Kazunori Uruma

Fast communication: Iterative partial matrix shrinkage algorithm for matrix rank minimization

This paper proposes a new matrix shrinkage algorithm for matrix rank minimization problems. The proposed algorithm provides a low rank solution by estimating a matrix rank and shrinking non-dominant singular values iteratively. We study the convergence properties of the algorithm, which indicate that the algorithm gives approximate low-rank solutions. Numerical results show that the proposed algorithm works efficiently for hard problems with low computing time.

Image colorization algorithm using series approximated sparse function

This paper deals with an image colorization and proposes a new image colorization algorithm. Assuming that the difference of color values between neighbor pixels is given as a monotonically increasing function of the difference of grayscale values between neighbor pixels, a colorization function is proposed, and the colorization problem is formulated as the weighted least squares problem using this function. In order to reduce the dependence on the value of a parameter in the algorithm, this paper utilizes a finite series approximation and provides a fast colorization algorithm. Numerical examples show that the proposed algorithm colorizes a grayscale image efficiently.

Image colorization based on the mixed l 0 /l 1 norm minimization

This paper proposes a new image colorization algorithm based on the mixed L0/L1 norm minimization. Introducing some assumptions, a problem of recovering a color image from a grayscale image with the small number of known color pixels is formulated as a mixed L0/L1 norm minimization, which is solved approximately by an iterative reweighted least squares (IRLS) algorithm. Numerical examples show that the proposed algorithm colorizes a grayscale image well using a small number of color pixels.

An image colorization algorithm using sparse optimization

This paper proposes a new digital image colorization algorithm using the sparse optimization. We deal with the colorization problem where a grayscale image is colorized using a full color image with a similar composition, and formulate this problem as the sparse optimization problem. We also provide an iterative reweighted least squares (IRLS) algorithm to solve this problem approximately, and the full color image is obtained in practical time. Numerical examples show that the proposed algorithm colorizes a grayscale image well.

High resolution depth image recovery algorithm based on the modeling of the sum of an average distance image and a surface image

This paper proposes a new depth image recovery algorithm which recovers a high resolution depth image using RGB color image from a very low resolution depth image. In order to achieve a high recovery performance, this paper represents the high resolution depth image as the sum of an average distance image and a surface image. Experimental examples show that the proposed algorithm achieves a high resolution depth recovery from a very low resolution depth image effectively.

Color image coding based on the colorization algorithm using multiple resolution images

This paper proposes a representative pixels (RP) extraction and colorization algorithm for the colorization-based digital image coding. In order to achieve a low computing time and to reduce the amount of information to represent the RP, the proposed algorithm uses the multiple resolution images obtained by simply multiple downsampling. Numerical examples show that the proposed algorithm extracts the RP and colorizes the luminance image fast and effectively.

Adaptive image inpainting algorithm based on generalized principal component analysis

This paper proposes an image inpainting algorithm based on generalized principal component analysis. Several inpainting algorithms have been proposed based on the assumption that an image can be modeled by the autoregressive (AR) model. However, their performances are not good enough to apply to natural photographs because they assume that images are modeled by the position-invariant linear model. To improve the inpainting quality, this work introduces a multiple AR model based inpainting based on the generalized principle component analysis (GPCA) and proposes a new multiple matrix rank minimization approach. A practical algorithm is provided based on the iterative partial matrix shrinkage (IPMS) algorithm, and numerical examples show that the effectiveness of the proposed algorithm.

High resolution depth image recovery algorithm using grayscale image

This paper proposes a depth image recovery algorithm which recovers depth images using grayscale images and low resolution depth images. Based on a image colorization technique, a depth value image recovery problem is formulated as a convex quadratic optimization problem, and a fast depth image recovery algorithm is proposed. Experimental results show that the proposed algorithm recovers a high resolution depth image from a very low resolution depth image effectively.

Subspace clustering and multiple matrix rank minimization approach to image inpainting algorithm

This paper proposes an image inpainting algorithm based on generalized principal component analysis (GPCA). Several inpainting algorithms have been proposed based on the assumption that an image can be modeled by the autoregressive (AR) model. However, their performances are not good enough to apply to natural photographs because they assume that images are modeled by the position-invariant linear model. To improve the inpainting quality, this work introduces a multiple AR model approach to image inpainting based on the generalized principle component analysis and proposes a new multiple matrix rank minimization approach. A practical algorithm is provided based on the iterative partial matrix shrinkage (IPMS) algorithm, and numerical examples show that the effectiveness of the proposed algorithm.

Singular values estimation approach to single-frame super-resolution

This paper deals with the problem of reconstructing a high-resolution digital image from a single low-resolution digital image and proposes a singular values estimation approach to the super-resolution. The proposed method estimates singular values in order to restore high-frequency components by using spline interpolation. Numerical examples show that the proposed method can restore high-resolution images efficiently.