Abstract: From linear layers and convolutions to self-attention, deep learning is implicitly Euclidean. But should it be? In this talk, I will dive into hyperbolic geometry for deep learning in computer vision. Hyperbolic learning is the key to unlocking hierarchical learning and making deep learning more robust. In the talk, I will place an emphasis on vision-language models, where hyperbolic embeddings are key for vision-language alignment. Lastly, I will dive into the many opportunities for hyperbolic learning in video understanding and how to get started in this field with our survey, tutorials, and software library.