Learning Image Embeddings using Convolutional Neural Networks for Improved Multi-Modal Semantics.

Douwe Kiela Léon Bottou

Published in: EMNLP (2014)

Keyphrases

multi modal
auto annotation
convolutional neural networks
image regions
multi modality
uni modal
low level
image data
input image
image classification
image features
image analysis
high dimensional
image segmentation
image processing
cross modal
visual recognition
high resolution
contrast enhancement
image collections
segmentation method
multiple modalities
image representation
convolutional network
feature selection