Using the Image-Text Relationship to Improve Multimodal Disaster Tweet Classification.
Tiberiu SoseaIustin SirbuCornelia CarageaDoina CarageaTraian RebedeaPublished in: ISCRAM (2021)
Keyphrases
- image classification
- image data
- image features
- image content
- single image
- image feature extraction
- high resolution
- image analysis
- input image
- image regions
- image retrieval
- image segmentation
- supervised machine learning
- pattern recognition
- support vector
- multiscale
- feature space
- feature extraction
- segmentation method
- machine learning
- low level
- preprocessing stage
- confidence scores
- information retrieval
- image representation
- visual features
- feature points
- training set
- web documents
- edge detection
- text mining
- classification accuracy
- image set
- image collections
- web images
- hyperspectral images
- feature vectors
- textural features
- scanned documents
- similarity measure
- decision trees