Introduction to Computer Vision with CNNs

Explore an introduction to computer vision with convolutional neural networks (CNNs) and transfer learning in this 36-minute talk from the Data Science Festival Summer School. Discover how CNNs have revolutionized computer vision tasks and learn practical techniques for implementing them in data science projects, even with limited training data. Gain insights into transfer learning, a powerful method for fine-tuning pre-trained CNNs on smaller datasets. Follow along as Chanuki Seresinhe, Data Scientist at Culture Trip, demonstrates real-world applications of CNNs in assessing image quality and extracting valuable insights from visual data. Delve into the history of CNNs, understand key concepts like activation functions, and learn step-by-step how to train your own CNN models. Uncover innovative uses of CNNs in research, including predicting image scenicness and classifying urban environment features. Whether you're new to computer vision or looking to expand your skills, this talk provides a comprehensive overview of CNNs and their practical applications in data science.