End-to-end learning of visual representations from uncurated instructional videos

Annotating videos is cumbersome, expensive and not scalable. Yet, many strong video models still rely on manually annotated data. With the recent introduction of the HowTo100M dataset, narrated videos now offer the possibility of learning video representations without manual supervision. In this wor...

Descripció completa

Dades bibliogràfiques
Autors principals: Miech, A, Alayrac, J-B, Smaira, L, Laptev, I, Sivic, J, Zisserman, A
Format: Conference item
Idioma:English
Publicat: IEEE 2020