How Do Humans Do It? | Introduction

แชร์
ฝัง
  • เผยแพร่เมื่อ 17 มิ.ย. 2024
  • First Principles of Computer Vision is a lecture series presented by Shree Nayar who is faculty in the Computer Science Department, School of Engineering and Applied Sciences, Columbia University. Computer Vision is the enterprise of building machines that “see.” This series focuses on the physical and mathematical underpinnings of vision and has been designed for students, practitioners, and enthusiasts who have no prior knowledge of computer vision.

ความคิดเห็น • 8

  • @kenjiarimura4522
    @kenjiarimura4522 5 หลายเดือนก่อน +1

    I ended here in this amazing channel because recently had an idea of a patent, but the professor Nayar Shree had already registered 20 years ago. 😅

  • @rendermanpro
    @rendermanpro 2 ปีที่แล้ว +6

    Amazing channel! Thanks for sharing

  • @marcusravi1684
    @marcusravi1684 2 ปีที่แล้ว +3

    God bless you, Sir.
    Lots of respect from Bangalore.

  • @ivanov67
    @ivanov67 2 ปีที่แล้ว

    Overall, human vision is still superior to what computer software can do. Human vision is optimal for survival, and that's what really matters.

  • @ivanov67
    @ivanov67 2 ปีที่แล้ว

    Oftentimes, human brain uses situational context when processing visual information, and that can help a lot. Computers can't do that yet.

  • @narendraparmar1631
    @narendraparmar1631 ปีที่แล้ว

    Great😄

  • @indio007
    @indio007 2 ปีที่แล้ว +3

    I think you overstate how human vision is flawed. It takes up a significant portion of this video.
    If you notice, all the examples are man-made tricks created for the very purpose of fooling us.
    Nature does not do this. It is not out tricking us.
    When vision is compared to the mind ,(and by extension rationality and reason) it makes exponentially less errors.
    Think about how many mental errors you make daily versus how many times you make errors in sense perception.
    The sums are not even close.
    There is also the part where you say machines are far more precise and less error prone in some regards following you pointing out are inability to quantify absolute values of length.
    Machines can't do this either with out using multiple instruments or without a priori knowledge of absolute distance.
    I should also add that it should come as no surprise that man-made inventions can ultimately do better at some tasks using properties that are also man-made.
    Absolute values (arb units) are also a man-made construct..
    Knowing them is nearly useless in the natural environment and are some we created not for survival but while we had a condition of leisure.
    Regardless, I love this channel and the videos. Thx for making them.