Building Footprint Segmentation

Library to train building footprint on satellite and aerial imagery.

Installation

pip install building-footprint-segmentation

Dataset

• Massachusetts Buildings Dataset
• Inria Aerial Image Labeling Dataset

Training

• Train With Config , Use config template for generating training config

• Train With Arguments

Visualize Training

Test images at end of every epoch. Follow Example

from building_footprint_segmentation.helpers.callbacks import CallbackList, TestDuringTrainingCallback

class TestCallback(TestDuringTrainingCallback):
    def inference(self, model, image, file_name, save_path, index):
        """
        
        :param model: the model used for training
        :param image: the images loaded by the test loader
        :param file_name: the file name of the test image
        :param save_path: path where to save the image
        :param index: 
        :return: 
        """
        # Define this method on how to handle the prediction at the end of every epoch

where_to_log_the_callback = r"path_to_log_callback"   
callbacks = CallbackList()

# Ouptut from all the callbacks caller will be stored at the path specified in log_dir
callbacks.append(TestCallback(where_to_log_the_callback))

Visualizing on Tensorboard

from building_footprint_segmentation.helpers.callbacks import CallbackList, TensorBoardCallback
where_to_log_the_callback = r"path_to_log_callback"   
callbacks = CallbackList()

# Ouptut from all the callbacks caller will be stored at the path specified in log_dir
callbacks.append(TensorBoardCallback(where_to_log_the_callback))

To view Tensorboard dash board

tensorboard --logdir="path_to_log_callback"

Segmentation for building footprint

• binary
• building with boundary (multi class segmentation)

Weight File

• RefineNet