 I'm Luca and I'm the founder of Beautify. Beautify is an AI company and we're specialized in computer vision and machine learning. We work with three types of clients, typically SMEs, startups, but also global companies that really want to put AI and machine learning at the core of their services and products. Today we're going to talk about video recognition, which is a very exciting field of research that has been recently revamped thanks to the deep learning revolution. Today in particular we're going to look at how to create a video recognition pipeline that ingests videos and outputs insights. This has been particularly difficult in the past because there were technical limitations, but now thanks to democratization of AI, extracting insights from videos is actually pretty easy and can create an immense amount of value for your clients. Video recognition is now so advanced for three main reasons or factors if you want. The first one is really about the democratization of AI and machine learning I was referring to. Now using machine learning is much easier than five to ten years ago. Secondly, we have an abundance of data, so we have very rich databases that are available to everyone, practitioners and practitioners and non-practitioners, and more importantly, third, we have very high quality sensors that really enable us to capture at a very global scale video, audio at very high quality.