Blind Image Inpainting with Sparse Directional Filter Dictionaries for Lightweight CNNs
Blind Image Inpainting with Sparse Directional Filter Dictionaries for Lightweight CNNs
复制标题
DOI：
10.1007/s10851-022-01119-6
复制
发表时间：
2022-05
影响因子：
2
通讯作者：
Jenny Schmalfuss;Erik Scheurer;Hengyuan Zhao;Nikolaos Karantzas;Andrés Bruhn;D. Labate
中科院分区：
文献类型：
--
作者：
Jenny Schmalfuss;Erik Scheurer;Hengyuan Zhao;Nikolaos Karantzas;Andrés Bruhn;D. Labate
近年来，基于深度学习架构的盲修复算法表现出了卓越的性能，在图像质量和运行时间方面通常优于基于模型的方法。然而，神经网络策略通常缺乏理论解释，这与基于模型的方法背后众所周知的理论形成鲜明对比。在这项工作中，我们利用两种方法的优势，将变换域方法和稀疏近似的理论概念整合到基于 CNN 的盲图像修复方法中。为此，我们提出了一种学习卷积核的新颖策略，该策略应用专门设计的滤波器字典，其元素与可训练权重线性组合。数值实验证明了该方法的竞争力。我们的结果表明，与传统 CNN 相比，修复质量不仅得到了提高，而且轻量级网络设计中的网络收敛速度也显着加快。我们的代码可在 https://github.com/cv-stuttgart/SDPF_Blind-Inpainting 获取。
Blind inpainting algorithms based on deep learning architectures have shown a remarkable performance in recent years, typically outperforming model-based methods both in terms of image quality and run time. However, neural network strategies typically lack a theoretical explanation, which contrasts with the well-understood theory underlying model-based methods. In this work, we leverage the advantages of both approaches by integrating theoretically founded concepts from transform domain methods and sparse approximations into a CNN-based approach for blind image inpainting. To this end, we present a novel strategy to learn convolutional kernels that applies a specifically designed filter dictionary whose elements are linearly combined with trainable weights. Numerical experiments demonstrate the competitiveness of this approach. Our results show not only an improved inpainting quality compared to conventional CNNs but also significantly faster network convergence within a lightweight network design. Our code is available at https://github.com/cv-stuttgart/SDPF_Blind-Inpainting.