SCViT: A Spatial-Channel Feature Preserving Vision Transformer for Remote Sensing Image Scene Classification.
Pengyuan LvWenjun WuYanfei ZhongFang DuLiangpei ZhangPublished in: IEEE Trans. Geosci. Remote. Sens. (2022)
Keyphrases
- scene classification
- remote sensing images
- spatial pyramid matching
- spatial layout
- object recognition
- image classification
- remote sensing
- natural scenes
- change detection
- visual words
- multispectral
- biologically inspired
- indoor outdoor
- image representation
- scene recognition
- image features
- computer vision
- bag of words
- spatial information
- feature vectors
- bag of features
- multi instance multi label learning
- natural images
- vision system
- face recognition
- image processing
- feature descriptors
- satellite images
- visual scene