ViNTER: Image Narrative Generation with Emotion-Arc-Aware Transformer.

Kohei Uehara Yusuke Mori Yusuke Mukuta Tatsuya Harada

Published in: CoRR (2022)

Keyphrases

image features
image data
single image
image segmentation
input image
image pixels
image regions
image classification
image analysis
image retrieval
image representation
multiscale
image content
image collections
high resolution
facial expressions
template matching
low level
segmentation algorithm
similarity measure
image matching
pixel values
database systems