Try Training with Distance Transformation as Label instead of Binary Label.

ManishSahu53 / Vector-Map-Generation-from-Aerial-Imagery-using-Deep-Learning-GeoSpatial-UNET

We propose a simple yet efficient technique to leverage semantic segmentation model to extract and separate individual buildings in densely compacted areas using medium resolution satellite/UAV orthoimages. We adopted standard UNET architecture, additionally added batch normalization layer after every convolution, to label every pixel in the image.

Apache License 2.0

91 stars 33 forks source link

Try Training with Distance Transformation as Label instead of Binary Label. #15

Open ManishSahu53 opened 4 years ago

ManishSahu53 commented 4 years ago

Use Distance Tranformation Map as label isntead of Binary Label