Gated PE-NL-MA: A multi-modal attention based network for video understanding.

Chengyang Xie Xiaoping Wang

Published in: Neurocomputing (2021)

Keyphrases

multi modal
semantic concepts
video search
audio visual
cross modal
multiple modalities
video data
high dimensional
image annotation
multi modality
key frames
single modality
video sequences
information theoretic
video clips
uni modal