Kok Wei

I tried with this approach, the output looks jagged, it wouldn't look natural in way: The code to reproduce it: img = np.zeros((800,800),dtype='uint8') point_x = [0, 100, 200, 250, 450,...

> > > Ok, so there is a logical error here while applying the chaikins algorithm. By writing the smooth function inside the `for i in (range(len(point_x)-3)):` you are basically...

> These can now be reproduced with BadPhotoCopy, right? Not exactly, we can reproduce a similar effect, but those noise look sparser (lighter). I will include another update to adjust...

I was testing on the effect earlier and was able to generate something like this:  Will something like this useful? Actually I couldn't access the full paper above, so...

Creating the style should be feasible, take this example: https://www.kaggle.com/patrickaudriaz/tobacco3482jpg  From here we can see several matching augmentations, such as page borders, ink bleed, dusty ink and etc. But...

Okay, so i think we can do it either way. If the source document is having certain format, such as formal email, quotation, newspaper and etc, then we need to...

So right now we are only looking at RVL-CDIP dataset right? I need some time to download it since the dataset size is large. As a first step, I will...

Here's the compiled list of images by looking only at rvl-cdip dataset `imagesA` folder : https://drive.google.com/drive/folders/17Z4VRqc1w9ZKpjoO3dulSqXTpxPVutMN?usp=sharing Some are not fully reproducible, i included them here since they look interesting enough....

> **Problem Faced** : While using the repo, it took some time to go through all the available augmentations and there examples as many markdowns don't have sample results, and...

> @kwcckw cool, will do that. though we can also push images to LFS if space is a concern, as the images are not more than 50-100Kb > > will...