Associating multiple vision transformer layers for fine-grained image representation.
Fayou SunNgo Hea ChoonYong Wee SekZuqiang MengPublished in: AI Open (2023)
Keyphrases
- fine grained
- image representation
- coarse grained
- multiscale
- image classification
- image content
- object recognition
- image features
- computer vision
- bag of words
- access control
- tightly coupled
- quadtree
- representation scheme
- image retrieval
- data lineage
- scene recognition
- visual words
- image processing
- scene classification
- receptive fields
- distributed systems
- markov random field
- bag of visual words
- pattern representation
- feature space
- linear quadtree