Enhancing Multi-modal Classification of Violent Events using Image Captioning.
Daniel Vallejo AldanaAdrián Pastor López-MonroyEsaú Villatoro-TelloPublished in: IberLEF@SEPLN (2023)
Keyphrases
- multi modal
- image classification
- single modality
- uni modal
- image data
- image features
- image analysis
- multiscale
- input image
- image content
- multi modality
- high dimensional
- image representation
- classification accuracy
- auto annotation
- audio visual
- cross modal
- image annotation
- edge detection
- low level
- image segmentation
- machine learning
- web images
- image processing
- image regions
- low level features
- feature vectors
- image retrieval
- feature space
- similarity measure
- multiple modalities
- fusing multiple
- feature selection