Skip to content

This is a modified version of LostGAN-v2 with part-wise segmentation masks

Notifications You must be signed in to change notification settings

rakshita111/Layout-to-Image-Synthesis

Folders and files

NameName
Last commit message
Last commit date

Latest commit

184c642 · Jul 24, 2023

History

6 Commits
Jul 24, 2023
Jul 24, 2023
Jul 24, 2023
Jul 24, 2023
Jul 24, 2023
Jul 24, 2023
Jul 24, 2023
Jul 24, 2023
Jul 24, 2023
Jul 24, 2023
Jul 24, 2023

Repository files navigation

This work is a modification of LostGAN-v2, where partwise segmentation masks are combined with the ISLA-Norm at the ResBlocks in the generator.

part_lostgan

Results

Comparison of the images generated by LostGAN-v2 and the modified LostGAN generated_images


Visualization of the partmasks for generated images

partmasks


Quantitative comparison of different layout-to-image synthesis methods on COCO-dataset (128x128 images) table

Installation

Check INSTALL.md for installation instructions.

1. Train models

python train.py --dataset coco --out_path outputs/

2. Run pretrained model

python test.py --dataset coco --model_path pretrained_model/G_coco.pth --sample_path samples/coco/

Reference

The code is majorly borrowed from LostGAN-v2.

About

This is a modified version of LostGAN-v2 with part-wise segmentation masks

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published