Search

In Spectral Reconstruction (SR), we recover hyperspectral images from their RGB counterparts. Most of the recent approaches are based on Deep Neural Networks (DNN), where millions of parameters are trained mainly to extract and utilize the contextual features in large image patches as part of the SR process. On the other hand, the leading Sparse Coding method ‘A+’—which is among the strongest point-based baselines against the DNNs—seeks to divide the RGB space into neighborhoods, where locally a simple linear regression (comprised by roughly 102 parameters) suffices for SR. In this paper, we explore how the performance of Sparse Coding can be further advanced. We point out that in the original A+, the sparse dictionary used for neighborhood separations are optimized for the spectral data but used in the projected RGB space. In turn, we demonstrate that if the local linear mapping is trained for each spectral neighborhood instead of RGB neighborhood (and theoretically if we could recover each spectrum based on where it locates in the spectral space), the Sparse Coding algorithm can actually perform much better than the leading DNN method. In effect, our result defines one potential (and very appealing) upper-bound performance of point-based SR.

Spectral reconstruction algorithms recover spectra from RGB sensor responses. Recent methods&mdash, with the very best algorithms using deep learning&mdash, can already solve this problem with good spectral accuracy. However, the recovered spectra are physically incorrect in that they do not induce the RGBs from which they are recovered. Moreover, if the exposure of the RGB image changes then the recovery performance often degrades significantly&mdash, i.e., most contemporary methods only work for a fixed exposure. In this paper, we develop a physically accurate recovery method: the spectra we recover provably induce the same RGBs. Key to our approach is the idea that the set of spectra that integrate to the same RGB can be expressed as the sum of a unique fundamental metamer (spanned by the camera&rsquo, s spectral sensitivities and linearly related to the RGB) and a linear combination of a vector space of metameric blacks (orthogonal to the spectral sensitivities). Physically plausible spectral recovery resorts to finding a spectrum that adheres to the fundamental metamer plus metameric black decomposition. To further ensure spectral recovery that is robust to changes in exposure, we incorporate exposure changes in the training stage of the developed method. In experiments we evaluate how well the methods recover spectra and predict the actual RGBs and RGBs under different viewing conditions (changing illuminations and/or cameras). The results show that our method generally improves the state-of-the-art spectral recovery (with more stabilized performance when exposure varies) and provides zero colorimetric error. Moreover, our method significantly improves the color fidelity under different viewing conditions, with up to a 60% reduction in some cases.

This paper reviews the second challenge on spectral reconstruction from RGB images, i.e., the recovery of whole- scene hyperspectral (HS) information from a 3-channel RGB image. As in the previous challenge, two tracks were provided: (i) a "Clean" track where HS images are estimated from noise-free RGBs, the RGB images are themselves calculated numerically using the ground-truth HS images and supplied spectral sensitivity functions (ii) a "Real World" track, simulating capture by an uncalibrated and unknown camera, where the HS images are recovered from noisy JPEG-compressed RGB images. A new, larger-than-ever, natural hyperspectral image data set is presented, containing a total of 510 HS images. The Clean and Real World tracks had 103 and 78 registered participants respectively, with 14 teams competing in the final testing phase. A description of the proposed methods, alongside their challenge scores and an extensive evaluation of top performing methods is also provided. They gauge the state-of-the-art in spectral reconstruction from an RGB image.

Recently Convolutional Neural Networks (CNN) have been used to reconstruct hyperspectral information from RGB images. Moreover, this spectral reconstruction problem (SR) can often be solved with good (low) error. However, these methods are not physically plausible: that is when the recovered spectra are reintegrated with the underlying camera sensitivities, the resulting predicted RGB is not the same as the actual RGB, and sometimes this discrepancy can be large. The problem is further compounded by exposure change. Indeed, most learning-based SR models train for a fixed exposure setting and we show that this can result in poor performance when exposure varies. In this paper we show how CNN learning can be extended so that physical plausibility is enforced and the problem resulting from changing exposures is mitigated. Our SR solution improves the state-of-the-art spectral recovery performance under varying exposure conditions while simultaneously ensuring physical plausibility (the recovered spectra reintegrate to the input RGBs exactly).