What we see in a photograph: content selection for image captioning.
Georgios BarlasChristos VeinidisAvi ArampatzisPublished in: Vis. Comput. (2021)
Keyphrases
- single image
- image data
- input image
- image classification
- image content
- image features
- image analysis
- region of interest
- image retrieval
- multiscale
- image collections
- image structure
- edge detection
- image representation
- image noise
- template matching
- vector field
- image pixels
- segmentation method
- image matching
- lighting conditions
- similarity measure
- image regions
- segmentation algorithm
- spatial information
- hough transform
- keypoints
- low level
- object recognition
- web images
- visual concepts
- metadata