First, what is computer vision and why is it one, so difficult and two, so important?
Short answer: Computer vision is incredibly fascinating (OBJECTIVELY), as you will see in this following article, in terms of being relevant to current and future technological applications and markets, and interesting in terms of the psycho-philosophical meaning behind what vision actually is and what is entails.
In essence, computer vision (CV) can be reduced to this quote from a TED talk by Dr. Fei-Fei Li, the foremost contributing computer scientist and academic on computer vision (5/23/2015):
To give a better understanding of the idea of computer vision, we really need to break down the idea behind what we mean by vision. And then, we can start talking about what we’re trying to replicate using computer vision.
So, what is vision? After a cursory perusal on BOS (Best Online Source), I’m inclined to suggest this definition:
‘Vision’ can be defined as the actual seeing and identification of visual images, and touches upon the processing of imaging and preparation for immediate reactions based on seen and identified visual images.
And… I think that’s a pretty good place to start.
So, in conjunction, here is a super interesting problem right now in computer vision:
Basically, as seeing humans, we mostly intuitively understand and process what we are seeing based on 100’s upon 1000’s of visual experiences that we collect from birth to childhood and onwards. This is a skill that is acquired through our human processors — the brain (mostly, TBD) and of course, the human body.
And in computer vision, we are trying to simplify this process, and also, eventually replicate this process, I imagine.
Anyhow, moving forward, who are the major players who are invested in computer vision?
> literally any self respecting cutting edge tech company
> and .. any technologically oriented research and/or academic setting (think tanks, universities, military, incognito labs, MI6 .. etc)
> perhaps, eventually, your thoughts.. and definitely this blog! (follow me for updates and information, please and thanks)
Moreover, who are the major academic labs and current sources of knowledge in computer vision?
- MIT CSAIL Computer Vision Research Group
- Berkeley Artificial Intelligence Research (BAIR)
- CI2CV Computer Vision Lab
- Computational Vision at Caltech
- UCLA Vision Lab
- HAL@Duke *this one is of personal opinion
(Source: https://deepomatic.com/en/top-5-computer-vision-labs-to-discover)
And so, what next if you and I are interested in learning more about hands on computer vision?
I’m really glad you’ve made it this far. I would definitely suggest that you go through this intro course on convolutional neural networks (the technology used at the current forefront of computer vision), taught by Dr. Fei-Fei Li herself, other professors, and some Stanford TAs, as a gentle way to get hands on experience with computer vision and how it is currently being studied.
Good luck, congrats on your explorations and interest, and let me know how the work goes!
