去模糊
计算机科学
人工智能
面子(社会学概念)
计算机视觉
图像处理
模式识别(心理学)
图像(数学)
图像复原
社会科学
社会学
作者
Rajeev Yasarla,Federico Perazzi,Vishal M. Patel
标识
DOI:10.1109/tip.2020.2990354
摘要
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 challenging 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: https://github.com/ rajeevyasarla/UMSN-Face-Deblurring
科研通智能强力驱动
Strongly Powered by AbleSci AI