Journal
JOURNAL OF MATHEMATICAL IMAGING AND VISION
Volume 65, Issue 2, Pages 323-339Publisher
SPRINGER
DOI: 10.1007/s10851-022-01119-6
Keywords
Deep learning; Image restoration; Inpainting; Neural networks; Sparse representations
Ask authors/readers for more resources
Blind inpainting algorithms based on deep learning architectures have shown remarkable performance, outperforming model-based methods in terms of image quality and run time. However, these neural network strategies often lack theoretical explanations, unlike model-based methods. This work combines the advantages of transform domain methods and sparse approximations with a CNN-based approach for blind image inpainting.
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.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available