Lex Fridman Podcast

#110 – Jitendra Malik: Computer Vision

Jul 21, 2020
Jitendra Malik, a distinguished professor at UC Berkeley and a pioneer in computer vision, shares his insights on the complexities of replicating human visual perception. He discusses the challenges of Tesla's Autopilot, emphasizing the gap between human and computer processing. Malik explores how integrated approaches and knowledge schemas can enhance action recognition. He critiques current evaluation methods, advocating for measures that reflect true understanding. Additionally, he highlights the importance of interdisciplinary research and the need for children's experiences in AI development.
Ask episode
AI Snips
Chapters
Books
Transcript
Episode notes
INSIGHT

Vision's Deceptive Simplicity

  • Computer vision is difficult because we do it subconsciously, making it seem simpler than it is.
  • Early AI researchers underestimated its complexity due to this effortless nature of human vision.
INSIGHT

Fallacy of the First Step

  • The "fallacy of the successful first step" describes how initial progress in vision can be misleadingly easy.
  • Achieving high accuracy becomes exponentially harder, requiring significantly more effort over time.
ANECDOTE

Tesla's Autopilot

  • Tesla's Autopilot, using a vision-based system with eight cameras and a neural network, tackles autonomous driving.
  • Jitendra Malik believes freeway driving is solvable, but full autonomy requires more than just vision.
Get the Snipd Podcast app to discover more snips from this episode
Get the app