abrliu / reading_group

Virginia Tech Vision and Learning Reading Group

Virginia Tech Vision and Learning Reading Group

Time

• Reading Group: Thursday 2PM - 3PM, Whittmore Hall 654
• Study Group: Friday 2PM - 3PM, Whittmore Hall 457B

Table of Contents

• Current Schedule (Fall 2019)
• Previous Meetings
• Mailing List
• Presenter
• Related Links
• FAQ
• About Us
• Suggested Papers

Current Schedule

Reading Group: Thursday 2PM - 3PM, Whittmore Hall 654

Please let Jinwoo know what paper you are going to present and your summary by the end of Sunday of your presentation week.

Date	Presenter	Topic
09/05	Joseph Messou	Fashion Retrieval via Graph Reasoning Networks on a Similarity Pyramid [Kuang et al., ICCV2019] [slides]
09/12	Jinwoo Choi	Temporal Attentive Alignment for Large-Scale Video Domain Adaptation [Chen et al., ICCV2019 Oral] [slides]
09/19	Esther Robb	AutoGAN: Neural Architecture Search for Generative Adversarial Networks [Gong et al., ICCV2019] [slides]
09/26	Yuliang Zou	Learning Spatial Common Sense with Geometry-Aware Recurrent Networks [Tung et al. CVPR2019 Oral] [slides]
10/03	Chen Gao	Learning Correspondence from the Cycle-consistency of Time [Wang et al., CVPR2019] [slides]
10/10	Xiaolong Li	Self-Supervised Correspondence in Visuomotor Policy Learning [Peter et al.]
10/17	Shih-Yang Su	MCP: Learning Composable Hierarchical Control with Multiplicative Compositional Policies [Peng et al., NeurIPS2019] [sides]
10/24	Badour AlBahar	StructureFlow: Image Inpainting via Structure-aware Appearance Flow [Ren et al., ICCV2019]
10/31	Si Chen	GA-Net: Guided Aggregation Net for End-to-end Stereo Matching [Zhang et al., CVPR2019] [slides]
11/07	-	No meeting - CVPR2020 deadline
11/14	-	No meeting - CVPR2020 deadline
11/21	-	No meeting - CVPR2020 supplementary deadline
11/28	-	No meeting - Happy Thanksgiving!
12/05	Jiarui Xu	TBD

Study Group: Friday 2PM - 3PM, Whittmore Hall 457B

Date	Instructor	Topic
09/06	Yann LeCun	Deep Learning, Convolutional Neural Networks, and Self-Supervised Learning
09/13	Oriol Vinyals	DeepMind AlphaStar, StarCraft, and Language
09/20	Peter Bartlett	Geometry of Deep Learning [Day 1 Part 1]
09/27	Yisong Yue	New Frontiers in Imitation Learning
10/04	N/A	Cancelled due to Aayush Bansal's visit
10/11	Jitendra Malik	AI Panel: Collins, Jordan, LeCun, Malik, Shenker
10/18	Jure Leskovec	Graph Representation Learning
10/25	Eric Jang	Tutorial on Normalizing Flows
11/01	Geoffrey Hinton	Andrew Ng interviews Geoffrey Hinton
11/08	-	No meeting - CVPR2020 deadline
11/15	-	No meeting - CVPR2020 deadline
11/22	TBD	TBD
11/29	-	No meeting - Happy Thanksgiving!
12/06	TBD	TBD

Previous Meetings

• Fall 2017
• Spring 2018
• Fall 2018
• Spring 2019

Mailing List

We use Google Groups to manage the mailing list: (link). You can click "Join Group" when you sign in with your Virginia Tech account.

Presenter

Please let Jinwoo know what paper you are going to present and your summary by the end of Sunday of your presentation week. Also, please send your slides or a link to them to Jinwoo once you finish preparing them.

• Badour AlBahar
• Si Chen
• Jinwoo Choi
• Chen Gao
• Xiaolong Li
• Joseph Messou
• Esther Robb
• Shih-Yang Su
• Jiarui Xu
• Yuliang Zou

Please contact Jinwoo if you want to be a presenter this semester!

Related Links

Resources

• Awesome Computer Vision
• Awesome Deep Vision
• Computer Vision Foundation open access

Similar reading group/seminars in other universities

• MIT Vision Seminars
• UIUC Vision Lunch
• UT-Austin CV Reading Group
• CMU VASC Seminar Series
• CMU ML Reading Group

Advanced CV courses

• Advanced Computer Vision (Jia-Bin Huang, Virginia Tech)
• Object and Activity Recognition Seminar (Trevor Darrell, UC Berkeley)
• Visual Learning and Recognition (Abhinav Gupta, CMU)
• Visual Recognition (Kristen Grauman, UT Austin)
• Advanced Computer Vision (Devi Parikh, Georgia Tech)
• Cutting-Edge Trends in Deep Learning and Recognition (Svetlana Lazebnik, UIUC)

FAQ

How is the presenters' order generated?

The presenters' order is generated from the presenters' list in a FIFO manner (but the list is initially generated randomly).

Who is responsible if I can not present at the scheduled time?

Yourself.

What should I do if I can not present at the scheduled time?

As early as possible, let the group organizer (ylzou -at- vt.edu) know about your situation. Also contact other presenters to see if they are willing to swap dates with you.

I have a question not listed here...

Then ask by sending an e-mail to the mailing list (vt-vision-and-learning-reading-group-g -at- vt.edu).

About Us

How it works?

We are a group that meets about once a week to discuss one to three relevant papers. For every meeting, one person will be in charge of selecting the paper(s) for that meeting, thoroughly understanding the work, and leading the discussion (either informally or via a presentation, whatever the leader thinks is best). The rest of the members will read over the paper(s) beforehand to gain a basic idea of the work. Then, on the day of the meeting, we will discuss the strengths, weaknesses, and techniques of the paper(s).

NOTE: Please tell the group organizer (chengao -at- vt.edu) which paper(s) you are going to present, and summarize the paper/talk in several sentences, before Monday of that week.

What we read?

We will be reading papers appearing in the leading computer vision conferences (e.g., CVPR, ICCV, ECCV, SIGGRAPH, SIGGRAPH Asia) and machine learning conferences (e.g., NIPS, ICML, ICLR, UAI, AAAI, IJCAI, AISTATS). Members are free to choose which paper(s) they will present (we can also provide suggestions), thus the specific topics will vary based on the members' interests.

Who can join?

We are open to everyone who is interested, whether you are an undergrad, a grad student, or VT staff, regardless of department. Anyone else in the Blacksburg area is also welcome. As long as you are interested in learning more about the fields (by reading cutting-edge research papers), you are welcome to join.

Suggested Papers

We maintain a pool of suggested papers here.

Credits: The contents and formats were modified from VT CVMLP Reading Group.