Abstract:In this paper, we propose a method that separates specular reflections from polarized color images. With line constraints in RGB color space and spatial do- main information, we used a Variational energy mini- mization approach to restore diffuse reflections. With anisotropic smoothing, our method preserves the shape of textures and successfully separates specular reflec- tions. Also, saturation of specular reflections is consid- ered using a simple inpainting method. space into two subspaces in each pixel: the specular line space and the diffuse plane space. The line space is parallel to the constraint line of polarization in each pixel, while the plane space represents a plane per- pendicular to the specular direction. In specular line space, our energy functional smoothes anisotropically the spatial variation of specular components depending on boundary values, and the direction of the smooth- ing is controlled by gradient information in the diffuse plane space. As a constraint, anisotropic homogeneous smoothness of the separated specular reflections is em- ployed in our energy functional and controlled by the gradient information of the polarization-varying spec- ular intensity in the polarized images. The narrow dynamic range of sensor and high in- cident angle can cause saturations with erroneous line constraints. To overcome this, a simple constrained Total Variation inpainting(9) method is applied dur- ing preprocessing. With this approach, our algorithm can deal with saturated specular images. In the cn- ergy minimization process following preprocessing, our method successfully separates specular reflections from the polarized images, which overall tcxtures and diffuse reflections.

Chromaticity-based separation of reflection components in a single image

Simple and Efficient Method for Specularity Removal in an Image

Separation Of Diffuse And Specular Reflection In Color Images

Saturation-preserving Specular Reflection Separation

Analysis and Synthesis of Multicolored Objects in A Single Image

Single Image Highlight Removal for Real-Time Image Processing Pipelines

Polarization Guided Specular Reflection Separation

Single Image Reflection Removal with Absorption Effect

An Illumination Distribution Preserved Colour Substitution Algorithm Based on Dichromatic Reflection Model

Highlight detection and removal based on chromaticity

Transferring Color Between Three-Dimensional Objects

Real-time highlight removal using intensity ratio

Dichromatic Based Rendering of Texture Images with High Color Fidelity

Reflection Component Separation Using Statistical Analysis and Polarisation.

Estimation of Spectral Reflectance of Object Surfaces with the Consideration of Perceptual Color Space

Separating Reflections From A Single Image Using Spatial Smoothness And Structure Information

Fast and High Quality Highlight Removal from A Single Image

Specular Highlight Removal for Real‐world Images

Variational Specular Separation Using Color and Polarization.

Accurate Image Specular Highlight Removal Based on Light Field Imaging

Light Field Imaging Based Accurate Image Specular Highlight Removal