Deep Amended Gradient Descent for Efficient Spectral Reconstruction from Single RGB Images

by   Zhiyu Zhu, et al.

This paper investigates the problem of recovering hyperspectral (HS) images from single RGB images. To tackle such a severely ill-posed problem, we propose a physically-interpretable, compact, efficient, and end-to-end learning-based framework, namely AGD-Net. Precisely, by taking advantage of the imaging process, we first formulate the problem explicitly based on the classic gradient descent algorithm. Then, we design a lightweight neural network with a multi-stage architecture to mimic the formed amended gradient descent process, in which efficient convolution and novel spectral zero-mean normalization are proposed to effectively extract spatial-spectral features for regressing an initialization, a basic gradient, and an incremental gradient. Besides, based on the approximate low-rank property of HS images, we propose a novel rank loss to promote the similarity between the global structures of reconstructed and ground-truth HS images, which is optimized with our singular value weighting strategy during training. Moreover, AGD-Net, a single network after one-time training, is flexible to handle the reconstruction with various spectral response functions. Extensive experiments over three commonly-used benchmark datasets demonstrate that AGD-Net can improve the reconstruction quality by more than 1.0 dB on average while saving 67× parameters and 32× FLOPs, compared with state-of-the-art methods. The code will be publicly available at


page 1

page 3

page 7

page 8

page 9


Semantic-embedded Unsupervised Spectral Reconstruction from Single RGB Images in the Wild

This paper investigates the problem of reconstructing hyperspectral (HS)...

Deep Posterior Distribution-based Embedding for Hyperspectral Image Super-resolution

In this paper, we investigate the problem of hyperspectral (HS) image sp...

MST++: Multi-stage Spectral-wise Transformer for Efficient Spectral Reconstruction

Existing leading methods for spectral reconstruction (SR) focus on desig...

Fast Fourier Intrinsic Network

We address the problem of decomposing an image into albedo and shading. ...

Snapshot Hyperspectral Imaging Based on Weighted High-order Singular Value Regularization

Snapshot hyperspectral imaging can capture the 3D hyperspectral image (H...

Learning to Recover Spectral Reflectance from RGB Images

This paper tackles spectral reflectance recovery (SRR) from RGB images. ...

Deep Diversity-Enhanced Feature Representation of Hyperspectral Images

In this paper, we study the problem of embedding the high-dimensional sp...

Please sign up or login with your details

Forgot password? Click here to reset