Learning to Deblur Face Images via Sketch Synthesis

• DOI: 10.1609/aaai.v34i07.6818
• Publication Date: 2020-06-29T18:36:17Z
• Authors: Songnan Lin; Jiawei Zhang; Jinshan Pan; Yicun Liu; Yongtian Wang; Jing Chen; Jimmy Ren
• Keywords: Deblurring; Motion blur; Sketch

The success of existing face deblurring methods based on deep neural networks is mainly due to the large model capacity. Few algorithms have been specially designed according domain knowledge images and physical properties process. In this paper, we propose an effective algorithm convolutional (CNNs). Motivated by conventional process which usually involves motion blur estimation latent clear image restoration, proposed first estimates a CNN then restores with estimated blur. However, estimating from blurry difficult as textures are scarce. As most share some common global structures can be modeled well sketch information, learn sketches so that help estimation. With blur, develop restoration CNN. Although involving several components, trained in end-to-end fashion. We analyze effectiveness each component show able deblur favorable performance against state-of-the-art methods.