5 Things that Will Never Be the Same After Computer Vision
It’s been almost 8 years since I started getting involved in the crazy world of Data Science.
I’ve been amazed hundreds of times with different technologies, their applications, and the impact they can have in the world around us.
But…
In my whole career, I fail to remember a piece of technology that is as disruptive as Computer Vision.
Why?
In a phrase:
I think Computer Vision can and will change the world.
I’m not one for writing articles about how much I love a piece of technology, but this time, I’ll make an exception.
Read along and you’ll get 3 great pieces of information:
- This great Computer Vision infographic our team has put together.
- 5 amazing applications to Computer Vision.
- A very special surprise for you, waiting at the end.
Without further ado… 5 Crazy Applications:
1. Self-Driving Cars:
Probably the most well-known application of Computer Vision.
We’ve all seen it: Uber, Google, Tesla. The dystopian future of self-driving cars is upon us.
How does it work?
Through Computer Vision, the car can easily recognize patterns on the road, such as lines on the pavement, other automobiles approaching or breaking, pedestrians crossing, etc. These images are then processed and the car will “react” accordingly.
How will this impact your life?
(A couple of stats that’ll blow your mind)
- Experts agree that there will be more than 10 million self-driving cars by 2020. 1 in 4 of every car being self-driven by 2030.
- Expectations are that self-driving cars will reduce accidents by a whopping 90% in the US alone. This will directly impact your annual car insurance cost, which is projected to go down by as much as $1,000.
- Self-Driving cars are expected to make mobility and parking more efficient. By how much? You will “get back” 50 minutes a day.
That’s 1. We’re just getting started.
2. Social Media and Marketing:
Let’s tackle this in two parts:
a) Social Media
b) Marketing
a. Social Media
Face Detection & Face Recognition algorithms are king here.
Facebook, Instagram, and Snapchat use them. I’m sure you’ve noticed.
When?
Selfies with Dog Ears…
That’s right.
Face Detection algorithms are used to ID the users face and adjust digital watermarks around it, so that when you move, the watermarks (dog ears) move around accordingly.
I have a question for you:
Have you uploaded a picture and Facebook automatically “knows” who’s there and tags the person?
If you answered “yes”, then the amazing technology behind this is in fact: Facial Recognition.
Now marketing…
b. Marketing
As if there wasn’t enough technology available for marketers to use on their products…
Drumroll….
Please welcome: Object Detection.
Currently being used in traffic control and surveillance, these techniques are now making a shift towards convenience stores.
Companies like Amazon with the Amazon Go's Just Walk Out Technology, a mix of Computer Vision and Deep Learning Algorithms, are making consumer’s shopping experience easier and more enjoyable, by eliminating lines and checkouts.
There are other applications as well.
Pinterest recently developed Lens, a Visual Discovery Tool similar in concept to Shazam.
What’s the difference then?
It doesn’t identify songs, it ID’s particular objects.
And the final example in Marketing is RealEyes or Unruly.
These crazy companies are using Emotion Recognition techniques to provide analytics and insights for businesses.
They use remote panels of users and connect them to webcams, to capture the real-time reaction of an individual towards a specific piece of content or product.
Marketing Researchers are probably drooling by now.
3. Robots
I’ve got two cool examples about Robots and Computer Vision:
- China and Japan are now using Computer Vision in robot waiters (Wait. What?!) Yes, yes – robot waiters use Computer Vision to recognize and interact with customers when they enter a store.
- The Dubai police force is deploying a life-sized robot police officer that uses Computer Vision to recognize criminals.
How is that even possible?
Well, in a nutshell, facial recognition works like this:
A face is detected from a video stream, based on the variation in pixel concentration, which then defines certain parameters that represent fixed points in a face: eyes, nose, and mouth.
Then, it can be recognized once these parameters are compared to thousands of images that the algorithm has access to. Finally, the robot establishes what the expressions might mean and even what the intent of the person may be.
If you’re thinking: “Wow, Robots are very cool”, then you’re reading my mind.
But, hey! What about humans then?
Don’t worry.
This next application is saving thousands of lives worldwide, as we speak.
4. Medicine
Some revolutionary technologies, like Microsoft’s InnerEye and IBM’s Watson Health Imaging, are reshaping the path of medical diagnosing.
Computer Vision software is being used to quickly process Medical Images and cross-reference them with thousands of malign formations.
Note that in this application there is no actual ‘vision’ or ‘sight’ involved – i.e. no cameras. Computer vision is used here as an algorithm that detects detect patterns and anomalies on Medical Images.
In a matter of seconds!
This way, doctors are able to accurately and efficiently ID potential threats to patients.
Amazing. Right?
5. Drones
This is one of my absolute favorites:
Drones.
Not only the military drones (which we won’t speak about here) but pocked sized ones.
These tiny pieces of state of the art technology are seriously packing!
Did you know? That…
- Through Computer Vision, a drone can figure out aspects of its surroundings to ensure it has a safe flight? It recognizes aspects like the floor, obstacles, hand-gesture control and user detection for a secure launch!
- They are becoming very relevant in other scenarios as well... For instance, the film industry or critical situations like natural disasters where rescuers use them to examine impacted zones.
That last application is a clear example of what I mean by changing the world.
By reshaping the way we approach technology and the way we view its applications, we can imagine new and amazing uses for it.
We just have to: 1) Know How they work. 2) Think of its applications. This way they can directly impact our lives in situations where the error of margin is close to zero.
There you go!
5 incredible applications of Computer Vision.
But…
Those are only 5 of the hundreds of applications that this technology can have.
Now, do you see why this will change the world?
Think of the possibilities!
Now I’m curious! Which was your favorite? Are you already involved with CV?
Please leave your comments and let’s share some great CV experiences!
And finally…
… Your surprise!
It’s somewhat related to drones but way more exciting.
Have I made enough pre-amble?
We launched an amazing Kickstarter that will explore the fantastic world of Computer Vision!
That’s right!
We are diving deep into CV and all its wonderful applications.
If you feel this is something you (or anyone around you) would want to explore, please check it out here: https://e.fnd.to/computervision
Happy Analyzing!
Kirill Eremenko & The SDS Team
References
Forbes.com (SDCs) & Forbes.com (Robots)
Applying Data Science to make work a more connected, human experience
6 年Great article!
7+ YoE in Business Intelligence
6 年Nice article! Few weeks back I completed your computer vision course on Udemy, which is a great course! Just wanted to ask do you plan to add more content to that course? And if you are open to suggestions would you consider adding some content about the applications of deep learning, specifically image recognition in medicine (point 4 in your article) ? That seems to be an area ripe for disruption using deep learning techniques and not to mention it is very exciting as well. Examples of its uses would be non invasive diagnosis, image guided radiotherapy. Would greatly appreciate it if you could provide any information or redirect me to some learning resources in this regard.
Sr. ML Scientist
7 年nice article