Using Artificial Tokens to Control Languages for Multilingual Image Caption Generation.
Satoshi TsutsuiDavid J. CrandallPublished in: CoRR (2017)
Keyphrases
- image data
- multiscale
- input image
- image analysis
- template matching
- image features
- image pixels
- image classification
- low level
- single image
- image content
- segmentation method
- image representation
- language independent
- test images
- multi lingual
- image segmentation
- region of interest
- image retrieval
- similarity measure
- multilingual documents
- feature points
- high resolution
- cross lingual
- bounding box
- region based image
- text generation
- language resources
- fusion method
- caption text
- image collections
- image matching
- hough transform
- image regions
- semantic information
- visual features
- control system