Episode #30: Computer vision

In this episode, we discuss computer vision, distinguishing it from computer graphics, and delve into the challenges of 3D computer vision. We also examine the current state of autonomous driving research and discuss deepfakes.

We host two experts in the field:

• Žan Gojčič a senior research scientist at Nvidia, with a research focus is on 3D deep learning, weakly supervised learning, and domain adaptation.
• Aljoša Osep a senior research scientist at Nvidia, working on learning to understand the dynamic world from raw and unlabeled streams of sensory data.

Timestamps:

1:30: What initially drew you to the field you are currently in, and what continues to motivate you in your research? 

6:35: Studying abroad

14:38: Why did you decide to go into industry instead of staying in academia?

20:02: What is computer vision (and the difference between computer vision and computer graphics) and what are some common applications of computer vision in today's world?

24:10: 3D computer vision and What needs to be “synthesized” to create a realistic scene?

27:09: Challenges in the filed

20:31: Current state of research in autonomous driving and challenges 

42:50: Deepfakes

49:34: Emerging trends, open problems

56:18: Book recommendations and closing words