Awesome

Scale-aware Detailed Matching for Few-Shot Aerial Image Semantic Segmentation

Codes and dataset (iSAID-5i) for Scale-aware Detailed Matching for Few-Shot Aerial Image Semantic Segmentation, and the work has been accepted by TGRS

the overall network:

<p align="left"> <img src="img/remote_sensing1.png" alt="the overall network" width="700px"> </p> some visualization results: the overall network: <p align="left"> <img src="img/remote_sensing_result.png" alt="the results" width="800px"> </p>

Training

cd scripts
sh train_group0.sh

Inference

If you want to test all of the saved models, you can use:

python test_all_frame.py

Environment

python == 3.7
pytorch1.0
torchvision,
pillow,
opencv-python,
pandas,
matplotlib,
scikit-image

Datasets and Data Preparation

The newly provied dataset iSAID-5i
(Password:nwpu) or iSAID-5i