Coarse-to-fine Task-driven Inpainting for Geoscience Images
Journal Article
·
· IEEE Transactions on Circuits and Systems for Video Technology
- Cleveland State Univ., Cleveland, OH (United States)
- Agency for Science, Technology and Research (A*STAR), Singapore. Centre for Frontier AI Research (CFAR)
- Wuhan Univ. (China)
- Beijing University of Technology (China)
- Brookhaven National Laboratory (BNL), Upton, NY (United States)
The processing and recognition of geoscience images have wide applications. Most of existing researches focus on understanding the high-quality geoscience images by assuming that all the images are clear. However, in many real-world cases, the geoscience images might contain occlusions during the image acquisition. This problem actually implies the image inpainting problem in computer vision and multimedia. As far as we know, all the existing image inpainting algorithms learn to repair the occluded regions for a better visualization quality, they are excellent for natural images but not good enough for geoscience images, and they never consider the following geoscience task when developing inpainting methods. Here, this paper aims to repair the occluded regions for a better geoscience task performance and advanced visualization quality simultaneously, without changing the current deployed deep learning based geoscience models. Because of the complex context of geoscience images, we propose a coarse-to-fine encoder-decoder network with the help of designed coarse-to-fine adversarial context discriminators to reconstruct the occluded image regions. Due to the limited data of geoscience images, we propose a MaskMix based data augmentation method, which augments inpainting masks instead of augmenting original images, to exploit the limited geoscience image data. The experimental results on three public geoscience datasets for remote sensing scene recognition, cross-view geolocation and semantic segmentation tasks respectively show the effectiveness and accuracy of the proposed method. The code is available at: https://github.com/HMS97/Task-driven-Inpainting.
- Research Organization:
- Brookhaven National Laboratory (BNL), Upton, NY (United States)
- Sponsoring Organization:
- National Science Foundation (NSF); USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
- Grant/Contract Number:
- SC0012704
- OSTI ID:
- 1993157
- Report Number(s):
- BNL-224638-2023-JAAM
- Journal Information:
- IEEE Transactions on Circuits and Systems for Video Technology, Journal Name: IEEE Transactions on Circuits and Systems for Video Technology Journal Issue: 12 Vol. 33; ISSN 1051-8215
- Publisher:
- IEEECopyright Statement
- Country of Publication:
- United States
- Language:
- English
Similar Records
Inpainting with sparse linear combinations of exemplars
Load Profile Inpainting for Missing Load Data Restoration and Baseline Estimation
Journal Article
·
Mon Dec 31 23:00:00 EST 2007
·
OSTI ID:960615
Load Profile Inpainting for Missing Load Data Restoration and Baseline Estimation
Journal Article
·
Thu Jul 06 20:00:00 EDT 2023
· IEEE Transactions on Smart Grid
·
OSTI ID:2329385