{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/learned-spectral-super-resolution","title":"Learned Spectral Super-Resolution","arxiv_id":"1703.09470","date":"2017-03-28","proceeding":null,"authors":["Silvano Galliani","Charis Lanaras","Dimitrios Marmanis","Emmanuel Baltsavias","Konrad Schindler"],"abstract":"We describe a novel method for blind, single-image spectral super-resolution.\nWhile conventional super-resolution aims to increase the spatial resolution of\nan input image, our goal is to spectrally enhance the input, i.e., generate an\nimage with the same spatial resolution, but a greatly increased number of\nnarrow (hyper-spectral) wave-length bands. Just like the spatial statistics of\nnatural images has rich structure, which one can exploit as prior to predict\nhigh-frequency content from a low resolution image, the same is also true in\nthe spectral domain: the materials and lighting conditions of the observed\nworld induce structure in the spectrum of wavelengths observed at a given\npixel. Surprisingly, very little work exists that attempts to use this\ndiagnosis and achieve blind spectral super-resolution from single images. We\nstart from the conjecture that, just like in the spatial domain, we can learn\nthe statistics of natural image spectra, and with its help generate finely\nresolved hyper-spectral images from RGB input. Technically, we follow the\ncurrent best practice and implement a convolutional neural network (CNN), which\nis trained to carry out the end-to-end mapping from an entire RGB image to the\ncorresponding hyperspectral image of equal size. We demonstrate spectral\nsuper-resolution both for conventional RGB images and for multi-spectral\nsatellite data, outperforming the state-of-the-art.","url_abs":"http://arxiv.org/abs/1703.09470v1","url_pdf":"http://arxiv.org/pdf/1703.09470v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"learned-spectral-super-resolution","repo_url":"https://github.com/Intelligent-Imaging-Center/Spectral-Reconstruction","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"spectral-super-resolution","task_name":"Spectral Super-Resolution"},{"task_slug":"super-resolution","task_name":"Super-Resolution"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1703.09470","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}