Computer Graphics in the Age of AI and Big Data （Hao Zhang）　

Computer graphics is traditionally defined as a field which covers all aspects of computer-assisted image synthesis. An introductory class to graphics mainly teaches how to turn an explicit model description including geometric and photometric attributes into one or more images. Under this classical and arguably narrow definition, computer graphics corresponds to a ``forward'' (synthesis) problem, which is in contrast to computer vision, which traditionally battles with the inverse (analysis) problem.In this talk, I would offer my view of what the NEW computer graphics is, especially in the current age of machine learning and data-driven computing. I will first remind ourselves several well-known data challenges that are unique to graphics problems. Then, by altering the above classical definition of computer graphics, perhaps only slightly, I show that to do the synthesis right, one has to first solve various inverse problems. In this sense, graphics and vision are converging, with data and learning playing key roles in both fields. A recurring challenge however is a general lack of “Big 3D Data”, which graphics research is expected to address. Finally, I want to explore a new perspective for the synthesis problem to mimic a higher-level human capability than pattern recognition and understanding.

Data Visualization: an Introduction （Oliver Deussen）　

In data visualization large amounts of data are converted into images in order to use the human eye to find interesting patterns that can later be analysed by data mining techniques. In my talk I will give brief overview about various data visualization techniques and will mention the most important research questions of this field.

The Many Faces of Stiffness （Uri Ascher）　

The words ”stiff”, “stiffness” , “stiffening”, etc., arise often in the visual computing context when simulating, calibrating and controlling dynamics. Instances include fluid flow (for example, simulations of water or smoke) and elastodynamics (for example, in deformable object simulations). But these words often have different meanings in different contexts. A subset of topics, on which we will concentrate, includes: 1. Textbook-type (decaying) numerical ODE stiffness; 2. Highly oscillatory stiffness; 3. Stiffness matrix; 4. Numerical stiffening Some of these terms are popular in scientific computing, while others arise in mechanical engineering. A potential confusion may arise in this way, and it gets serious when more than one meaning is encountered in the context of one application or project. Such is the case with the simulation of deformable objects in visual computing, where all of the above appear in one way or another under one roof. In this lecture I will describe the meaning of stiffness in each of these topics, how they arise, how they are related, what practical challenges they bring up, and how these challenges are handled in context. The concepts and their evolution will be demonstrated. It is about meshes - their resolution and spectral properties - both in time and in space.

Physics Based Computer Animation Fundamentals （Paul Kry）　

This 80 minute course will introduce the main concepts of physically based computer animation, with a focus on elastic and rigid body systems. We will examine different numerical integration techniques and see how the eigenvalue decomposition of a linear system can be used to understand the long term behavior of different stepping methods. We will look at real examples of larger stiff systems that benefit from iterative linear solvers, and see how constraints and constraint stabilization help and where they fail. Various concepts will be demonstrated with interactive 2D examples. Additional resources for future study will be included throughout the course. To conclude I will summarize avanced topics that build on this material.

Research on Interactive Virtual Reality （Weiwei Xu）

Virtual reality (VR) and augmented reality (AR) technologies are rapidly evolving toward the goal of universalization and commercialization. However, the production efficiency of virtual reality content has not met the rapid development of virtual reality technology, and has become a bottleneck for the widespread application of virtual reality technology. This talk will introduce our research on rapid modeling and physical simulation, which provide technical support for interactive virtual reality content production. More specifically, the talk will include: 1) modeling motions based on multi-view images, which realizes joint modeling of geometric details and motion functions of moving objects; 2) fast physical simulation based on Krylov subspace and region decomposition, which effectively improves the speed of object motion simulation and thus is suitable for interactive applications in virtual reality.

Animating Cameras: Interactive Virtual Cinematography （Marc Christie）　

In this talk I will present the scientific and technical challenges pertained to virtual cinematography. By focusing on an approach that consists in extracting and formalizing cinematographic knowledge from real movies (camera placement, motions and cuts, and actor staging), and transposing this knowledge to virtual and interactive environments, we will review the main approaches and present recent advances. We will also draw a clear picture of challenges that need to be addressed in the future, and show that this topic calls for a multidisciplinary approach ranging from cognitive sciences to semiotics, multimedia analysis and computer graphics

The importance of geometry in stylization of images （Ariel Shamir）

Many popular techniques for stylizing images and videos appeared in the last few years. Many of them are based on neural networks and apply texture synthesis methods. In this talk I will present the ideas behind image stylization using neural networks as well as some earlier works, not relying on neural networks, for stylization and abstraction of portraits and videos. The key insight in these works is that texture transfer is not always enough, and geometric structure should also be considered to capture artistic style.

Human Perception of 3D Shapes （Manfred Lau）　

I will discuss my recent research direction of learning human-perceptual measures of 3D shapes. The key idea is to explore the human perception of 3D shapes by learning with raw data and without any pre-defined computational description that is typical in existing work. I have applied this idea to study several shape perception problems: human-perceived tactile and softness properties of virtual shapes, style-similarity metrics of 3D shapes, and the aesthetics of 3D shapes. For each of these problems, I will describe the data collection step for collecting human data of the perceptual property, the machine learning step for learning a perceptual measure from the data, and show the results and applications of the learned perceptual measure.

Generative Adversarial Networks and their Applications （Dani Lischinski）　

In this tutorial talk we will describe Generative Adversarial Networks (GANs). We will start by explaining the motivation and the theory behind these networks, which are considered by some as one of the most important developments in Artificial Intelligence. Next we will survey various extensions and applications of GANs, including the recent exciting cycleGAN/dualGAN architecture.

Modeling Real-World Scenes （Liangliang Nan）　

Capturing the real world scenes in the 3D format has been made possible by advances in scanning and photogrammetric technologies. This has attracted increasing interests in acquiring, analyzing, and modeling real-world scenes. However, obtaining a faithful 3D representation of real-world scenes still remains an open problem. In this talk, I would like to share my experiences in the past few years in reconstructing urban scenes. I will present algorithms for reconstructing coarse models and algorithms for enriching the coarse models with fine details. In the end, we will discuss the trend and some topics for the future research.

Coupled Object and Actor Reconstruction from Monocular Input （Niloy Mitra）　

A long-standing challenge in scene analysis is the recovery of scene arrangements under moderate to heavy occlusion, directly from monocular video. The problem is inherently ill-posed. Over decades, researchers have studied the use of various regularizers in the form of transformation groups (e.g., symmetry types), data priors (e.g., database shapes), functional priors (e.g., object affordance), etc. In the last years, we have been investigating the utility of physics- and interaction-based priors to recover occluded geometry along with scene actors in an effort to extract semantic scene understanding from raw (dynamic) inputs. In this talk, I will report our recent results on this topic. For more information, please visit http://geometry.cs.ucl.ac.uk/.

Clustering for Graphical Applications （Daniel Cohen-Or）　

In recent years, computer graphics has been enjoyed and benefitted from advances in machine learning, and more and more graphics techniques are based on learning techniques. In this talk, I will present three novel graphical methods that are based on non-parametric clustering technique: Collection distilling, Data-driven morphing, and weak Convexity Shape Decomposition.

上机实验课程 （Jiacheng Ren）　

上机实验课程的主要学习内容为，基础三维模型和材质的制作，简单灯光和渲染的设置，以及画面构图和构成的学习。本课程的目的是带领大家初步进入三维世界，培养三维空间想象力和动手能力，为今后进入计算机图形学等方向的学习和研究，提供必要的技术和艺术方面的帮助。
课程使用的软件为3dmax，考虑到只有三个晚上近六个课时的学习时间，以及大部分学生没有接触过此类软件，所以课程会从最基本的软件操作开始，到复杂模型制作，循循渐进。用一些比较有意思的实例，快速掌握软件操作的同时，分析其制作的过程和思路，着重培养大家独立思考和学习的能力。最终可以用所学到的制作方法，举一反三，实现自己想要创作的三维物体，也为课程学习结束后，自行学习打下基础。