Topic:

An explicit and scene-adapted definition of convex self-similarity prior with application to unsupervised Sentinel-2 super-resolution  

 

Speaker:

林家祥 教授 Prof. Chia-Hsiang Lin

 

成功大學電機工程學系

 Dept. of Electrical Engineering,

National Cheng Kung University

 

Language :

Chinese 

 

Date:

2020/06/05

 

Time:

14:20~

 

Location: 

Delta building R216, NTHU

 

Abstract:

Sentinel-2 satellite, launched by the European Space Agency, plays a critical role in various Earth observation missions. However, the spatial resolutions of Sentinel-2 images are different across its 12 spectral bands, meaning that there is no pixel in such imagery. To facilitate the analysis of such multi-resolution images, super-resolving (SR) of the low-resolution bands to a higher resolution is desired. Without relying on big data, we computationally achieve this SR task from a single dataset. As in many image restoration inverse problems, we exploit image self-similarity, a commonly observed property in natural images. However, the design of self-similarity regularization in non-diagonal inverse problems is challenging; often, a self-similarity based denoiser is plugged into the algorithmic iterations, without a guarantee of convergence in general. For the first time, we explicitly define the concept of self-similarity as a convex function, built explicitly on a self-similarity graph that can be directly learned from the Sentinel-2 images. Remarkably, the function is scene-adapted, unlike widely used sparsity or total-variation regularization schemes. We then develop a fast algorithm, termed Sentinel-2 super-resolution via scene-adapted self-similarity (SSSS), which efficiently and exactly solves three involved different types of very large-scale matrix inversions. We experimentally show the superiority of SSSS over four commonly observed scenes, indicating the potential usage of our newly introduced convex self-similarity regularization in other ill-posed imaging inverse problems.

 

Short bio: 

Chia-Hsiang Lin (S’10–M’18) received the B.S. degree in electrical engineering and the Ph.D. degree in communications engineering from National Tsing Hua University (NTHU), Taiwan, in 2010 and 2016, respectively. 

 

He is currently an Assistant Professor with the Department of Electrical Engineering and the Miin Wu School of Computing (College X), National Cheng Kung University (NCKU), Taiwan. Before joining NCKU, he held research positions at Virginia Tech, Arlington, VA, USA (2015-2016), NTHU (2016-2017), The Chinese University of Hong Kong, HK (2014 and 2017), and the University of Lisbon (ULisboa), Lisbon, Portugal (2017-2018). He was an Assistant Professor with the Center for Space and Remote Sensing Research, National Central University, Taiwan, in 2018, and a Visiting Professor with ULisboa in 2019.

 

Dr. Lin received the Prize Paper Award for Interactive Session from IEEE Geoscience and Remote Sensing Society (GRS-S), in 2020, and The 3rd Place (Objective) and 5th Place (Subjective) from AIM Real World Super-Resolution Challenge at IEEE International Conference on Computer Vision (ICCV), in 2019. He received the Ministry of Science and Technology (MOST) Young Scholar Fellowship, together with the Einstein Grant Award, for the period of 2018–2023. In 2016, he was a recipient of the Outstanding Doctoral Dissertation Award from the Chinese Image Processing and Pattern Recognition Society and the Best Doctoral Dissertation Award from the IEEE GRS-S.