Enhanced Deep Image Prior for Unsupervised Hyperspectral Image Super-Resolution
Chinese Academy of Sciences · Aerospace Information Research Institute · +6 more institutions
Abstract
Depending on a large-scale paired dataset of low-resolution hyperspectral image (LrHSI), high-resolution multispectral image (HrMSI), and corresponding high-resolution hyperspectral image (HrHSI), the supervised paradigm has achieved impressive performance in the hyperspectral image super-resolution (HISR). However, the intrinsic data-intensive manner hinders its further application in real scenarios. Fortunately, deep image prior (DIP) allows us to achieve unsupervised super-resolution (SR) by solely utilizing degraded observations. However, its potential to accurately model complicated hyperspectral priors is still not fully exploited due to the following two factors: 1) existing methods tend to reconstruct…
Citation impact
- FWCI
- 141.33
- Percentile
- 100%
- References
- 114
Authors
6- JLJiaxin LiCorresponding
Chinese Academy of Sciences, Aerospace Information Research Institute
- KZKe Zheng
Liaocheng University
- LGLianru Gao
Chinese Academy of Sciences, Aerospace Information Research Institute
- ZHZhu Han
University of Chinese Academy of Sciences
- ZLZhi Li
Chinese Academy of Sciences, Aerospace Information Research Institute
Topics & keywords
- Hyperspectral imaging
- Artificial intelligence
- Image (mathematics)
- Computer science
- Image resolution
- Computer vision
- Remote sensing
- Pattern recognition (psychology)