CIC-BART-SSA: Controllable Image Captioning with Structured Semantic Augmentation.
Kalliopi BasiotiMohamed A. AbdelsalamFederico FancelluVladimir PavlovicAfsaneh FazlyPublished in: CoRR (2024)
Keyphrases
- input image
- image analysis
- multiscale
- image data
- image classification
- image features
- image pixels
- single image
- template matching
- low level
- edge detection
- segmentation method
- image retrieval
- feature points
- image content
- hough transform
- image collections
- image structure
- semantically meaningful
- multiresolution
- natural language
- high level
- keypoints
- image representation
- high resolution
- key frames
- visual concepts
- image processing