Interreflection Removal Using Fluorescence

Fu, Ying; Lam, Antony; Matsushita, Yasuyuki; Sato, Iwao; Sato, Yoichi

doi:10.1007/978-3-319-10602-1_14

Cited by 13 publications

(18 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This phenomenon has been studied in the domain of computer vision, mainly in order to remove this effect from the images to be able to retrieve the shape of an object in an image (shape-from-shading methods) [1,2,4,[25][26][27]. Recently, interreflections have been exploited in the context of color camera calibration [28].…”

Section: Interreflectionsmentioning

confidence: 99%

Deep spectral reflectance and illuminant estimation from self-interreflections

et al. 2018

View full text Add to dashboard Cite

In this work, we propose a CNN-based approach to estimate the spectral reflectance of a surface and the spectral power distribution of the light from a single RGB image of a V-shaped surface. Interreflections happening in a concave surface lead to gradients of RGB values over its area. These gradients carry a lot of information concerning the physical properties of the surface and the illuminant. Our network is trained with only simulated data constructed using a physics-based interreflection model. Coupling interreflection effects with deep learning helps to retrieve the spectral reflectance under an unknown light and to estimate the spectral power distribution of this light as well. In addition, it is more robust to the presence of image noise than the classical approaches. Our results show that the proposed approach outperforms the state of the art learning-based approaches on simulated data. In addition, it gives better results on real data compared to other interreflection-based approaches.

show abstract

Section: Interreflectionsmentioning

confidence: 99%

Deep spectral reflectance and illuminant estimation from self-interreflections

et al. 2018

View full text Add to dashboard Cite

show abstract

“…3 State of the Art Many methods have been proposed in the literature in order to model interreflections in computer vision applications. Some approaches focus on removing the effect of interreflections [6,[10][11][12][13]16], whereas others use it as extra information that may help in solving other problems such as light SPD estimation or surface spectral reflectance estimation [9]. Moreover, the proposed systems differ in many ways regarding the number of color channels or wavelengths, the number of light bounces they handle, and the number and the type of assumptions they pose:…”

Section: Definitions and Notationsmentioning

confidence: 99%

“…The main applications for this research were in shape from shading [4,6,10] and color constancy domains [7][8][9]. Nevertheless, the research in this area has become less popular recently, focusing mainly on the adaptation of early proposed models for different types of applications and needs [11][12][13]. This seems to follow a general trend in computer vision, as recent research in this area is focusing more on pattern recognition approaches and less on physics-based approaches.…”

Section: Introductionmentioning

confidence: 99%

Interreflections in Computer Vision: A Survey and an Introduction to Spectral Infinite-Bounce Model

Deeb¹,

Muselet²,

Hébert³

et al. 2017

J Math Imaging Vis

View full text Add to dashboard Cite

Interreflections are observed on concave objects or when multiple objects are located closely. In a vision system, interreflections can largely affect color values captured by the camera. Due to this fact, modeling interreflections is important for many vision applications. In this paper, we consider the problem of treating and modeling interreflections in the domain of computer vision. First, a survey of existing approaches in the state of the art is given. These approaches are detailed, discussed and compared. Most of the state of the art models take into consideration only two bounces of light between surface elements. We, afterwards, introduce a new interreflection model based on radiometric definitions. This model is the first one that takes into consideration an infinite number of light bounces between surface elements while providing image RGB values as a result. The accuracy of our model is studied by comparing it to real camera outputs. Thanks to our new model, the importance of using infinite bounces of light while studying interreflection, instead of only two bounces is demonstrated.

show abstract

“…Lam and Sato [5] and Fu et al [6] predicted the appearance of relit fluorescent scenes after all information of spectra were recovered in the scene. Fu et al [25] removed scene interreflection using fluorescence.…”

Section: Related Workmentioning

confidence: 99%

Reflectance and Fluorescence Spectral Recovery via Actively Lit RGB Images

Lam

Sato

et al. 2016

IEEE Trans. Pattern Anal. Mach. Intell.

Self Cite

View full text Add to dashboard Cite

In recent years, fluorescence analysis of scenes has received attention in computer vision. Fluorescence can provide additional information about scenes, and has been used in applications such as camera spectral sensitivity estimation, 3D reconstruction, and color relighting. In particular, hyperspectral images of reflective-fluorescent scenes provide a rich amount of data. However, due to the complex nature of fluorescence, hyperspectral imaging methods rely on specialized equipment such as hyperspectral cameras and specialized illuminants. In this paper, we propose a more practical approach to hyperspectral imaging of reflective-fluorescent scenes using only a conventional RGB camera and varied colored illuminants. The key idea of our approach is to exploit a unique property of fluorescence: the chromaticity of fluorescent emissions are invariant under different illuminants. This allows us to robustly estimate spectral reflectance and fluorescent emission chromaticity. We then show that given the spectral reflectance and fluorescent chromaticity, the fluorescence absorption and emission spectra can also be estimated. We demonstrate in results that all scene spectra can be accurately estimated from RGB images. Finally, we show that our method can be used to accurately relight scenes under novel lighting.

show abstract

Interreflection Removal Using Fluorescence

Cited by 13 publications

References 23 publications

Deep spectral reflectance and illuminant estimation from self-interreflections

Deep spectral reflectance and illuminant estimation from self-interreflections

Interreflections in Computer Vision: A Survey and an Introduction to Spectral Infinite-Bounce Model

Reflectance and Fluorescence Spectral Recovery via Actively Lit RGB Images

Contact Info

Product

Resources

About