Sparse representation and learning in visual recognition: Theory and applications

Ubiquitous motion blur easily fails multi-frame super-resolution (MFSR). Our method proposed in this paper tackles this issue by optimally searching least blurred pixels in MFSR. An EM framework is proposed to guide residual blur estimation and high-resolution image reconstruction. To suppress noise, we employ a family of sparse penalties as natural image priors, along with an effective solver. Theoretical analysis is performed on how and when our method works. The relationship between estimation errors of motion blur and the quality of input images is discussed. Our method produces sharp and higher-resolution results given input of challenging low-resolution noisy and blurred sequences.

/pdf/handling-motion-blur-in-multi-frame-super-resolution-4wsivug2m9.pdf

Handling motion blur in multi-frame super-resolution

This paper comprehensively reviews the recent development of image deblurring, including nonblind/blind, spatially invariant/variant deblurring techniques. Indeed, these techniques share the same objective of inferring a latent sharp image from one or several corresponding blurry images, while the blind deblurring techniques are also required to derive an accurate blur kernel. Considering the critical role of image restoration in modern imaging systems to provide high-quality images under complex environments such as motion, undesirable lighting conditions, and imperfect system components, image deblurring has attracted growing attention in recent years. From the viewpoint of how to handle the illposedness which is a crucial issue in deblurring tasks, existing methods can be grouped into ve categories: Bayesian inference framework, variational methods, sparse representation-based methods, homographybased modeling, and region-based methods. In spite of achieving a certain level of development, image deblurring, especially the blind case, is limited in its success by complex application conditions which make the blur kernel hard to obtain and be spatially variant. We provide a holistic understanding and deep insight into image deblurring in this review. An analysis of the empirical evidence for representative methods, practical issues, as well as a discussion of promising future directions are also presented.

/pdf/recent-progress-in-image-deblurring-3v4bl9vvjr.pdf

Recent Progress in Image Deblurring

Handling Motion Blur inMulti-Frame Super-Resolution

In this paper we present a new patch-based empirical Bayesian video denoising algorithm. The method builds a Bayesian model for each group of similar space-time patches. These patches are not motion-compensated, and therefore avoid the risk of inaccuracies caused by motion estimation errors. The high dimensionality of spatiotemporal patches together with a limited number of available samples poses challenges when estimating the statistics needed for an empirical Bayesian method. We therefore assume that groups of similar patches have a low intrinsic dimensionality, leading to a spiked covariance model. Based on theoretical results about the estimation of spiked covariance matrices, we propose estimators of the eigenvalues of the a priori covariance in high-dimensional spaces as simple corrections of the eigenvalues of the sample covariance matrix. We demonstrate empirically that these estimators lead to better empirical Wiener filters. A comparison on classic benchmark videos demonstrates improved visual quality and an increased PSNR with respect to state-of-the-art video denoising methods.

https://hal.archives-ouvertes.fr/hal-01674474/document

Video Denoising via Empirical Bayesian Estimation of Space-Time Patches

Camera shake during long exposure is ineluctable in light-limited situations, and results in a blurry observation. Recovering the blur kernel and the latent image from the blurred image is an inherently ill-posed problem. In this paper, we analyze the image acquisition model to capture two blurred images simultaneously with different blur kernels. The image pair is well-aligned and the kernels have a certain relationship. Such strategy overcomes the challenge of blurry image alignment and reduces the ambiguity of blind deblurring. Thanks to the aided hardware, the algorithm based on such image pair can give high-quality kernel estimation and image restoration. The experiments on both synthetic and real images demonstrate the effectiveness of our image capture strategy, and show that the kernel estimation is accurate enough to restore superior latent image, which contains more details and fewer ringing artifacts.

Exploring aligned complementary image pair for blind motion deblurring

Robust blind motion deblurring using near-infrared flash image

We present an effective patch-based video denoising algorithm that exploits both local and nonlocal correlations. The method groups 3D shape-adaptive patches, whose surrounding cubic neighborhoods along spatial and temporal dimensions have been found similar by patch clustering. Such grouping results in 4D data structures with arbitrary shapes. Since the obtained 4D groups are highly correlated along all the dimensions, they can be represented very sparsely with a 4D shape-adaptive DCT. The noise can be effectively attenuated by transform shrinkage. Experimental results on a wide range of videos show that this algorithm provides significant improvement over the state-of-the-art denoising algorithms in terms of both objective metric and subjective visual quality.

Video denoising using shape-adaptive sparse representation over similar spatio-temporal patches

The invention provides a complementary blurred image acquisition system and a blurred image recovery method using the complementary blurred image acquisition system The system comprises a spectroscope, a first prism combination and a second prism combination The method comprises the following steps of: establishing the complementary blurred image acquisition system; acquiring two degraded images of the same scene under the influence of motion blur; obtaining an initial blur core and the estimation of a clear image by taking the relationship between two blur cores as a constraint; thinning the blur cores; and performing deconvolution on the image by the finally obtained blur cores to obtain an ideal image deblurring result According to the complementary blurred image acquisition system, the images of the two groups of prisms are arranged side by side, and images of a 90-degree rotation relationship can be shot by a camera The motion blurred image recovery method is simple, effective, and high in robustness and real-time performance; and on the premise of low complexity, the recovery effect of the motion blurred image can be improved remarkably

Complementary blurred image acquisition system and blurred image recovery method using complementary blurred image acquisition system

The invention provides a non-uniform motion blurred image restoration method. The method includes selecting a plurality of overlapped image blocks in a blurred image; figuring out and selecting partial blur kernels of the plurality of the image blocks and deblurred image blocks by the uniform blurred image restoration method; establishing corresponding relationship between the partial blur kernels of the plurality of the image blocks and camera global motion parameters according to the partial blur kernels of the plurality of the image blocks and the deblurred image blocks, and estimating initial camera global motion parameters according to the corresponding relationship; iterating the initial camera global motion parameters and the deblurred image blocks to acquire an estimated final camera global motion parameters; calculating the blurred image to acquire a final deblurred image according to the final camera global motion parameter. The embodiment of the method is based on a motion blurred image imaging model, the corresponding relationship between the partial blur kernels and the camera global motion parameters is combined, and rapid, effective, robust low-error blurred image restoration can be realized.

Wen Li

Papers

Exploring aligned complementary image pair for blind motion deblurring

Robust blind motion deblurring using near-infrared flash image

Video denoising using shape-adaptive sparse representation over similar spatio-temporal patches

Complementary blurred image acquisition system and blurred image recovery method using complementary blurred image acquisition system

Non-uniform motion blurred image restoration method