Check out our latest research on weakly-supervised 3D shape completion.


t-SNE, 3D Vision and Being a Good CVPR Citizen — Notes from CVPR’18

Last week, I attended my very first CVPR in Salt Lake City, where I also presented my work on weakly-supervised 3D shape completion. In the course of the week, I attended several tutorials as well as all oral and poster sessions. In this article, I want to share my notes and some general comments.

As hard-working PhD student, I was well-prepared to take notes at last week's CVPR'18. While my choice of taking notes directly in LaTeX might not have been optimal, I was able to write down the most important points of most talks, spotlights and tutorials I attended. Additionally, it allows to easily share my notes as PDF.

My personal highlights were definitely the two tutorials on interpretability, held on Monday, and the panel on being a "Good Citizen of CVPR":

Regarding interpretability, I can recommend two talks in particular: Been Kim's "Introduction to Interpretable Machine Learning" and Laurens van der Maaten's "Dos and Don'ts of using t-SNE to Understand Vision Models".

From Friday's panel on good CVPR-citizenship, I can highly recommend Bill Freeman's and Jitendra Malik's opinions on good papers, Sven Dickinson's ideas on mentorships, Vladlen Koltun's view on good research and David Forsyth's overview of the CVPR community.

The notes found below are not necessarily complete or correct. Please refer to the corresponding papers or talks for details. Furthermore, the expressed views are my personal ones and are not intended to offend anybody; they also do not reflect the views of my employer.

Download Notes (or browse below)

What is your opinion on this article? Did you find it interesting or useful? Let me know your thoughts in the comments below or get in touch with me: