Boaz Arad (Ben-Gurion University)
Tuesday, 29.11.2016, 11:30
Hyperspectral (HS) images or "hyperspectral data-cubes" contain radiance spectrum information at high spectral resolution for each point in the scene. Until recently, acquiring such information involved expensive, bulky equipment which required long exposure times - making HS imaging impractical for "natural" imaging (ground-level, horizontally viewed scenes). The talk will present a novel methodology, allowing high-accuracy estimation of HS information from unseen scenes using only RGB data acquired from a consumer-grade camera. This is achieved by collection of a highly generalizable HS prior and leveraging the sparsity of HS signals via overcomplete dictionaries.
(see ECCV2016: "Sparse Recovery of Hyperspectral Signal from Natural RGB Images")