Face Inpainting with Deep Generative Models

Face Inpainting with Deep Generative Models
复制标题

使用深度生成模型进行面部修复

DOI:
10.2991/ijcis.d.191016.003
复制
发表时间:
2019-10
影响因子:
2.9
通讯作者:
Junqiu Li
Junqiu Li
中科院分区:
计算机科学4区
文献类型:
--
作者:
Zhenping Qiang;Libo He;Qinghui Zhang;Junqiu Li

文献摘要

参考文献

相似文献

从受损图像中进行人脸语义修复是计算机视觉中的一个具有挑战性的问题,具有许多实际应用。与已有的自然图像修复不同,人脸修复任务通常需要根据已有的视觉数据对缺失区域进行语义填充。本文提出了一种新的基于深度生成模型的人脸修复算法,在图像生成模型中增加了结构损失约束,以确保生成的图像具有与待修复人脸图像尽可能相似的结构。同时,在受损图像中计算不同的权值,以增强修复边界的边缘一致性。在不同人脸数据集上的实验和定性和定量分析表明,该算法能够生成视觉上令人满意的人脸补齐。
Semantic face inpainting from corrupted images is a challenging problem in computer vision and hasmany practical applications. Different fromwell-studied nature image inpainting, the face inpainting task often needs to fill pixels semantically into a missing region based on the available visual data. In this paper, we propose a new face inpainting algorithm based on deep generative models, which increases the structural loss constraint in the image generation model to ensure that the generated image has a structure as similar as possible to the face image to be repaired. At the same time, different weights are calculated in the corrupted image to enforce edge consistency at the repair boundary. Experiments on different face data sets and qualitative and quantitative analyses demonstrate that our algorithm is capable of generating visually pleasing face completions.
DOI: 10.1145/1201775.882269
发表时间: 2003-07
期刊: ACM SIGGRAPH 2003 Papers
影响因子: --
作者:
P. Pérez;Michel Gangnet;A. Blake
通讯作者: P. Pérez;Michel Gangnet;A. Blake
DOI: 10.1007/s00371-007-0175-y
发表时间: 2008-01-01
期刊: VISUAL COMPUTER
影响因子: 3.5
作者:
Ersotelos, Nikolaos;Dong, Feng
通讯作者: Dong, Feng
DOI: 10.1561/2200000006
发表时间: 2009-01-01
影响因子: 32.8
作者:
Bengio, Yoshua
通讯作者: Bengio, Yoshua
DOI: 10.1145/882262.882269
发表时间: 2003-07-01
影响因子: 6.2
作者:
Pérez, P;Gangnet, M;Blake, A
通讯作者: Blake, A
DOI: 10.5201/ipol.2016.163
发表时间: 2016-01-01
影响因子: 1
作者:
Di Martino, J. Matias;Facciolo, Gabriele;Meinhardt-Llopis, Enric
通讯作者: Meinhardt-Llopis, Enric