DeepTriNet: A Tri-Level Attention Based DeepLabv3+ Architecture for Semantic Segmentation of Satellite Images.
Tareque Bashar OviShakil MosharrofNomaiya BashreeMd Shofiqul IslamMuhammad Nazrul IslamPublished in: CoRR (2023)
Keyphrases
- satellite images
- semantic segmentation
- multispectral
- remote sensing
- street scenes
- change detection
- superpixels
- label transfer
- pixel level
- object categories
- image understanding
- conditional random fields
- high resolution
- image data
- computer vision
- higher order
- object detection
- object classes
- multi class
- weakly supervised
- image analysis