图像补全_图像修复：EdgeConnect

用看智障的眼神看

于 2021-01-11 22:54:25 发布

阅读量2.8k

点赞数 1

CC 4.0 BY-SA版权

文章标签：图像补全

本文链接：https://blog.csdn.net/weixin_36074841/article/details/112576447

EdgeConnect是一种新型的图像修复方法，利用深度学习生成边缘以改善图像补全的质量。它由边缘生成器和图像完成网络组成，能更好地恢复图像的精细细节，避免过度平滑和模糊。在CelebA、Places2和Paris StreetView等数据集上，该模型表现出优于当前最佳技术的定量和定性结果。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

应该有一些人或专栏写过这篇论文了。

链接：EdgeConnect: Generative Image Inpainting with Adversarial Edge Learning

代码链接：knazeri/edge-connect

摘要：

Over the last few years, deep learning techniques have yielded significant improvements in image inpainting. However, many of these techniques fail to reconstruct reasonable structures as they are commonly over-smoothed and/or blurry. This paper develops a new approach for image inpainting that does a better job of reproducing filled regions exhibiting fine details. We propose a two-stage adversarial model EdgeConnect that comprises of an edge generator followe