Implement image pyramid (Gaussian and Laplacian) as discussed in the class - may refer to paper. You are required to apply this for image compression and mosacing. Implement image inpainting. Implement image analogies.

Note: Arguments in <> brackets are mandatory whereas arguments asked in [] have default values hard-coded

To run image mosaicing algorithm:
python3 mosaicing.py --source <SOURCE_IMAGE_PATH> --dest <SOURCE_IMAGE2_PATH> [--levels no_of_levels_of_pyramid] --output OUTPUT_IMAGE_FOLDER

To run image compression algorithm:
python3 compression_median_cut_binning.py --source <SOURCE_IMAGE_PATH> [--levels no_of_levels_of_pyramid] [--to_keep no._of_bins/colors retained] [--mode median_cut/binning] --output OUTPUT_IMAGE_FOLDER

To run image analogy algorithm:
python3 image_analogies.py --source <SOURCE_IMAGE_PATH> --source_filtered <FILTERED_SOURCE_IMAGE_PATH> --dest <DEST_UNFILTERED_PATH> [--levels no_of_levels_of_pyramid] [--mode RGB/YIQ] --output OUTPUT_IMAGE_FOLDER [--resize_height resize_height_to_this] [--kappa hyperparameter]

Notes:

1. YIQ/RGB - Use RGB mode only if task is of transferring color details.
2. resize_height - Kept low as task is computationally very expensive
3. kappa - hyperparameter -> not to be changed if beginner

To run image inpainting algorithm:

python3 inpaint.py

Note: The file works with hard_coded image path. Please make necessary changes to make it an argument. I am too lazy to make this change :p