Tri-RAT: optimizing the attention scores for image captioning.
You YangYongzhi AnJuntao HuLongyue PanPublished in: Int. J. Multim. Inf. Retr. (2022)
Keyphrases
- image data
- image features
- single image
- input image
- image analysis
- image content
- image classification
- image pixels
- multiscale
- image regions
- region of interest
- test images
- edge detection
- image collections
- template matching
- image representation
- spatial information
- keypoints
- segmentation method
- post processing
- high resolution
- grey level
- image segmentation
- feature points
- low level
- image retrieval
- confidence scores
- segmentation algorithm
- ranked list
- visual attention
- object recognition
- similarity measure