U-Finger: Multi-Scale Dilated Convolutional Network for Fingerprint Image Denoising and Inpainting
This addresses fingerprint image restoration for biometric applications, but it is incremental as it builds on existing convolutional network approaches.
The paper tackled fingerprint image denoising and inpainting by developing a multi-scale convolutional network called U-Finger, which achieved second place in a competition with metrics like MSE of 0.0231 and PSNR of 16.9688 dB.
This paper studies the challenging problem of fingerprint image denoising and inpainting. To tackle the challenge of suppressing complicated artifacts (blur, brightness, contrast, elastic transformation, occlusion, scratch, resolution, rotation, and so on) while preserving fine textures, we develop a multi-scale convolutional network, termed U- Finger. Based on the domain expertise, we show that the usage of dilated convolutions as well as the removal of padding have important positive impacts on the final restoration performance, in addition to multi-scale cascaded feature modules. Our model achieves the overall ranking of No.2 in the ECCV 2018 Chalearn LAP Inpainting Competition Track 3 (Fingerprint Denoising and Inpainting). Among all participating teams, we obtain the MSE of 0.0231 (rank 2), PSNR 16.9688 dB (rank 2), and SSIM 0.8093 (rank 3) on the hold-out testing set.