An Eye for Self-Driving Cars
Computer Vision: field of Computer Science to enable computers to see, has numerous applications. One of the greatest applications is making Self-Driving Cars a reality. I consider it as one of the great applications of Computer Vision because as per research driverless cars are poised to save 10 million lives per decade – and 50 million lives around the world in half a century. This is a great contribution to mankind.
The idea of self-driving cars on our streets seemed like a crazy sci-fi fantasy just a few short years ago. A self-driving car is capable of sensing the environment and navigating with little or no human input.
The two main approaches used by Self-driving Car Engineers are –
- Deep learning using computer vision
Engineers have been working on perfecting robotics approach for years, but the outcome was not fruitful. Advancement in infrastructure enabled them to use deep learning for computer vision. The rapid progress made in recent years in artificial intelligence (AI) and autonomous technology proves that the self-driving car is becoming a reality.
Computer Vision provides eyes to self-driving cars, these are great for spotting things like lane lines on the highway, pedestrian, speed signs, and traffic lights. Some researchers think that, with better machine vision, they can use cameras to identify everything they see and navigate accordingly.
Manufacturers use both computer vision and sensor fusion for developing the solution. In the below table, you can see the comparison of preferred approaches (deep learning and robotics), from different car makers standpoints in the development of autonomous technology:
Computer vision is going to contribute in many fields, not like humans, who can get overwhelmed. A computer can see many things at once, in high detail and analyse without getting “tired”. The accuracy of computer analysis can bring tremendous time savings and quality improvements.
By: Sumit Ranjan