MTLE: A Multitask Learning Encoder of Visual Feature Representations for Video and Movie Description

Open in new window