Watch a 33-minute conference talk from Data Science Conference Europe 2023 exploring how transformer architectures have revolutionized computer vision applications. Delve into the fascinating transition of transformer models from their original language processing roots to becoming powerful tools in visual tasks. Learn about the fundamental architecture of vision transformers, understand the mechanics of self-attention mechanisms, and discover why these models achieve remarkable performance in visual applications. Presented by Ivan Biliskov in Belgrade, this talk provides a comprehensive overview of how transformer technology is reshaping the landscape of computer vision and setting new benchmarks in the field.