BES-Net: Boundary Enhancing Semantic Context Network for High-Resolution Image Semantic Segmentation.
Fenglei ChenHaijun LiuZhihong ZengXichuan ZhouXiaoheng TanPublished in: Remote. Sens. (2022)
Keyphrases
- semantic segmentation
- high resolution
- image representation
- image data
- image content
- input image
- conditional random fields
- image classification
- image collections
- image set
- super resolution
- image features
- image retrieval
- image segmentation
- feature extraction
- object categories
- superpixels
- multiscale
- segmentation method
- low resolution
- test images
- low level
- scene classification
- image regions
- visual features
- region of interest
- visual words
- feature vectors
- object recognition
- bounding box
- similarity measure