TRCaptionNet: A novel and accurate deep Turkish image captioning model with vision transformer based image encoders and deep linguistic text decoders.
Serdar YildizAbbas MemisSongül VarliPublished in: Turkish J. Electr. Eng. Comput. Sci. (2023)
Keyphrases
- image data
- image analysis
- single image
- input image
- image segmentation
- image content
- template matching
- statistical model
- multiscale
- image classification
- test images
- color vision
- image collections
- reconstruction method
- region of interest
- bayesian framework
- image pixels
- energy functional
- segmentation method
- image representation
- segmentation algorithm
- feature points
- probabilistic model
- similarity measure
- image retrieval
- energy function
- prior model
- image regions
- image processing
- image matching
- visual perception
- image features
- high resolution
- information retrieval
- neural network
- natural language
- probability density function
- multiresolution
- natural language processing
- edge detection
- scale space
- graph cuts