Fakultät Informatik

Master-Seminar – Machine Learning in Graphics

Lecturer

 N. Thuerey,  R. Westermann,  M. Niessner,  Kiwon Um

Studies

Master Informatics

Time, Place

Mondays 16:00-18:00, Seminarraum MI   02.13.010

Begin

24.04.2017

Content

In this course, students will autonomously investigate recent research about machine learning techniques in the computer graphics area. Independent investigation for further reading, critical analysis, and evaluation of the topic are required.

Requirements

  • The participants have to present their results in a talk, which should last 30 minutes. Talks will be given in English. The slides should be sent one week before the talk; otherwise, the talk will be canceled.
  • A short report (approximately 3-4 pages in  the ACM SIGGRAPH format) should be prepared and sent within two weeks after the talk.

Preliminary Schedule

24.03.2017

Deadline for sending an e-mail with 3 preferences

31.03.2017

Notification of assigned paper

Papers

Id

Paper

01

2016, Dong et al.,  Image Super-Resolution Using Deep Convolutional Networks, IEEE Transactions on Pattern Analysis and Machine Intelligence

02

2017, Dahl et al.,  Pixel Recursive Super Resolution, arXiv.org

03

2016, Yan et al.,  Automatic Photo Adjustment Using Deep Neural Networks, ACM Trans. Graph.

04

2015, Gryka et al.,  Learning to Remove Soft Shadows, ACM Trans. Graph.

05

2014, Goodfellow et al.,  Generative Adversarial Networks, Advances in Neural Information Processing Systems 27 (NIPS 2014)

06

2016, Ruder et al.,  Artistic Style Transfer for Videos, arXiv.org
(optional) 2015, Gatys et al.,  A Neural Algorithm of Artistic Style, arXiv.org

07

2016, Ren et al.,  Image Based Relighting Using Neural Networks, ACM Trans. Graph.

08

2015, Kalantari et al.,  A Machine Learning Approach for Filtering Monte Carlo Noise, ACM Trans. Graph.

09

2017, Zheng and Zheng,  NeuroLens: Data-Driven Camera Lens Simulation Using Neural Networks, Computer Graphics Forum

10

2015, Ladický et al.,  Data-driven Fluid Simulations Using Regression Forests, ACM Trans. Graph.

11

2017, Kim et al.,  Category-Specific Salient View Selection via Deep Convolutional Neural Networks, Computer Graphics Forum

12

2015, Mnih et al.,  Human-Level Control Through Deep Reinforcement Learning, Nature

13

2015, Guo et al.,  3D Mesh Labeling via Deep Convolutional Neural Networks, ACM Trans. Graph.

14

2016, Simo-Serra et al.,  Learning to Simplify: Fully Convolutional Networks for Rough Sketch Cleanup, ACM Trans. Graph.

15

2016, Nishida et al.,  Interactive Sketching of Urban Procedural Models, ACM Trans. Graph.

16

2016, Zeng et al.,  3DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructions, arXiv.org

17

2016, Peng et al.,  Terrain-adaptive Locomotion Skills Using Deep Reinforcement Learning, ACM Trans. Graph.

18

2016, Holden et al.,  A Deep Learning Framework for Character Motion Synthesis and Editing, ACM Trans. Graph.

References

 

News

Matthias Niessner, our new Professor from Stanford University, offers a number of interesting topics for  master theses.

 

A new PhD/PostDoc position on  Computational Fabrication and 3D Printing is available at the Computer Graphics & Visualization group.

 

A new PhD position is available at the games engineering group.  Check it out here.