On the Co-Selection of Vision Transformer Features and Images for Very High-Resolution Image Scene Classification.
Souleyman ChaibDou El Kefel MansouriIbrahim OmaraAhmed HagagSahraoui DhelimDjamel Amar BensaberPublished in: Remote. Sens. (2022)
Keyphrases
- scene classification
- image features
- image classification
- scene categories
- bag of features
- spatial layout
- test images
- object recognition
- image representation
- input image
- image data
- visual words
- image collections
- image regions
- spatial pyramid matching
- sample images
- indoor outdoor
- computer vision
- natural scenes
- image content
- image analysis
- keypoints
- scene recognition
- sift descriptors
- scene categorization
- natural images
- image retrieval
- visual features
- image matching
- bag of words
- spatial information
- scene images
- feature extraction
- low level
- feature vectors
- feature representation
- image database
- salient regions
- object detection
- multiscale
- image patches
- illumination conditions
- single image
- image descriptors
- feature points
- color features
- texture features
- pixel level
- color histogram
- biologically inspired
- feature descriptors
- image structure
- lighting conditions
- vision system
- action recognition
- set of training images
- object categories