Computer vision has made great leaps forward in recent years because of deep learning, thus granting computers a higher grade in understanding visual scenes. The potentialities of deep learning in vision tasks are great: allowing a computer to visually perceive and understand its surroundings is a capability that opens the door to new artificial intelligence applications in both mobility (for instance, self-driving cars can detect if an appearing obstacle is a pedestrian, an animal or another vehicle from the camera mounted on the car and decide the correct course of action) and human-machine interaction in everyday-life contexts (for instance, allowing a robot to perceive surrounding objects and successfully interact with them).
After presenting ConvNets and how they operate in the first chapter, we now intend to create a quick, easy...