README.md

CSCI-594-FALL23

There are two models: the baseline model and the model we are proposing. Additionally, experiments with the models were done to generate Monet-like images and Bob Ross-like images.

Links to the dataset:

1. Link to the Kaggle Competition with the Monet Dataset
2. Link to the Bob Ross dataset

Links to annotated custom datasets:

1. Link to the Segmented Monet Dataset
2. Link to the Segmented Photos (using Segformer, ADE20K)

Files are:

• Baselines:
  1. Baseline_monet_style => Implementation of the Baseline model to generate Monet-like images
  2. Baseline-bob-ross_style => Implementation of the Baseline model to generate Bob Ross-like images
• Our implementation:
  1. SegCycleGAN_monet_style => Implementation of the CycleGAN with the segmented features to generate Monet-like images
  2. SegCycleGAN-bob-ross_style => Implementation of the CycleGAN with the segmented features to generate Bob Ross-like images