ETH Zurich - D-INFK - IVC - CVG - Lectures - Deep Learning Seminar

Deep Learning Seminar


Course title: Deep Learning for Computer Vision: Seminal Work
Course ID: 263-5904-00L
Lecturers: Dr. Iro Armeni
Teaching assistants: Daniel Thul, and Sebastian Beetschen, Luca Cavalli, Mihai Dusmanu, Cathrin Elich, Taein Kwon, Sandro Lombardi, Rémi Pautrat, Denys Rozumnyi, Paul-Edouard Sarlin, Silvan Weder
Venue: Mo 16-18h, offered online (students will receive information about joining)

This seminar covers seminal papers on the topic of deep learning for computer vision. The students will present and discuss the papers and gain an understanding of the most influential research in this area - both past and present. The objectives of this seminar are two-fold. Firstly, the aim is to provide a solid understanding of key contributions to the field of deep learning for vision (including a historical perspective as well as recent work). Secondly, the students will learn to critically read and analyse original research papers and judge their impact, as well as how to give a scientific presentation and lead a discussion on their topic.

Each student chooses one paper from the provided collection to present during the course of the seminar. The students will be supported in the preparation of their presentation by the seminar assistants.

News

  • 24.02. Paper assignment posted.
  • 22.02. Introductory class.

Important information

  • Class Material: Class material will be posted on Moodle (link). We will also use Moodle to submit assignments.
  • Assignments: Each student will present one paper. Everybody is encouraged to read each paper before it is being presented and engage in a discussion following the presentations. To foster interesting discussions, each paper will also be assigned two "critics" who study the paper and prepare questions for the discussion.
  • Grading: Each student will be graded based on both their presentation (80%) and their participation in the assigned discussions (20%).
  • Attendance: Attendance is required to pass the course (3 absences allowed).
  • The class is held online.

Slides

To be announced

Schedule

Date Topic Presenter Critics Assistant
22.02. Introduction
01.03. #01. ImageNet Classification with Deep Convolutional Neural Networks Menelaos K. Dominik A., Elham A. Daniel Thul
01.03. #02. U-Net: Convolutional Networks for Biomedical Image Segmentation N/A Martin B. Iro Armeni
08.03. #03. Deep residual learning for image recognition Dominik A. Menelaos K., Weirong C. Denys Rozumnyi
08.03. #04. MobileNetV2: Inverted Residuals and Linear Bottlenecks Benjamin J. Daniele C. Daniel Thul
15.03. #05. Attention Is All You Need Wenhao X. David N., Dominik A. Taein Kwon
15.03. #06. 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks Dominik H. Amir R., Jiaqi C. Iro Armeni
22.03. #08. Dropout: a simple way to prevent neural networks from overfitting Eshaan M. Martin B., Elena I. Luca Cavalli
29.03. #09. A Metric Learning Reality Check Amir R. Elias R., Menelaos K. Mihai Dusmanu
29.03. #10. MentorNet: Learning Data-Driven Curriculum for Very Deep Neural Networks on Corrupted Labels Martin B. Mihai B., Soley H. Sebastian Beetschen
05.04. No Class
12.04. #11. Visualizing and Understanding Convolutional Networks Elena I. Zhiyin Q., Tiancheng H. Luca Cavalli
12.04. #12. Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation Daniele C. Benjamin J., Amir R. Daniel Thul
12.04. #13. Conditional Random Fields as Recurrent Neural Networks Jiaqi C. Soley H., Paula W. Iro Armeni
19.04. No Class
26.04. #14. Learning Convolutional Neural Networks for Graphs Mihai B. Wenhao X., Zhiyin Q. Denys Rozumnyi
26.04. #15. A Style-Based Generator Architecture for Generative Adversarial Networks Elham A. Dominik H., Eshaan M. Rémi Pautrat
26.04. #16. Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks Tiancheng H. Jiaqi C., Shengqu C. Cathrin Elich
03.05. #16B. Semantic Image Synthesis with Spatially-Adaptive Normalization Elias R. Anuj P., Elena I. Paul-Edouard Sarlin
03.05. #17. DSAC-differentiable RANSAC for Camera Localization Kouroche B. Elham A., Benjamin J. Paul-Edouard Sarlin
03.05. #18. SuperPoint: Self-Supervised Interest Point Detection and Description Soley H. Eshaan M., Daniele C. Rémi Pautrat
10.05. #19. DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation Paula W. Weirong C., Mihai B. Silvan Weder
10.05. #20. NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis Shengqu C. Tiancheng H., Kouroche B. Silvan Weder
17.05. #21. AtlasNet: A Papier-Mâché Approach to Learning 3D Surface Generation Zhiyin Q. Paula W., Dominik H. Sandro Lombardi
17.05. #22. BSP-Net: Generating Compact Meshes via Binary Space Partitioning Weirong C. Anuj P., Xuran L. Paul-Edouard Sarlin
24.05. No Class
31.05. #23. Unsupervised Learning of Probably Symmetric Deformable 3D Objects from Images in the Wild Xuran L. Shengqu C., Wenhao X. Cathrin Elich
31.05. #24. Unifying Deep Local and Global Features for Image Search Anuj P. Kouroche B., Elias R. Mihai Dusmanu

© CVG, ETH Zürich webmaster: psarlin@inf.ethz.ch