Abstract:Due to the physical nature of color images, color image processing such as denoising and inpainting has shown extensive and versatile possibilities over grayscale image processing. The monochromatic and the concatenation model have been widely used to process color images by processing each color channel independently or concatenating three color channels as one unified one and then used existing grayscale image processing methods directly without specific operations. These above schemes, however, have some limitations: (1) they would destroy the inherent correlation among three color channels since they cannot represent color images holistically; (2) they usually focus on one specific handcrafted prior such as smoothness, low-rankness, or even deep prior and thus failing to fuse deep and handcrafted priors of color images flexibly. To conquer these limitations, we propose one unified model to integrate deep prior and low-rank quaternion prior (DLRQP) for color image processing under the plug-and-play (PnP) framework. Specifically, the quaternion representation with low-rank constraint is introduced to denote the color image in a holistic way and one advanced denoiser is adopted to explore the deep prior in an iterative process. To tightly approximate the quaternion rank, one nonconvex penalty function is further utilized. We derive an alternate iterative approach to tackle the proposed model. We empirically demonstrate that our model can achieve superior performance over existing methods on both color image denoising and inpainting tasks.

Color Occlusion Face Recognition Method Based on Quaternion Non-Convex Sparse Constraint Mechanism

Local Spatial Continuity Steered Sparse Representation for Occluded Face Recognition

Face recognition with contiguous occlusion using linear regression and level set method

Color Face Recognition Based on Quaternion Zernike Moment Invariants and Quaternion BP Neural Network

Cancelable color face recognition using trinion gyrator transform and randomized nonlinear PCANet

Color face recognition based on quaternion matrix representation

Quasi Non-Negative Quaternion Matrix Factorization with Application to Color Face Recognition

Color Face Recognition using High-Dimension Quaternion-based Adaptive Representation

Color Facial Expression Recognition by Quaternion Convolutional Neural Network With Gabor Attention

Generalized Two-Dimensional Quaternion Principal Component Analysis with Weighting for Color Image Recognition

Color Image Classification Via Quaternion Principal Component Analysis Network.

Learning Spatially Decoupled Color Representations for Facial Image Colorization

Quaternion matrix regression for color face recognition

Face Detection Based on Vector Quantization in Color Images

Obtain Method of Quaternion Matrix Orthogonal Eigenvector Set and Its Application in Color Face Recognition

Study of Occluded Robust Face Recognition Approach Based On Homotopy Algorithm and Color Information Fusion

Quaternion Scalar and Vector Norm Decomposition: Quaternion PCA for Color Face Recognition

Quaternion-based 2D-DOST and Stacked Principal Component Analysis Network for Multimodal Face Recognition

Quaternion Based Maximum Margin Criterion Method for Color Face Recognition

Face Recognition Against Occlusions Via Colour Fusion Using 2D-MCF Model and SRC

Deep and Low-Rank Quaternion Priors for Color Image Processing