English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

High-resolution (HR) hyperspectral face image plays an important role in facerelated computer vision tasks under uncontrolled conditions, such as low-lightenvironment and spoofing attacks. However, the dense spectral bands ofhyperspectral face images come at the cost of limited amount of photons reacheda narrow spectral window on average, which greatly reduces the spatialresolution of hyperspectral face images. In this paper, we investigate how toadapt the deep learning techniques to hyperspectral face image super-resolution(HFSR), especially when the training samples are very limited. Benefiting fromthe amount of spectral bands, in which each band can be seen as an image, wepresent a spectral splitting and aggregation network (SSANet) for HFSR withlimited training samples. In the shallow layers, we split the hyperspectralimage into different spectral groups. Then, we gradually aggregate the neighborbands at deeper layers to exploit spectral correlations. By this spectralsplitting and aggregation strategy (SSAS), we can divide the originalhyperspectral image into multiple samples (\emph{from less to more}) to supportthe efficient training of the network and effectively exploit the spectralcorrelations among spectrum. To cope with the challenge of small trainingsample size (S3) problem, we propose to expand the training samples by aself-representation model and symmetry-induced augmentation. Experiments showthat SSANet can well model the joint correlations of spatial and spectralinformation. By expanding the training samples, SSANet can effectivelyalleviate the S3 problem.

From Less to More: Spectral Splitting and Aggregation Network for Hyperspectral Face Super-Resolution