Transformer-based local-global guidance for image captioning.
Hashem ParvinAhmad Reza Naghsh-NilchiHossein Mahvash MohammadiPublished in: Expert Syst. Appl. (2023)
Keyphrases
- input image
- image data
- image classification
- multiscale
- image segmentation
- image features
- image noise
- image content
- single image
- image analysis
- low level
- template matching
- feature points
- image representation
- global information
- pixel values
- test images
- multiresolution
- image collections
- image regions
- hough transform
- spatial information
- image retrieval
- object detection
- image pixels
- region of interest
- keypoints
- segmentation method
- binary images
- segmentation algorithm
- edge detection