive camera model and its applications [J]. Journal of Electronic Imaging. 26(2), 2017. (SCI)
[6] Shen Cai*, Chenhao Wang, Yan Yan, Yuncai Liu. Analysis of the Pencil of Conics with Double Complex Contact and its Application to Camera Calibration [J]. Journal of Shanghai Jiao Tong University (Science), 18(1), 1-6, 2013. (EI)
[7] Shen Cai*, Longxiang Huang, Yuncai Liu. Revisiting Principal-Axes Aligned Conics: A Unified Algebraic Constraint and Geometric Interpretation for Camera Calibration[C]. IEEE International Conference on Information and Automation, 691-697, Shenyang, 2012. (EI)
[8] Shen Cai, and Zhuping Zang*. A Deformed Chessboard Pattern for Automatic Camera Calibration[C]. ICAICTE, 588-591, 2013.
[9] Cao Gu, Haikuan Du, Shen Cai and Xiaogang Chen*. Joint Multiple Image Parametric Transformation Estimation via Convolutional Neural Networks[J]. IEEE Access, 2018, volume6:18822-18831. (SCI)
[10] Shen Cai*, Zhanhao Wu. Camera calibration with coplanar conics: a unified explanation and ambiguity analysis[J]. IPSJ Transactions on Computer Vision and Applications, 10(14), 2018. (EI)
[11] Hui Cao, Haikuan Du, Siyu Zhang, Shen Cai*. InSphereNet: a Concise Representation and Classification Method for 3D Object. International Conference on Multimedia Modeling, Daejeon, 2020. (CCF-C)
[12] Siyu Zhang, Hui Cao, Yuqi Liu, Shen Cai*, Yanting Zhang, Yuanzhan Li, Xiaoyu Chi. SN-Graph: a Minimalist 3D Object Representation for Classification. International Conference on Multimedia & Expo (ICME), Shenzhen, 2021. (CCF-B, oral)
[13] Yuanzhan Li, Yuqi Liu, Yujie Lu, Siyu Zhang,Shen Cai*, Yanting Zhang. High-fidelity 3D Model Compression based on Key Spheres. Data Compression Conference (DCC), Salt Lake City, 2022. (CCF-B, full paper/oral presentation)
[14] Yongdong Huang, Yuanzhan Li, Xulong Cao, Siyu Zhang, Shen Cai*, Ting Lu, Yuqi Liu. An Efficient End-to-End 3D Voxel Reconstruction based on Neural Architecture Search. International Conference on Pattern Recognition (ICPR), Montreal, 2022. (CCF-C, oral)
[15] Xulong Cao, Yao Huang, Yongdong Huang, Yuanzhan Li, Shen Cai*. LDAM: Line Descriptors Augmented by Attention Mechanism. International Conference on Digital Image Processing (ICDIP), Wuhan, 2022. (EI)
[16] Yuqi Liu, Yin Wang, Haikuan Du, Shen Cai*. Spherical Transformer: Adapting Spherical Signal to CNNs. Chinese Conference on Pattern Recognition and Computer Vision (PRCV), Shenzhen, 2022. (CCF-C)
近期主持的科研项目:
1. 国家自然科学基金青年基金项目,“基于单应计算的视觉平面几何表示与认知研究”,2018.01-2020.12 (主持)
2. 中央高校基本科研业务费专项资金资助,“大规模图像基元与真实物体之间对应问题的研究”,2014.01-2016.12(主持)
3. 东华大学青年教师入校启动基金,2014.01-2014.12(主持)
4. 模式识别国家重点实验室开放课题基金 "基于几何基元对应的快速物体检测",2016.01-2017.12(主持)
5. 国家自然科学基金面上项目,“基于约束条件的非负矩阵分解算法及其在纤维自动识别中的应用研究”,2015.01-2017.12 (参与)
6. 东华大学非线性科学研究所专项资金“几何视觉中非线性约束的最优求解问题”,2014.01-2016.12(参与)
7. 多RGBD深度相机的基于机器人的鲁棒标定、外参数标定。(公司课题)
8. 人脸检测及特定动作识别系统 (公司课题)
9. VR头盔定位精度评测系统 (公司课题)
10. 变焦相机的快速标定算法 (公司课题)
11. 物流过程中的行为识别系统 (公司课题)
12. 激光点云与全景图像融合重建的高精度视觉定位算法 (公司课题)
获奖:
1. 阿里巴巴菜鸟网络2019全球科技挑战赛:智能体积测量。冠军(1/487)。
联系方式:cs@dhu.edu.cn
松江校区1号学院楼245室
非常欢迎数学专业本科生报考研究生或进入实验室实习!