Fast super-resolution via dense local training and inverse regressor search

Regression-based Super-Resolution (SR) addresses the upscaling problem by learning a mapping function (i.e. regressor) from the low-resolution to the high-resolution manifold. Under the locally linear assumption, this complex non-linear mapping can be properly modeled by a set of linear regressors d...

Descripción completa

Detalles Bibliográficos
Autores: Pérez-Pellitero, Eduardo, Salvador, Jordi, Torres-Xirau, Iban, Ruiz Hidalgo, Javier|||0000-0001-6774-685X, Rosenhahn, Bodo
Tipo de recurso: artículo
Fecha de publicación:2014
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/27529
Acceso en línea:https://hdl.handle.net/2117/27529
https://dx.doi.org/10.1007/978-3-319-16811-1_23
Access Level:acceso abierto
Palabra clave:Computer vision
Image processing
Visió per ordinador
Imatges -- Processament
Àrees temàtiques de la UPC::Enginyeria de la telecomunicació::Processament del senyal::Processament de la imatge i del senyal vídeo
Àrees temàtiques de la UPC::So, imatge i multimèdia::Creació multimèdia::Imatge digital
Descripción
Sumario:Regression-based Super-Resolution (SR) addresses the upscaling problem by learning a mapping function (i.e. regressor) from the low-resolution to the high-resolution manifold. Under the locally linear assumption, this complex non-linear mapping can be properly modeled by a set of linear regressors distributed across the manifold. In such methods, most of the testing time is spent searching for the right regressor within this trained set. In this paper we propose a novel inverse-search approach for regression-based SR. Instead of performing a search from the image to the dictionary of regressors, the search is done inversely from the regressors’ dictionary to the image patches. We approximate this framework by applying spherical hashing to both image and regressors, which reduces the inverse search into computing a trained function. Additionally, we propose an improved training scheme for SR linear regressors which improves perceived and objective quality. By merging both contributions we improve speed and quality compared to the state-of-the-art.