Presentation + Paper
14 May 2019 Deep learning for fast super-resolution reconstruction from multiple images
Author Affiliations +
Abstract
Recent advancements in single-image super-resolution reconstruction (SRR) are attributed primarily to convolutional neural networks (CNNs), which effectively learn the relation between low and high resolution and allow for obtaining high-quality reconstruction within seconds. SRR from multiple images benefits from information fusion, which improves the reconstruction outcome compared with example-based methods. On the other hand, multiple-image SRR is computationally more demanding, mainly due to required subpixel registration of the input images. Here, we explore how to exploit CNNs in multiple-image SRR and we demonstrate that competitive reconstruction outcome can be obtained within seconds.
Conference Presentation
© (2019) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Michal Kawulok, Pawel Benecki, Krzysztof Hrynczenko, Daniel Kostrzewa, Szymon Piechaczek, Jakub Nalepa, and Bogdan Smolka "Deep learning for fast super-resolution reconstruction from multiple images", Proc. SPIE 10996, Real-Time Image Processing and Deep Learning 2019, 109960B (14 May 2019); https://doi.org/10.1117/12.2519579
Lens.org Logo
CITATIONS
Cited by 7 scholarly publications.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Image processing

Super resolution

Convolution

Image fusion

Image registration

Image filtering

Convolutional neural networks

Back to Top