网格生成是自主可控CAD/CAE工业软件研发的重要前处理步骤。课题组基于加权排序思想，提出了六面体网格的奇异结构简化方法，对解决六面体网格生成难题提供了新途径;提出了基于边界元与标架场的高质量四边结构网格自动生成方法、基于插值体细分的高阶网格生成方法，为解决结构化网格生成的自动化难题进行了有益探索。上述系列成果均发表在 CAD、CAGD 等本领域权威期刊上，并成功应用于国防基础科研核科学挑战专题项目以及国家数值风洞重大工程，为自主可控国产工业软件的研发做出了贡献。
Gang Xu*, Ran Ling, Yongjie Jessica Zhang, Zhoufang Xiao, Zhongping Ji, Timon Rabczuk, "Singularity structure simplification of hexahedral mesh via weighted ranking," Computer-Aided Design, 2021, 130: 102946. (CCF B)
Zhoufang Xiao, Shouping He, Gang Xu*, Jianjun Chen, Qing Wu, "Boundary element-based automatic domain partitioning approach for semi-structured quad mesh generation," Engineering Analysis with Boundary Elements, 2020, 113, 133-144.
Jin Xie, Jinlan Xu, Zhenyu Dong, Gang Xu*, Chongyang Deng, Bernard Mourrain, Yongjie Jessica Zhang, "Interpolatory Catmull-Clark volumetric subdivision over unstructured hexahedral meshes for modeling and simulation applications," Computer Aided Geometric Design,2020, 80, 101867, Special issue of GMP2020.
Zhoufang Xiao, Gang Xu*, Jianjun Chen*, Qing Wu, Shai Zhou, "A tailored fast multipole boundary element method for viscous layer mesh generation," Engineering Analysis with Boundary Elements, 2019, 99, 268-280.
Z Xiao, C Ollivier-Gooch, JDZ Vazquez, "Anisotropic Tetrahedral Mesh Adaptation with Improved Metric Alignment and Orthogonality," Computer-Aided Design,143, 103136.
智能制造与虚拟现实是数字经济产业的重要组成部分。复杂产品的计算域高质量参数化则是困扰智能制造与虚拟现实中高精度仿真分析方法向前发展的关键瓶颈问题之一。课题组在国际上最早开始了这一问题的研究，开辟了“适合分析的计算域参数化”这一研究方向，研究了计算域参数化对等几何仿真模拟精度的影响，并定义了“适合分析的参数化”的评判度量；创新性地提出了约束优化、变分调和映射、边界重新参数化等一系列构造高质量计算域参数化的理论和方法，并解决了复杂拓扑平面区域的参数化难题, 为任意复杂计算域的参数化问题提供了基本框架, 从而为高精度仿真分析提供了重要几何基础，丰富了数字几何计算基础理论。在本系列成果中，三篇论文入选ESI热点论文和高被引论文，三篇论文曾分别入选计算机辅助工程领域的TOP期刊CMAME和计算机辅助设计领域的TOP期刊CAD；近五年所发表论文中引用次数最多的25篇；论文列表(一篇在以国内机构为第一署名单位发表的 CAD 论文中排名第 1,SCI 他引次数为 93 次； 一篇在以国内机构为第一单位发表的 CMAME 论文中排名第 3，SCI 他引次数为 85 次)；
对于具有一致拓扑的CAD数字几何模型, 课题组提出了“分析重用”这一概念，给出了基于径向基函数的拓扑一致体参数化方法, 并给出了基于预计算的等几何分析重用方法，提高了体参数化与等几何分析求解的效率, 可推动等几何分析方法在实际工程中的应用。此外，基于新型样条模型，课题组提出了基于三角化区域上的复杂拓扑多元样条的等几何分析方法、基于扩 展 Loop 细分模式的新型等几何分析方法、计算域与物理域样条空间相异的扩展等几何分析方法等一系列新型高精度仿真分析求解方法，在求解精度、计算 效率和应用广度方面均有显著提高。上述系列成果均发表在 CAD、CMAME、IJNME 等本领域顶级期刊上。
S Wang, J Ren, X Fang, H Lin, G Xu, H Bao, J Huang , "IGA-suitable planar parameterization with patch structure simplification of closed-form polysquare," Computer Methods in Applied Mechanics and Engineering, 2022, 392: 114678.
Jinlan Xu, Chengnan Ling, Gang Xu*, Zhongping Ji, Xiangyang Wu,Timon Rabczuk, "Dynamic spline bas-relief modeling with isogeometric collocation method," Computer Aided Geometric Design, Volume 81, August 2020, 101913.
Gang Xu, Ming Li, Bernard Mourrain, Timon Rabczuk, Jinlan Xu, Stephane P.A. Bordas, "Constructing IGA-suitable planar parameterization from complex CAD boundary by domain partition and global/local optimization," Computer Methods in Applied Mechanics and Engineering , 2018, 328, 175-200.
Gang Xu, Tsz-Ho Kwok, Charlie C.L. Wang, "Isogeometric computation reuse method for complex objects with topology-consistent volumetric parameterization," Computer-Aided Design, 2017, 91, 1-13.
Gang Xu, Bernard Mourrain, Regis Duvigneau, Andre Galligo, "Analysis-suitable volume parameterization of multi-block computational domain in isogeometric applications.," Computer-Aided Design, 2013, 45(2), 395-404. Special issue on SPM'2012 Symposium on Solid and Physical Modeling (The paper was nominated as one of the Best Paper Awards Candidates of the conference)
Gang Xu, Bernard Mourrain, Regis Duvigneau, Andre Galligo, "Parameterization of computational domain in isogeometric analysis: Methods and comparison," Computer Methods in Applied Mechanics and Engineering, 2011, 200(23-24): 2021-2031.
主要研究深度学习在计算机视觉中的应用，并与计算机图形学技术相结合，涉及智能计算艺术、视频行为分析、艺术机器人等课题。课题组已在视觉质量评价与增强、图像生成与艺术风格迁移、视频人体检测与分析、虚拟试衣、绘画机器人等领域取得了丰硕成果。
Fei Gao, Xingxin Xu, Jun Yu, Meimei Shang, Xiang Li, and Dacheng Tao, "Complementary, Heterogeneous and Adversarial Networks for Image-to-Image Translation," IEEE Transactions on Image Processing, vol. 30, pp. 3487 - 3498, 2021. paper project
Jun Yu, Xingxin Xu, Fei Gao*, Shengjie Shi, et al. "Towards Realistic Face Photo-Sketch Synthesis via Composition-Aided GANs," IEEE Transactions on Cybernatics, DOI: 10.1109/TCYB.2020.2972944. (Corresponding Author) project paper_arxiv paper_ieee
Fei Gao, Jingjie Zhu, Zeyuan Yu, Peng Li, Tao Wang,
"Making Robots Draw A Vivid Portrait In Two Minutes,"
in the Proceedings of the 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2020),
pp. 9585-9591, Las Vegas, USA, 2020.
paper_iros
paper_arxiv
project
妙绘艺术 微信小程序
Lin Zhao, Meimei Shang, Fei Gao*, et al. "Representation Learning of Image Composition for Aesthetic Prediction," Computer Vision and Image Understanding (CVIU), vol. 199, 103024, Oct. 2020. project paper
Wang G, Wang Y, Zhang H, Gu R, Hwang JN. "Exploit the connectivity: Multiobject tracking with trackletnet," Proceedings of the 27th ACM International Conference on Multimedia (ACM MM)., 2019: 482-490.