How to use CV techniques that can change how you see the world
Introducing Computer Vision
If you are looking for a way to improve your resume, this is not the right article.
If you are curious about computer vision, read on!
Computer Vision, aka CV, is the field of study that pursues the enterprise of teaching computers how to see and extract information from digital visual content (photos, videos, etc.).
A little background
The history of out topic starts with first experiments in the 1950s, when some early models of neural networks were used to detect the edges of simple objects and categorize them into categories such as circles and squares.
We had to wait a few more years (1970s) before meeting the first commercial version of computer vision, with the creation of a software able to interpret typed or handwritten text using optical character recognition.
Trending AR VR Articles:
1. Oculus Go, the Nintendo Switch of VR
2. Expert View: 3 ways VR is transforming Learning & Development
3. Five Augmented Reality Uses That Solve Real-Life Problems
4. Virtual Reality Headsets: What are the Options? Which is Right For You?
Nowadays computer vision is widely used for face detection and profile matching (Facebook), for content control (Instagram) and more. The efficiency and capacity of the technology are always improved, even thanks to the high amount of data that users upload every day in their social media platforms, which are used to train the CV systems.
More has yet to come: by 2022, the CV hardware and software market is expected to reach $48.6 billion volume, with an increased impact in people’s everyday life.
How does CV work?
The tremendous growth in achieving computer vision results was made thanks to the iterative learning process made possible with neural networks. It starts with a curated dataset with information that helps the machine learn a specific topic.
Each image needs to be tagged with metadata that indicates the correct answer. When a neural network runs through data and signals it’s found an image with an object; it’s the feedback that is received regarding if it was correct or not that helps it improve. Neural networks are using pattern recognition to distinguish many different pieces of an image. Instead of a programmer defining the attributes that make that specific object, the machines learn from the millions of images uploaded.
Some pretty cool examples of Computer Vision in practice
- **Google Translate App** can now automatically detect languages so you can point your camera at a flyer or sign and get results in your native tongue even if you don’t know what language you’re reading. Check this cool music video that shows the capabilities.
- **Facial recognition systems** are technologies capable of identifying or verifying a person from a digital image or a video frame from a video source. Check how Sighthound used this technology to create a simple and effective concierge system here.
3. **Autonomous vehicles** In the world of autonomous vehicles, computer vision is often referred to as “perception”, because cameras are the primary (but not the only) tool that vehicle uses to perceive its environment.
Cameras are key to a variety of essential tasks for autonomous vehicles: lane finding, road curvature estimation, obstacle detection and classification, traffic sign detection and classification, traffic light detection and classification, and more. Tesla, for example, equips its cars with “eight surround cameras that provide 360 degrees of visibility around the car at up to 250 meters of range.”
See the streets in the eyes of Tesla Autopilot here!
You want to learn more about Computer Vision? Keep following us and in these series of Computer Vision articles you’ll discover step by step how to build your own computer vision system.
Reach to us for any questions or suggestions!
bucharest@rinftech.com
Don’t forget to give us your 👏 !
