Graphics as Visual Computing

Hui Huang        Ruizhen Hu
  Why should we learn Computer Graphics?

Learn a new perspective/view of computing.
How to represent 3D objects in a computer?
How to present visual information?
What to do with 3D objects with a computer?
How to connect virtual and real world with computer graphics?

 

Course Description


Pre-requisites

 

    • - Linear Algebra
    • - Calculus
    • - Algorithm and Program Design

Pre-
course

 

- Introduction to Computer Graphics

(The first semester of Junior year,

Major of Computer Science and Software Engineering)

 

Time/Location

Content

• March 14th 15:50 - 17:20 (624)
• March 18th  09:30 - 12:30 (938)
• March 21st  15:50 - 17:20 (624)

Creativity, Creative Thinking and Problem Solving

 

How to find good ideas? How to evaluate if a new idea is a good one?

In this seminar I will talk about what creativity is and what kinds of creativity there are. We will speak about creativity techniques and evaluation methods for ideas. Together we will develop creative solutions to all kinds of daily problems, we will play creativity games and have a lot of fun together. At the end of the seminar participants will understand how to improve their creative abilities, how to find new and even unexpected solutions to given problems. Participants will also get a good feeling for the needed effort to realize an idea.

• Start from April 11th (The sixth week)

• Tuesday 3:50-5:20pm, Room 624
Seminar (6 topics - 12 weeks – 24 presentations)

• Point Clouds
• Surface Reconstruction
• Geometric Elements
• Shape Semantics
• Scene Modeling & Understanding
• Image & Video

What is “Advanced”?

Introduction to Computer Graphics

Advanced Computer Graphics

Course Contents

2D/3D concepts, graphic processing pipelines, theories and methods

General questions of Graphics
Design ideas and theories
Recent development of Graphics

Course Objectives

Know the digital expression of visual information

Know the methodology of visual expression and how to do scientific research

Teaching Methods

Study

Research and study

Lecture

Paper study (presentations and reports)

Grading

Class participation +Practice

Class Presentation (Discussion) + Study Report

Seminar

- Presentation Schedule: 30 minutes presentation and 15 minutes Q&A for each speaker
- Presentation topic: choose from the recommended papers
- Presentation requirements: prepare PPT in English, which contains research background, relevant works, main contributions, key techniques, experimental evaluations and applications.

- Grading: all audiences will rate the presentation at the provided evaluation form; interactions with the audiences will also be considered. )

April 11th

1. Tips on How to Read and Present a Paper
2. Co-Segmentation of 3D Shapes via Subspace Clustering (Speaker: Ruizhen Hu)

April 18th

1. Analysis and Controlled Synthesis of Inhomogeneous Textures (Speaker: Yang Zhou)
2. As-Rigid-As-Possible Shape Interpolation / Surface Modeling / Shape Manipulation (Speaker: Shengqiu Huang)

April 25th

1. Efficiently Combining Positions and Normals for Precise 3D Geometry (Speaker: Kaixi Xie)
2. Shading-aware Multi-view Stereo (Speaker: Xiaohui Zhou)

May 2nd

1. Robust Moving Least-Squares Fitting with Sharp Features (Speaker: Timur Valiullin)

2. Surface Reconstruction from Unorganized Points (Speaker: Lilia Nizamova)

May 9th

1. Poisson Surface Reconstruction (Speaker: Zhijie Wu)

2. Real-time 3D Reconstruction at Scale using Voxel Hashing (Speaker: Guohang Yan)

May 16th

1. A Volumetric Method for Building Complex Models from Range Images (Speaker: Yunpeng Cui)

2. Laplacian Surface Editing (Speaker: Valerii Bagaveev)

May 23th

1. Connectivity Shapes (Speaker: Anna Repina)

2. Learning 3D Mesh Segmentation and Labeling (Speaker: Ayrat Mutygullin)

June 6th 1. Poisson Image Editing (Speaker: Lingli He)
2. GrabCut - Interactive Foreground Extraction using Iterated Graph Cuts (Speaker: Weizhi An)
June 13th 1. Image Halftoning (Speaker: Daniel Cohen-Or)
2. Color Quantization (Speaker: Daniel Cohen-Or)
June 20th 1. Learning How Objects Function via Co-analysis of Interactions (Speaker: Zihao Yan)
2. Global Contrast based Salient Region Detection (Speaker: Jianhong Yuan)
June 27th

1. Semantic Scene Completion from a Single Depth Image (Speaker: Qiuyang Chen)
2. Geometric Context from a Single Image (Speaker: Hao Yang)

 

Grading Policy



10%

 



40%


50%

 

 

 

+10%

 

Class Participation

Paper Presentation

Final Study Report

Final Demo Bonus

- Class Attendance: a student should report absences to instructor before the class, but the maximum amount of leave during this course is TWO. Absence from class without such notification will lead to course failure.

- Discussion Participation: try to actively participate in all the discussions after each presentation.

- The paper of your presentation is chosen from the paper list, consisting of 6 topics.

- According to the performance of your presentation, instructor and your classmates will rate your presentation at the evaluation form.

 

 

- Based on the paper you presented, read at least TWO more related papers.

- Write a study report using the provided latex template. The report can be written in English or Chinese.

 

 

 

 

- We encourage you to implement the algorithm or find & run the source codes of the paper you present if available to better understand the proposed method. Showing your demo with some new results which are not presented in the paper will earn extra bonus scores.

 

 

Presentation Requirements

- Suppose the audiences have no any relevant knowledge background, you need to explain the problem clearly enough to let them understand.

- The presentation should better start from explaining the key problem that need to be solved in the paper and the key idea of the proposed algorithm, i.e., the big picture. Then you illustrate the details with examples.

- Before presentation, read through the paper you are going to present carefully and repeatedly, so that you can be clear about all the details of the techniques proposed in the paper and able to answer all the questions raised by the audiences.

- Use the attached template to make your PPT in English.

- One week before your presentation, contact the instructor to check whether your prepared PPT and contents meet the requirements or not. You can then refine them according to the instructor’s comments.

Report Requirements

  1. - Give a concise introduction to the paper you presented, including the problem background, key ideas, technical contributions and so on.
  2. - Give more detailed explanations of those two or more additional papers you read. Please avoid extensively copying text or photos from the original papers in your report. 
  3. - In the conclusion section of your report, you need to summarize and analyze these papers as well as the relevant research topics. Personal research experience and possible improvements you can think for those works should also be included in this section.
  4. - You need to write your report using the latex template provided and the final report should be longer than four pages.  

  5. Marking Criteria:  According to the report requirements above, your report grade will be assigned by instructors. Plagiarism is NOT allowed and will lead to course failure.
    How to submit: Send your report by email to ruizhen.hu@szu.edu.cn with subject in the form of “[Visual Computing] Report_Name_StudentID ”.
    Deadline: July 1st, 2017


Downloads

Creativity.zip

Papers.zip Report_template_en.zip Report_template_cn.zip Presentation_template.pptx
Fundamentals of Computer Graphics (4th Edition).pdf Paperlist.docx Report_evaluation.pdf Course_evaluation.pdf  


 

Textbook recommendation

VCC Introduction

Cutting cross multiple disciplines such as computer graphics, computer vision and visualization, Visual Computing Research Center (VCC) conducts deep research on 3D acquisition/understanding/modeling and visual analytics/computing, with applications in virtual reality, smart city, E-commercial, manufacture, education, art and entertainment.

VCC has a world-class research team with many of the researchers receiving their higher education from U.S., Canada, and Europe. The center enjoys close collaboration with leading universities and companies world-wide, playing a central role in bringing together research, education, industry, and government sectors that related to 3D technologies.

VCC publishes in top venues in all of the computer graphics, computer vision and visualization fields, such as ACM SIGGRAPH, ACM SIGGRAPH Asia, ACM TOG and IEEE TVCG. For more details, please visit ourResearch page.