{"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/multimodal-image-super-resolution-via-joint","title":"Multimodal Image Super-resolution via Joint Sparse Representations induced by Coupled Dictionaries","arxiv_id":"1709.08680","date":"2017-09-25","proceeding":null,"authors":["Pingfan Song","Xin Deng","João F. C. Mota","Nikos Deligiannis","Pier Luigi Dragotti","Miguel R. D. Rodrigues"],"abstract":"Real-world data processing problems often involve various image modalities\nassociated with a certain scene, including RGB images, infrared images or\nmulti-spectral images. The fact that different image modalities often share\ncertain attributes, such as certain edges, textures and other structure\nprimitives, represents an opportunity to enhance various image processing\ntasks. This paper proposes a new approach to construct a high-resolution (HR)\nversion of a low-resolution (LR) image given another HR image modality as\nreference, based on joint sparse representations induced by coupled\ndictionaries. Our approach, which captures the similarities and disparities\nbetween different image modalities in a learned sparse feature domain in\n\\emph{lieu} of the original image domain, consists of two phases. The coupled\ndictionary learning phase is used to learn a set of dictionaries that couple\ndifferent image modalities in the sparse feature domain given a set of training\ndata. In turn, the coupled super-resolution phase leverages such coupled\ndictionaries to construct a HR version of the LR target image given another\nrelated image modality. One of the merits of our sparsity-driven approach\nrelates to the fact that it overcomes drawbacks such as the texture copying\nartifacts commonly resulting from inconsistency between the guidance and target\nimages. Experiments on real multimodal images demonstrate that incorporating\nappropriate guidance information via joint sparse representation induced by\ncoupled dictionary learning brings notable benefits in the super-resolution\ntask with respect to the state-of-the-art. Of particular relevance, the\nproposed approach also demonstrates better robustness than competing\ndeep-learning-based methods in the presence of noise.","url_abs":"http://arxiv.org/abs/1709.08680v2","url_pdf":"http://arxiv.org/pdf/1709.08680v2.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":"multimodal-image-super-resolution-via-joint","repo_url":"https://github.com/pingfansong/CDLSR","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"dictionary-learning","task_name":"Dictionary Learning"},{"task_slug":"image-super-resolution","task_name":"Image Super-Resolution"},{"task_slug":"super-resolution","task_name":"Super-Resolution"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}