MSG-SR-Net: A Weakly Supervised Network Integrating Multiscale Generation and Superpixel Refinement for Building Extraction From High-Resolution Remotely Sensed Imageries.
Xin YanLi ShenJicheng WangXu DengZhilin LiPublished in: IEEE J. Sel. Top. Appl. Earth Obs. Remote. Sens. (2022)
Keyphrases
- weakly supervised
- superpixels
- multiscale
- remotely sensed
- change detection
- natural images
- object class
- object segmentation
- remotely sensed images
- infrared
- image processing
- conditional random fields
- image representation
- topic models
- segmentation algorithm
- image patches
- super resolution
- scale space
- remote sensing
- image segmentation
- segmentation method
- graph structure
- semi supervised
- learning algorithm