Deblurring Face Images using Uncertainty Guided Multi-Stream Semantic Networks [article]

Rajeev Yasarla
2020 arXiv   pre-print
We propose a novel multi-stream architecture and training methodology that exploits semantic labels for facial image deblurring. The proposed Uncertainty Guided Multi- Stream Semantic Network (UMSN) processes regions belonging to each semantic class independently and learns to combine their outputs into the final deblurred result. Pixel-wise semantic labels are obtained using a segmentation network. A predicted confidence measure is used during training to guide the network towards the
more » ... ng regions of the human face such as the eyes and nose. The entire network is trained in an end- to-end fashion. Comprehensive experiments on three different face datasets demonstrate that the proposed method achieves significant improvements over the recent state-of-the-art face deblurring methods. Code is available at: rajeevyasarla/UMSN-Face-Deblurring
arXiv:1907.13106v2 fatcat:2f6cp2tfdbfnnnxrpg7zj5gy4m