Fast-N-Squeeze: Towards Real-Time Spectral Reconstruction From RGB Images

Mirko Agarla, Simone Bianco, Marco Buzzelli, Luigi Celona, Raimondo Schettini; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2022, pp. 1132-1139

Abstract

We present an efficient method for the reconstruction of multispectral information from RGB images, as part of the NTIRE 2022 Spectral Reconstruction Challenge. Given an input image, our method determines a global RGB-to-spectral linear transformation matrix, based on a search through optimal matrices from training images that share low-level features with the input. The resulting spectral signatures are then adjusted by a global scaling factor, determined through a lightweight SqueezeNet-inspired neural network. By combining the efficiency of linear transformation matrices with the data-driven effectiveness of convolutional neural networks, we are able to achieve superior performance than winners of the previous editions of the challenge.

Related Material

[pdf]
[bibtex]
@InProceedings{Agarla_2022_CVPR, author = {Agarla, Mirko and Bianco, Simone and Buzzelli, Marco and Celona, Luigi and Schettini, Raimondo}, title = {Fast-N-Squeeze: Towards Real-Time Spectral Reconstruction From RGB Images}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2022}, pages = {1132-1139} }