课程题目：Computational Fabrication, from Appearance to Deformation
课程摘要：I will introduce two 3D coloring techniques and a pneumatic soft objects fabrication method. Currently, color 3D printing is still unavailable for low cost 3D printers. We proposed two 3D surface coloring techniques, hydrographics and thermoforming. By simulating the physical process, we calculate pre-distorted images which can wrap around the object and stretch in the way we have designed in physical realization, thus give single color 3D printed objects the desired texture. I will also introduce a solution for design and fabrication of soft pneumatic objects with desired deformations. Given a 3D object with its rest and deformed target shapes, our two-step method automatically optimizes the chamber structure and material distribution inside the object volume so that the fabricated object can deform to all the target deformed poses with controlled air injection. The design results can be fabricated with 3D printing and deformed by a controlled pneumatic system.
课程题目：Learning to Estimate 3D Human Pose and Shape from 2D Image
课程摘要：Recovering 3D human pose is a challenging problem with many applications. It has been conditionally solved by motion capture systems or depth sensors. This talk will discuss the more challenging case of using a single RGB camera: going directly from 2D appearance to 3D geometry. While deep learning approaches have shown remarkable abilities to solve 2D vision problems, it is difficult for them to directly learn and predict 3D geometry due to the lack of training data and higher dimensionality and nonlinearity of the solution space. In this talk, I will introduce our recent efforts toward 3D human pose and shape prediction from a single image, which solve the aforementioned challenges by integrating deep learning with geometric models as well as end-to-end learning using weakly-annotated data and multi-view geometry.
讲者简介：周晓巍，浙江大学计算机学院CAD&CG国家重点实验室研究员、国家青年千人计划入选者。2008年获浙江大学信息工程专业学士学位，2013年获香港科技大学电子及计算机工程博士学位。2014年至2017年在美国宾夕法尼亚大学计算机及信息科学系、GRASP机器人实验室从事博士后研究。主要研究领域为计算机视觉、机器人感知、医学图像分析，尤其在三维物体识别、人体姿态估计、图像匹配等方面取得了一系列重要成果。策划和组织了Geometry Meets Deep Learning Workshops，并长期担任PAMI、IJCV、TIP等二十余种SCI期刊审稿人以及CVPR、ICCV、IJCAI等计算机领域顶级会议程序委员会委员。
课程题目：A Geometric View to Optimal Transportation and Generative Model
课程摘要：In this talk, we introduce the intrinsic relations between optimal transportation and convex geometry, especially the variational approach to solve Alexandrov problem: constructing a convex polytope with prescribed face normals and volumes. This leads to a geometric interpretation to generative models, and leads to a novel framework for generative models. By using the optimal transportation view of GAN model, we show that the discriminator computes the Kantorovich potential, the generator calculates the transportation map. For a large class of transportation costs, the Kantorovich potential can give the optimal transportation map by a close-form formula. Therefore, it is sufficient to solely optimize the discriminator. This shows the adversarial competition can be avoided, and the computational architecture can be simplified. Preliminary experimental results show the geometric method outperforms WGAN for approximating probability measures with multiple clusters in low dimensional space.
讲者简介：雷娜，大连理工大学国际信息与软件学院教授，博士生导师，兼任北京成像技术高精尖创新中心研究员；中国工业与应用数学学会几何设计与计算专业委员会委员；中国数学会计算机数学专业委员会委员；中国计算机学会计算机视觉专业委员会委员; 美国数学会 Mathematical Review评论员。纽约州立大学石溪分校计算机系访问教授；德克萨斯大学奥斯汀分校计算工程与科学研究所JTO research fellow；清华大学数学科学中心访问教授；中科院数学与系统科学研究院访问学者。IEEE Transactions on Visualization and Computer Graphics，Computer Aided Geometric Design, Graphical Models， The Visual Computer, Journal of Computational and Applied Mathematics, Journal of Systems Science and Complexity, SCIENCE CHINA Mathematics等国际期刊审稿人。研究方向为：应用现代微分几何和代数几何的理论与方法解决工程及医学领域的问题，主要聚焦于计算共形几何、计算拓扑、符号计算及其在计算机图形学、计算机视觉、几何建模和医学图像中的应用。
课程摘要：Machine learning has demonstrated being highly successful at solving many real-world applications ranging from information retrieval, data mining, and speech recognition, to computer graphics, visualization, and human-computer interaction.. However, most users often treat the machine learning model as a “black box” because of its incomprehensible functions and unclear working mechanism. Without a clear understanding of how and why the model works, the development of high-performance models typically relies on a time-consuming trial-and-error procedure. This talk presents the major challenges of interactive machine learning and exemplifies the solutions with several visual analytics techniques and examples, including model understanding and diagnosis.
讲者简介：Shixia Liu is a tenured associate professor at Tsinghua University. Her research interests include visual text analytics, visual social analytics, visual behavior analytics, graph visualization, and tree visualization. Before joining Tsinghua University, she worked as a lead researcher at Microsoft Research Asia and a research staff member at IBM China Research Lab. Shixia is one of the Papers Co-Chairs of IEEE VAST 2016 and 2017. She is an associate of IEEE Transactions on Visualization and Computer Graphics and is on the editorial board of Information Visualization. She was the guest editor of ACM Transactions on Intelligent Systems and Technology and Tsinghua Science and Technology. She was the program co-chair of PacifcVis 2014 and VINCI 2012. Shixia was in the Steering Committee of VINCI 2013. She is on the organizing committee of IEEE VIS 2015 and 2014. She is/was in the Program Committee for CHI 2019, 2018, InfoVis 2015, 2014, VAST 2018, 2015, 2014, KDD 2015, 2014, 2013, ACM Multimedia 2009, SDM 2008, ACM IUI 2011, 2009, PacificVis 2008, 2009, 2010, 2011, PAKDD 2013, VISAPP 2012, 2011, VINCI 2011.
课程题目：Robust mesh generation and applications to geometry processing
授课讲者1：Xifeng Gao - New York University, Florida State University, USA
授课讲者2：Daniele Panozzo - New York University, USA
课程摘要：We will overview of recent advancements in robust geometry processing algorithms. The talk will cover the robust generation of surface and volumetric meshes and their applications to computer graphics and mechanical engineering applications. The frontal lectures will be complemented by a hands-on introduction of libigl, an open-source geometry processing library (github.com/libigl/libigl).
讲者1简介：Xifeng Gao is now holding a PostDoc position at the Courant Institute of Mathematical Sciences of New York University. Dr. Gao will join the Department of Computer Science at FSU in Fall 2018 as an Assistant Professor. He received his Ph.D. degree in 2016 and won the best Ph.D. dissertation award from the Department of Computer Science at the University of Houston. Dr. Gao has wide research interests that are related to geometry processing, such as Computer Graphics, Visualization, Multimedia Processing, Medical Imaging, Information Forensics, and Digital Fabrication. His research works have been published in several leading Journals, e.g., ACM TOG, ACM TOMM, CGF, and IEEE TVCG. More details about his research can be found on his homepage.
讲者2简介：Daniele Panozzo is an Assistant Professor of Computer Science at the Courant Institute of Mathematical Sciences in New York University. Prior to joining NYU he was a postdoctoral researcher at ETH Zurich (2012-2015). He earned his PhD in Computer Science from the University of Genova (2012) and his doctoral thesis received the EUROGRAPHICS Award for Best PhD Thesis (2013). Daniele’s research interests are in digital fabrication, geometry processing, architectural geometry and discrete differential geometry. He received the EUROGRAPHICS Young Researcher Award in 2015, the NSF CAREER Award in 2017, and his work has been covered by Swiss National Television and various national and international printed media. Daniele is leading the development of libigl (https://github.com/libigl/libigl), an award-winning (EUROGRAPHICS Symposium of Geometry Processing Software Award, 2015) open-source geometry processing library that supports academic and industrial research and practice. Daniele is chairing the Graphics Replicability Stamp (http://www.replicabilitystamp.org), which is an initiative to promote reproducibility of research results and to allow scientists and practitioners to immediately beneﬁt from state-of-the-art research results.