Skip to content

Generating Maps from Satellite images using the Pix2Pix GAN.

Notifications You must be signed in to change notification settings

jatin67singh/GANS-Satellite_to_Maps

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

GANS-Satellite_to_Maps

--> Generating Maps from Satellite images using the Pix2Pix GAN.
Developed and Trained a model that can convert aerial satellite imagery ("input") into map routes ("output"). The architecture for the generator is a U-Net and the Discriminator is a PatchGAN. I have saved the model checkpoint while training(pix2pix_15000.pth), that can be used as a pre-trained model. However, it can be trained from scratch.

Training parameters

adv_criterion = nn.BCEWithLogitsLoss()
recon_criterion = nn.L1Loss()
lambda_recon = 200

n_epochs = 20
input_dim = 3
real_dim = 3
display_step = 200
batch_size = 4
lr = 0.0002
target_shape = 256
device = 'cpu' (could be cuda if your device supports!)

Dimensions

Shape of the input tensor: torch.Size([1, 3, 256, 256])
Shape of the Output tensor: torch.Size([1, 3, 256, 256])

Dataset

--> Maps.zip
It contains two folders, one for each training and validation data.
The image tensor is horizontally concatenated. The condition(satellite image) is left half of the Image and the label(map) is the right half of each image in the dataset.

Training

Screenshot 2024-05-22 at 4 59 43 PM

About

Generating Maps from Satellite images using the Pix2Pix GAN.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published