Self-Guiding Multimodal LSTM - when we do not have a perfect training dataset for image captioning.

Yang Xian Yingli Tian

Published in: CoRR (2017)

Keyphrases

training dataset
image data
single image
image features
segmentation method
image analysis
image content
input image
image representation
image retrieval
image pixels
multiscale
region of interest
image classification
image regions
low level
training data
image segmentation
edge detection
pixel values
high resolution
similarity measure
feature points
image quality
training samples
semi supervised
image collections