Graph Learning and its Application in Robotics

Presentation by Dr. Chen Wang as part of the Air Lab Summer School 2020.
Sessions list, overviews, and links to repos: https://theairlab.org/summer2020

This tutorial introduces the basics of Graph Neural Networks and its application to robotics such as feature matching. The tutorial presents algorithms presented in the following papers.

GCN, ICLR 2017: https://arxiv.org/pdf/1609.02907.pdf
SAGE, NIPS 2017: https://arxiv.org/pdf/1706.02216.pdf
GAT, ICLR 2018: https://arxiv.org/pdf/1710.10903.pdf
SuperGlue, CVPR 2020 https://arxiv.org/pdf/1911.11763.pdf
APPNP, ICLR 2019 https://arxiv.org/pdf/1810.05997.pdf

Chapters:
00:00 Intro
02:38 Basics of Graph
10:52 Graph Convolutional Network(GCN)
28:35 Graph Attention Network (GAT)
43:33 Graph Sampling (SAGE)
52:02 Application in feature matching (SuperGlue)
01:03:30 Assignment: implementation of APPNP

Slides: https://entuedu-my.sharepoint.com/:p:...
Exercise: https://classroom.github.com/a/h3ktCkJ5

Air Lab Website: https://theairlab.org
Twitter:   / airlabcmu  
Facebook:   / airlabcmu  
Medium:   / airlabcmu