Advanced Computer Graphics

CS05145 高级计算机图形学

时间:周六 14:00-16:40
教室:3C202

教师:黄章进

论文报告顺序

2017年12月23日8:30-,电3楼632会议室:

  1. SF17011006 何良杰 & SF17011019 武胜 VNect: Real-time 3D Human Pose Estimation with a Single RGB Camera
  2. SF17011014 孙波 Real-time Geometry, Albedo and Motion Reconstruction Using a Single RGBD Camera
  3. SF17011009 李世栋 A Deep Learning Approach for Generalized Speech Animation
  4. SF17011001 柴业钦 & SF17011022 许心悦 Synthesizing Obama: Learning Lip Sync from Audio
  5. SF17011011 马帅奇 & SF17011015 孙晓东 DeepSketch2Face: A Deep Learning Based Sketching System for 3D Face and Caricature
  6. SF17011013 潘朝阳 Phace: Physics-based Face Modeling and Animation
  7. SF17011008 李盼 & SF17011026 张孝天 Learning a model of facial shape and expression from 4D scans
  8. SF17011016 王广玉 Example-Based Synthesis of Stylized Facial Animations
  9. SF17011025 张杰 Bringing Portraits to Life
  10. SF17011010 罗斌 & SF17011030 卓圣洁 Real-time 3D Eyelids Tracking from Semantic Edges
  11. SF17011005 管海 & SF17011020 夏青 Online Generative Model Personalization for Hand Tracking
  12. SF17011027 章翔 ClothCap: Seamless 4D Clothing Capture and Retargeting
  13. SF17011003 单兴兴 DeepLoco: Dynamic Locomotion Skills Using Hierarchical Deep Reinforcement Learning
  14. SF17011007 姜钰爽 & SF17011012 马勇 GRASS: Generative Recursive Autoencoders for Shape Structures
  15. SG13011409 叶勇 Data-Driven Synthesis of Smoke Flows with CNN-based Feature Descriptors
  16. SG15011002 吴勇 & SG15011061 陈芳 BendSketch: Modeling Freeform Surfaces Through 2D Sketching
  17. SF17011021 邢佑君 & SF17011023 余潮海 VoCo: Text-based Insertion and Replacement in Audio Narration
  18. SG15011059 谢伟 Computational Zoom: A Framework for Post-Capture Image Composition
  19. SF17011029 朱检检 Unmixing-Based Soft Color Segmentation for Image
  20. SF17011017 王晖 Globally and Locally Consistent Image Completion
  21. SF17011002 陈健鹏 Deep Bilateral Learning for Real-Time Image Enhancement
  22. SF17011018 吴启雄 Real-Time User-Guided Image Colorization with Learned Deep Priors
  23. SF17011028 郑玉山 Deep High Dynamic Range Imaging of Dynamic Scenes
  24. SF17011024 张华利 HDR image reconstruction from a single exposure using deep CNNs

 


序号

学号

姓名

论文题目

1 SF17011001 柴业钦    

 

18. Synthesizing Obama: Learning Lip Sync from Audio
2 SF17011002 陈健鹏    

 

25. Deep Bilateral Learning for Real-Time Image Enhancement
3 SF17011003 单兴兴    

 

DeepLoco: Dynamic Locomotion Skills Using Hierarchical Deep Reinforcement Learning
4 SF17011004 顾辉    

 

未入学
5 SF17011005 管海    

 

39. Online Generative Model Personalization for Hand Tracking
6 SF17011006 何良杰    

 

1. VNect: Real-time 3D Human Pose Estimation with a Single RGB Camera
7 SF17011007 姜钰爽    

 

8. GRASS: Generative Recursive Autoencoders for Shape Structures
8 SF17011008 李盼    

 

37. Learning a model of facial shape and expression from 4D scans
9 SF17011009 李世栋    

 

16. A Deep Learning Approach for Generalized Speech Animation
10 SF17011010 罗斌    

 

43. Real-time 3D Eyelids Tracking from Semantic Edges
11 SF17011011 马帅奇    

 

29. DeepSketch2Face: A Deep Learning Based Sketching System for 3D Face and Caricature
12 SF17011012 马勇    

 

同 姜钰爽
13 SF17011013 潘朝阳    

 

31. Phace: Physics-based Face Modeling and Animation
14 SF17011014 孙波    

 

2. Real-time Geometry, Albedo and Motion Reconstruction Using a Single RGBD Camera
15 SF17011015 孙晓东    

 

同 马帅奇
16 SF17011016 王广玉    

 

33. Example-Based Synthesis of Stylized Facial Animations
17 SF17011017 王晖    

 

22. Globally and Locally Consistent Image Completion
18 SF17011018 吴启雄    

 

26. Real-Time User-Guided Image Colorization with Learned Deep Priors
19 SF17011019 武胜       同 何良杰
20 SF17011020 夏青    

 

同 管海
21 SF17011021 邢佑君       19. VoCo: Text-based Insertion and Replacement in Audio Narration
22 SF17011022 许心悦       同 柴业钦
23 SF17011023 余潮海       同 邢佑君
24 SF17011024 张华利       44. HDR image reconstruction from a single exposure using deep CNNs
25 SF17011025 张杰       42. Bringing Portraits to Life
26 SF17011026 张孝天       同 李盼
27 SF17011027 章翔       15. ClothCap: Seamless 4D Clothing Capture and Retargeting
28 SF17011028 郑玉山       Deep High Dynamic Range Imaging of Dynamic Scenes
29 SF17011029 朱检检       10. Unmixing-Based Soft Color Segmentation for Image
30 SF17011030 卓圣洁       同 罗斌
31 SG13011409 叶勇    

 

11. Data-Driven Synthesis of Smoke Flows with CNN-based Feature Descriptors
32 SG15011002 吴勇    

 

28. BendSketch: Modeling Freeform Surfaces Through 2D Sketching
33 SG15011059 谢伟    

 

5. Computational Zoom: A Framework for Post-Capture Image Composition
34 SG15011061 陈芳

 

 

 

同 吴勇

 


论文报告 列表

时间:待定

要求:

  1. 自行分组报告论文,每组不超过2人,申领论文时把组员学号和姓名一起发到老师邮箱
  2. 论文选自Siggraph 2017 (http://kesen.realtimerendering.com/sig2017.htmlACM Digital Library )共34篇,和Siggraph Asia 2017(http://kesen.realtimerendering.com/siga2017Papers.htm)共10篇(具体见论文列表),每组选择一篇感兴趣的论文报告。
  3. 为解决选择相同论文的冲突,请大家在email中提供按优先级排列的拟报告论文列表。 根据先到先得原则按发email的时间顺序来确定报告论文列表。 如果所选论文都已被认领,会告知再次提供一份拟报告论文列表。只有在所有组确定完所报告的论文后,才能更换已确定的论文。
  4. 每篇论文报告时间为10-15分钟(包括展示video),讨论2-3分钟时间。
  5. 由于每篇论文的报告时间有限,比较短,因此报告的重点在于:简单介绍作者和单位情况; 主要讲一下这篇文章在做什么(What)?为什么做这个问题(motivation-Why)?这个问题以前有些什么工作? 难点在哪里?这篇文章解决了什么样的难点或提出什么样的新方法(selling points)?局限和未来工作(limitation and future work)? 细节若太多不用展开,关键要让大家知道这篇文章的主要思想和价值所在。
  6. 请大家尽量在网上去找关于你所负责报告论文的相关资料。如各个作者的主页,项目网页。 很多作者会把论文、video、甚至将PPT放在他的主页。讲解论文时,论文中的图片、视频均播放给大家看一下。 “一图胜千言”,视频又能胜于图片。
  7. 书面报告要求采用科研论文格式(标题、摘要、关键字、正文、参考文献等)撰写, 内容可以是所报告论文的概览、同主题多篇论文的综述等。
  8. 评分:PPT口头报告 + 书面报告

 

 

Last update : 12/20/2017