In the second installment of our 2-part event on deep computer vision, we have a hands-on lecture on the fundamentals of deep computer vision. (First part: https://www.meetup.com/AI-Lab/events/254880165/)
We start by introducing the convolutional neural network, the building block of most modern day computer vision algorithms. You will learn how CNNs work and you will build your own image classifier in TensorFlow/Keras. (1,5 hours)
After the break, we will dive into YOLO (You Only Look Once) – the state of the art algorithm used to detect and localize objects in real time. You will understand the important concepts used and will see a demo of YOLO in action. (1,5 hours)
Note that while everyone can follow the lecture as it will be presented in relatively simple language, you will get the most out of the lecture if you can code in Python and are willing to follow along with coding during the lecture. Skeleton code in Python Notebooks will be provided and we will fill in the relevant code together.