This project has been created for my submission to the 2018 ChaLearn LAP Track 3.
The objective of this project is to train an algorithm that can inpaint and denoise fingerprint images that contain artifacts like noise, scratches, etc. to improve the performance of subsequent operations like fingerprint verification that are typically applied on such images.
The submission finished at the 10th position in the Feedback phase of the competition. It is enough to conclude that the GAN approach indeed does have potential to be used for fingerprint inpainting and denoising.
