Deeply Learned Filter Response Functions for Hyperspectral Reconstruction

Nie, Shijie; Gu, Lin; Zheng, Yinqiang; Lam, Antony; Ono, Nobutaka; Sato, Iwao

doi:10.1109/cvpr.2018.00501

Cited by 76 publications

(63 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such transformation is the nonlocal Laplacian method that has been widely used in image processing [31,32], which can also be called graph Laplacian and behaves as the discrete approximation to the second term in (5). We denote it as GL to simplify the notation:…”

Section: Nonlocal Second-order Regularizer For Hsi Inpaintingmentioning

confidence: 99%

“…The former is dominated by the well-known deep learning-based methods, which enjoy high popularity due to their strong power on nonlinear learning capabilities. In practice, several network architecutres, such as convolutional neural nets (CNNs) [5,6] and generative adversarial nets (GANs) [7], have been successfully extended to the field of hyperspectral image reconstruction. However, although excellent performance can be achieved, deep learning-based methods suffer from two drawbacks that hinder its usage in singleton inpainting problem, especially in non-commecial home computers.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Manifold-Based Nonlocal Second-Order Regularization for Hyperspectral Image Inpainting

Zheng

Jiang

et al. 2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Section: Nonlocal Second-order Regularizer For Hsi Inpaintingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Manifold-Based Nonlocal Second-Order Regularization for Hyperspectral Image Inpainting

Zheng

Jiang

et al. 2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

“…In the area of color imaging, Henz et al [40] proposed an overall color filter array optimization method using deep convolutional neural networks. Similarly, Nie et al [41] jointly optimized filters and recovered spectral images by optimizing the weights in CNN. Li et al [42] adopted sparse representation to model the pipeline of color imaging and demosaicking, and then optimized filters arrays via minimizing mutual coherence.…”

Section: Related Workmentioning

confidence: 99%

Optimized Multi-Spectral Filter Arrays for Spectral Reconstruction

Xie

et al. 2019

Sensors

View full text Add to dashboard Cite

Multispectral filter array (MSFA)-based imaging is a compact, practical technique for snapshot spectral image capturing and reconstruction. The imaging and reconstruction quality is highly influenced by the spectral sensitivities and spatial arrangement of channels on MSFAs, and the used reconstruction method. In order to design a MSFA with high imaging capacity, we propose a sparse representation based approach to optimize spectral sensitivities and spatial arrangement of MSFAs. The proposed approach first overall models the various errors associated with spectral reconstruction, and then uses a global heuristic searching method to optimize MSFAs via minimizing the estimated error of MSFAs. Our MSFA optimization method can select filters from off-the-shelf candidate filter sets while assigning the selected filters to the designed MSFA. Experimental results on three datasets show that the proposed method is more efficient, flexible, and can design MSFAs with lower spectral construction errors when compared with existing state-of-the-art methods. The MSFAs designed by our method show better performance than others even using different spectral reconstruction methods.

show abstract

“…However, this supposition is not true since cameras generally have different spectral sensitivity functions some of which are not similar to CIE [3]. This causes the implementations to produce inaccurate re- sults [24]. Therefore, the sensitivity function of the camera must be used together with RGB data to perform spectral reconstruction.…”

Section: Introductionmentioning

confidence: 99%

Towards Spectral Estimation from a Single RGB Image in the Wild

Kaya

Can

Timofte

2019

2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)

View full text Add to dashboard Cite

In contrast to the current literature, we address the problem of estimating the spectrum from a single common trichromatic RGB image obtained under unconstrained settings (e.g. unknown camera parameters, unknown scene radiance, unknown scene contents). For this we use a reference spectrum as provided by a hyperspectral image camera, and propose efficient deep learning solutions for sensitivity function estimation and spectral reconstruction from a single RGB image. We further expand the concept of spectral reconstruction such that to work for RGB images taken in the wild and propose a solution based on a convolutional network conditioned on the estimated sensitivity function. Besides the proposed solutions, we study also generic and sensitivity specialized models and discuss their limitations. We achieve state-of-the-art competitive results on the standard example-based spectral reconstruction benchmarks: ICVL, CAVE, NUS and NTIRE. Moreover, our experiments show that, for the first time, accurate spectral estimation from a single RGB image in the wild is within our reach. 1

show abstract

Deeply Learned Filter Response Functions for Hyperspectral Reconstruction

Cited by 76 publications

References 26 publications

Manifold-Based Nonlocal Second-Order Regularization for Hyperspectral Image Inpainting

Manifold-Based Nonlocal Second-Order Regularization for Hyperspectral Image Inpainting

Optimized Multi-Spectral Filter Arrays for Spectral Reconstruction

Towards Spectral Estimation from a Single RGB Image in the Wild

Contact Info

Product

Resources

About