Hello trend,
This is Satya Mallick from LearnOpenCV.com.
We shared several videos from our courses during the Thanksgiving course sale to give you a sense of the content we cover.
- The theory of image formation
- How color is recorded by a digital camera.
- Bias-Variance Tradeoff in Machine Learning.
- Good Practices for Training Neural Networks.
- Demystifying Neural Networks
During this holiday sale, we have been sharing stories that show us the breadth of computer vision as a field.
A few days back, we learned how the work of a computer vision grad student, Katie Bouman, made the first image of a black hole possible.
Today, we will look Computer Vision applied to reconstructing protein macrocules using Cryogenic Electron Microscopy (Cyro-EM).
That's literally the other end of the spectrum!
Before we jump into today's newsletter, here is a quick reminder that OpenCV Holiday Sale ends at the stroke of midnight, Pacific Time. We have a 20% off on all courses and bundles and you get an additional 5% discount with coupon code YES5 at checkout.
| BEGIN YOUR AI JOURNEY |
Not Interested? Click on this link to stop future Holiday Sale emails.
Vision in the Small
Most people are familiar with computer vision applications in industrial automation, safety/security, manufacturing, agriculture, medical diagnostics, etc.
The impact of computer vision is much broader.
I published a few papers in the field of Cryo-EM during my Ph.D. It is a technique used to solve the 3D structure of protein macromolecules.
The wavelength of electrons is much shorter than the wavelength of photons. Therefore, an Electron Microscope that uses a beam of electrons can resolve finer details in microscopic samples than an optical light microscope.
In the 1970s, researchers developed Cryo-EM's practical and theoretical foundation. They freeze samples of protein macromolecules using liquid ethane and then photograph them using an Electron Microscope.
How cool is that?
The field was still emerging when I published a paper titled Detecting particles in cryo-EM micrographs using learned features in the Journal of Structural Biology (2004).
It essentially implemented Haar Cascades (Viola & Jones) to solve an object detection problem in the field. I worried about the scientific contribution of the paper because I merely took a standard algorithm in computer vision and applied it to a new field.
The reviewers, on the other hand, thought my paper did considerable service to the Cryo-EM community by bringing a state-of-the-art vision algorithm to the field.
They awarded my paper the cover of that Journal of Structural Biology issue, a prestigious honor for a grad student.
Sometimes you contribute by being an inventor. At other times you contribute by being a messenger.
I also worked on one of the most challenging problems in the field with my collaborators -- 3D reconstruction of protein macromolecules using Cryo-EM images.
One evening when we got some excellent results, I remember my advisor, Prof. David Kriegman, telling me that anybody in Cryo-EM who develops algorithms for reconstructing protein macromolecule at atomic resolution (4 Angstroms) could win a Nobel Prize.
I did not quite believe him, and I never worked in the area after my Ph.D. as I thought working in a fledgling research area like Cryo-EM would not be a good career choice.
In 2017, Jacques Dubochet, Joachim Frank, and Richard Henderson won the Nobel prize in Chemistry for their pioneering work in Cryo-EM.
Computer Vision and Deep Learning algorithms now dominate Cryo-EM. In retrospect, working in Cryo-EM would have been an equally good career choice.
In many new fields, Computer Vision and AI, techniques are yet to be widely adopted. It takes time for ideas to move to adjacent areas.
The messengers who take ideas learned in Computer Vision and AI and apply them to these new fields will have rewarding careers.
The AI revolution has just started, and it is having a massive impact across many industries.
| JOIN THE AI REVOLUTION |
Cheers!
Satya
Courses / Webinar / Jobs
Facebook / LinkedIn / Twitter / Instagram
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.