Language Does More Than Describe: On The Lack Of Figurative Speech in Text-To-Image Models.
Ricardo KleinleinCristina Luna JiménezFernando Fernández MartínezPublished in: CoRR (2022)
Keyphrases
- image content
- english text
- input image
- multiscale
- image features
- text to speech synthesis
- single image
- image segmentation
- text to speech
- random fields
- image data
- image representation
- bayesian framework
- image retrieval
- image classification
- low level
- image collections
- language generation
- web images
- image analysis
- high resolution
- information retrieval
- computational linguistics
- statistical model
- region of interest
- complex background
- hough transform
- markov random field
- text mining
- pixel values
- image regions
- segmentation method
- segmentation algorithm
- language acquisition
- edge detection
- similarity measure
- scanned documents
- text input
- probabilistic model