Pre-training image-language transformers for open-vocabulary tasks

We present a pre-training approach for vision and language transformer models, which is based on a mixture of diverse tasks. We explore both the use of image-text captioning data in pre-training, which does not need additional supervision, as well as object-aware strategies to pre-train the model. W...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2022-09
Hauptverfasser:	Piergiovanni, A J, Kuo, Weicheng, Angelova, Anelia
Format:	Artikel
Sprache:	eng
Schlagworte:	Training Transformers Vision Visual tasks
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!