UC San Diego Tsinghua University CVPR2022 "Artificial Intelligence" tutorial

Source: Specialty

This article is a tutorial, it is recommended to read for 5 minutes

Several scholars from UC San Diego Tsinghua University told the tutorial of "Artificial Intelligence", and the 200+pages PPT is worthy of attention.

In recent years, people have become more and more interested in artificial intelligence research in computer vision. A number of embedded AI seminars and challenges have been held in the research community, including the broad strategy learning of ICLR 2022, IROS 2020 OCRTOC: Open Cloud Robot Watch Organization Challenge, CVPR 2019 habitat: Embedded Agent Challenges and seminars, as well as embedded AI seminars in 2020 and 2021. Computer vision is now an important module of artificial intelligence research, but we still lack a basic tutorial to guide researchers, especially those researchers with visual and machine learning backgrounds to begin in this field.

In particular, with the latest progress of physical simulation and rendering technology, the body AI in the virtual environment has made many impressive progress. These platforms make many visual robot problems possible, and these problems cannot be conducted in a large scale in the real world. Fast speed, easier parallelization, simpler data collection and lower cost essence to allow embedded AI studies in simulation to establish larger communities, with different researcher backgrounds, improved code sharing and standards The benchmark. However, the virtual environment also has its own problems, such as simulation parameters and domain gaps, which are worth noting when building and using them.

Our tutorial aims to provide entry guidelines for computer vision researchers to study visual issues of body agents in the environment, and highlighting common problems encountered when using these environments. This tutorial will focus on the principles of cross -platform sharing and teach the concept of multiple simulation environments.

https://ai-workshops.github.io/building-nd-rking-nvired

Speaker:

Directory content:

- END -