# Blind Image Inpainting with Sparse Directional Filter Dictionaries for Lightweight CNNs

arXiv: https://arxiv.org/abs/2205.06597 · HTML: https://mlsanity.com/paper/2205.06597/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 49 references that are in the map.
- First posted on arXiv 13 May 2022.

## Builds on (most cited first, 13 of 13)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [Going Deeper with Convolutions](https://mlsanity.com/paper/1409.4842/) (2014)
- [Context Encoders: Feature Learning by Inpainting](https://mlsanity.com/paper/1604.07379/) (2016)
- [Free-Form Image Inpainting with Gated Convolution](https://mlsanity.com/paper/1806.03589/) (2018)
- [Generative Image Inpainting with Contextual Attention](https://mlsanity.com/paper/1801.07892/) (2018)
- [Image Inpainting for Irregular Holes Using Partial Convolutions](https://mlsanity.com/paper/1804.07723/) (2018)
- [A Survey on Neural Network Interpretability](https://mlsanity.com/paper/2012.14261/) (2020)
- [Contextual Residual Aggregation for Ultra High-Resolution Image Inpainting](https://mlsanity.com/paper/2005.09704/) (2020)
- [Structured Receptive Fields in CNNs](https://mlsanity.com/paper/1605.02971/) (2016)
- [VCNet: A Robust Approach to Blind Image Inpainting](https://mlsanity.com/paper/2003.06816/) (2020)
- [Deep Blind Image Inpainting](https://mlsanity.com/paper/1712.09078/) (2017)
- [Can fully convolutional networks perform well for general image restoration problems?](https://mlsanity.com/paper/1611.04481/) (2016)

## Built on by (most cited first, 1 of 1)

- [BVINet: Unlocking Blind Video Inpainting with Zero Annotations](https://mlsanity.com/paper/2502.01181/) (2025)

## Abstract (arXiv)

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.