Seeing Beyond the Patch: Scale-Adaptive Semantic Segmentation of High-resolution Remote Sensing Imagery based on Reinforcement Learning.
Yinhe LiuSunan ShiJunjue WangYanfei ZhongPublished in: ICCV (2023)
Keyphrases
- remote sensing imagery
- remote sensing
- semantic segmentation
- high resolution
- reinforcement learning
- change detection
- superpixels
- multispectral
- remote sensing images
- satellite images
- conditional random fields
- hyperspectral
- low resolution
- image processing
- weakly supervised
- remote sensing data
- super resolution
- scene classification
- image patches
- high frequency
- image analysis
- multiscale
- state space
- wavelet transformation
- object categories
- supervised learning
- computer vision
- data sets
- bit rate
- scale space
- object classes
- feature vectors
- training data