Global and Local Attention-Based Free-Form Image Inpainting
- Authors
- Uddin, S.M.N.; Jung, Yong Ju
- Issue Date
- Jun-2020
- Publisher
- MDPI
- Keywords
- Attention module; Convolutional neural networks (CNN); Free-form mask; Image inpainting; Mask update
- Citation
- Sensors, v.20, no.11, pp.1 - 27
- Journal Title
- Sensors
- Volume
- 20
- Number
- 11
- Start Page
- 1
- End Page
- 27
- URI
- https://scholarworks.bwise.kr/gachon/handle/2020.sw.gachon/77458
- DOI
- 10.3390/s20113204
- ISSN
- 1424-8220
- Abstract
- Deep-learning-based image inpainting methods have shown significant promise in both rectangular and irregular holes. However, the inpainting of irregular holes presents numerous challenges owing to uncertainties in their shapes and locations. When depending solely on convolutional neural network (CNN) or adversarial supervision, plausible inpainting results cannot be guaranteed because irregular holes need attention-based guidance for retrieving information for content generation. In this paper, we propose two new attention mechanisms, namely a mask pruning-based global attention module and a global and local attention module to obtain global dependency information and the local similarity information among the features for refined results. The proposed method is evaluated using state-of-the-art methods, and the experimental results show that our method outperforms the existing methods in both quantitative and qualitative measures. © 2020 by the authors. Licensee MDPI, Basel, Switzerland.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - IT융합대학 > 소프트웨어학과 > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.