Image super-resolution remains an important research topic to overcome the limitations of physical acquisition systems, and to support the development of high resolution displays. Previous example-based super-resolution approaches mainly focus on analyzing the co-occurrence properties of low resolution and high-resolution patches. Recently, we proposed a novel single image super-resolution approach based on linear manifold approximation of the high-resolution image-patch space [1]. The image super-resolution problem is then formulated as an optimization problem of searching for the best matched high resolution patch in the manifold for a given low-resolution patch. We developed a novel technique based on the l1 norm sparse graph to learn a set of low dimensional affine spaces or tangent subspaces of the high-resolution patch manifold. The optimization problem is then solved based on the learned set of tangent subspaces. In this paper, we build on our recent work as follows. First, we consider and analyze each tangent subspace as one point in a Grassmann manifold, which helps to compute geodesic pairwise distances among these tangent subspaces. Second, we develop a min-max algorithm to select an optimal subset of tangent subspaces. This optimal subset reduces the computational cost while still preserving the quality of the reconstructed high-resolution image. Third, and to further achieve lower computational complexity, we perform hierarchical clustering on the optimal subset based on Grassmann manifold distances. Finally, we analytically prove the validity of the proposed Grassmann-distance based clustering. A comparison of the obtained results with other state-of-the-art methods clearly indicates the viability of the new proposed framework.
Deep learning has been successfully applied to image super resolution (SR). In this paper, we propose a deep joint super resolution (DJSR) model to exploit both external and self similarities for SR. A Stacked Denoising Convolutional Auto Encoder (SDCAE) is first pre-trained on external examples with proper data augmentations. It is then fine-tuned with multi-scale self examples from each input, where the reliability of self examples is explicitly taken into account. We also enhance the model performance by sub-model training and selection. The DJSR model is extensively evaluated and compared with state-of-the-arts, and show noticeable performance improvements both quantitatively and perceptually on a wide range of images.
Text image super-resolution is a challenging yet open research problem in the computer vision community. In particular, low-resolution images hamper the performance of typical optical character recognition (OCR) systems. In this article, we summarize our entry to the ICDAR2015 Competition on Text Image Super-Resolution. Experiments are based on the provided ICDAR2015 TextSR dataset and the released Tesseract-OCR 3.02 system. We report that our winning entry of text image super-resolution framework has largely improved the OCR performance with low-resolution images used as input, reaching an OCR accuracy score of 77.19%, which is comparable with that of using the original high-resolution images 78.80%.
Single image super-resolution (SR) aims to estimate a high-resolution (HR) image from a lowresolution (LR) input. Image priors are commonly learned to regularize the otherwise seriously ill-posed SR problem, either using external LR-HR pairs or internal similar patterns. We propose joint SR to adaptively combine the advantages of both external and internal SR methods. We define two loss functions using sparse coding based external examples, and epitomic matching based on internal examples, as well as a corresponding adaptive weight to automatically balance their contributions according to their reconstruction errors. Extensive SR results demonstrate the effectiveness of the proposed method over the existing state-of-the-art methods, and is also verified by our subjective evaluation study.
We propose a deep learning method for single image super-resolution (SR). Our method directly learns an end-to-end mapping between the low/high-resolution images. The mapping is represented as a deep convolutional neural network (CNN) that takes the low-resolution image as the input and outputs the high-resolution one. We further show that traditional sparse-coding-based SR methods can also be viewed as a deep convolutional network. But unlike traditional methods that handle each component separately, our method jointly optimizes all layers. Our deep CNN has a lightweight structure, yet demonstrates state-of-the-art restoration quality, and achieves fast speed for practical on-line usage. We explore different network structures and parameter settings to achieve trade-offs between performance and speed. Moreover, we extend our network to cope with three color channels simultaneously, and show better overall reconstruction quality.
Deep learning techniques have been successfully applied in many areas of computer vision, including low-level image restoration problems. For image super-resolution, several models based on deep neural networks have been recently proposed and attained superior performance that overshadows all previous handcrafted models. The question then arises whether large-capacity and data-driven models have become the dominant solution to the ill-posed super-resolution problem. In this paper, we argue that domain expertise represented by the conventional sparse coding model is still valuable, and it can be combined with the key ingredients of deep learning to achieve further improved results. We show that a sparse coding model particularly designed for super-resolution can be incarnated as a neural network, and trained in a cascaded structure from end to end. The interpretation of the network based on sparse coding leads to much more efficient and effective training, as well as a reduced model size. Our model is evaluated on a wide range of images, and shows clear advantage over existing state-of-the-art methods in terms of both restoration accuracy and human subjective quality.
A trainable filter-based higher-order Markov Random Fields (MRFs) model - the so called Fields of Experts (FoE), has proved a highly effective image prior model for many classic image restoration problems. Generally, two options are available to incorporate the learned FoE prior in the inference procedure: (1) sampling-based minimum mean square error (MMSE) estimate, and (2) energy minimization-based maximum a posteriori (MAP) estimate. This letter is devoted to the FoE prior based single image super resolution (SR) problem, and we suggest to make use of the MAP estimate for inference based on two facts: (I) It is well-known that the MAP inference has a remarkable advantage of high computational efficiency, while the sampling-based MMSE estimate is very time consuming. (II) Practical SR experiment results demonstrate that the MAP estimate works equally well compared to the MMSE estimate with exactly the same FoE prior model. Moreover, it can lead to even further improvements by incorporating our discriminatively trained FoE prior model. In summary, we hold that for higher-order natural image prior based SR problem, it is better to employ the MAP estimate for inference.
In this paper we present seven techniques that everybody should know to improve example-based single image super resolution (SR): 1) augmentation of data, 2) use of large dictionaries with efficient search structures, 3) cascading, 4) image self-similarities, 5) back projection refinement, 6) enhanced prediction by consistency check, and 7) context reasoning. We validate our seven techniques on standard SR benchmarks (i.e. Set5, Set14, B100) and methods (i.e. A+, SRCNN, ANR, Zeyde, Yang) and achieve substantial improvements.The techniques are widely applicable and require no changes or only minor adjustments of the SR methods. Moreover, our Improved A+ (IA) method sets new state-of-the-art results outperforming A+ by up to 0.9dB on average PSNR whilst maintaining a low time complexity.
We study the problem of evaluating super resolution methods. Traditional evaluation methods usually judge the quality of super resolved images based on a single measure of their difference with the original high resolution images. In this paper, we proposed to use both fidelity (the difference with original images) and naturalness (human visual perception of super resolved images) for evaluation. For fidelity evaluation, a new metric is proposed to solve the bias problem of traditional evaluation. For naturalness evaluation, we let humans label preference of super resolution results using pair-wise comparison, and test the correlation between human labeling results and image quality assessment metrics' outputs. Experimental results show that our fidelity-naturalness method is better than the traditional evaluation method for super resolution methods, which could help future research on single-image super resolution.
A large number of image super resolution algorithms based on the sparse coding are proposed, and some algorithms realize the multi-frame super resolution. In multi-frame super resolution based on the sparse coding, both accurate image registration and sparse coding are required. Previous study on multi-frame super resolution based on sparse coding firstly apply block matching for image registration, followed by sparse coding to enhance the image resolution. In this paper, these two problems are solved by optimizing a single objective function. The results of numerical experiments support the effectiveness of the proposed approch.
Restricted Boltzmann Machine (RBM) is an importan- t generative model modeling vectorial data. While applying an RBM in practice to images, the data have to be vec- torized. This results in high-dimensional data and valu- able spatial information has got lost in vectorization. In this paper, a Matrix-Variate Restricted Boltzmann Machine (MVRBM) model is proposed by generalizing the classic RBM to explicitly model matrix data. In the new RBM model, both input and hidden variables are in matrix forms which are connected by bilinear transforms. The MVRBM has much less model parameters, resulting in a faster train- ing algorithm while retaining comparable performance as the classic RBM. The advantages of the MVRBM have been demonstrated on two real-world applications: Image super- resolution and handwritten digit recognition.
In this paper, we propose a new super resolution technique based on the interpolation followed by registering them using iterative back projection (IBP). Low resolution images are being interpolated and then the interpolated images are being registered in order to generate a sharper high resolution image. The proposed technique has been tested on Lena, Elaine, Pepper, and Baboon. The quantitative peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) results as well as the visual results show the superiority of the proposed technique over the conventional and state-of-art image super resolution techniques. For Lena's image, the PSNR is 6.52 dB higher than the bicubic interpolation.
Local learning of sparse image models has proven to be very effective to solve inverse problems in many computer vision applications. To learn such models, the data samples are often clustered using the K-means algorithm with the Euclidean distance as a dissimilarity metric. However, the Euclidean distance may not always be a good dissimilarity measure for comparing data samples lying on a manifold. In this paper, we propose two algorithms for determining a local subset of training samples from which a good local model can be computed for reconstructing a given input test sample, where we take into account the underlying geometry of the data. The first algorithm, called Adaptive Geometry-driven Nearest Neighbor search (AGNN), is an adaptive scheme which can be seen as an out-of-sample extension of the replicator graph clustering method for local model learning. The second method, called Geometry-driven Overlapping Clusters (GOC), is a less complex nonadaptive alternative for training subset selection. The proposed AGNN and GOC methods are evaluated in image super-resolution, deblurring and denoising applications and shown to outperform spectral clustering, soft clustering, and geodesic distance based subset selection in most settings.
In this paper, we propose a novel approach to hyperspectral image super-resolution by modeling the global spatial-and-spectral correlation and local smoothness properties over hyperspectral images. Specifically, we utilize the tensor nuclear norm and tensor folded-concave penalty functions to describe the global spatial-and-spectral correlation hidden in hyperspectral images, and 3D total variation (TV) to characterize the local spatial-and-spectral smoothness across all hyperspectral bands. Then, we develop an efficient algorithm for solving the resulting optimization problem by combing the local linear approximation (LLA) strategy and alternative direction method of multipliers (ADMM). Experimental results on one hyperspectral image dataset illustrate the merits of the proposed approach.
Despite the great advances made in the field of image super-resolution (ISR) during the last years, the performance has merely been evaluated perceptually. Thus, it is still unclear whether ISR is helpful for other vision tasks. In this paper, we present the first comprehensive study and analysis of the usefulness of ISR for other vision applications. In particular, six ISR methods are evaluated on four popular vision tasks, namely edge detection, semantic image segmentation, digit recognition, and scene recognition. We show that applying ISR to input images of other vision systems does improve their performance when the input images are of low-resolution. We also study the correlation between four standard perceptual evaluation criteria (namely PSNR, SSIM, IFC, and NQM) and the usefulness of ISR to the vision tasks. Experiments show that they correlate well with each other in general, but perceptual criteria are still not accurate enough to be used as full proxies for the usefulness. We hope this work will inspire the community to evaluate ISR methods also in real vision applications, and to adopt ISR as a pre-processing step of other vision tasks if the resolution of their input images is low.
Recent work in image processing suggests that operating on (overlapping) patches in an image may lead to state-of-the-art results. This has been demonstrated for a variety of problems including denoising, inpainting, deblurring, and super-resolution. The work reported in [1,2] takes an extra step forward by showing that ordering these patches to form an approximate shortest path can be leveraged for better processing. The core idea is to apply a simple filter on the resulting 1D smoothed signal obtained after the patch-permutation. This idea has been also explored in combination with a wavelet pyramid, leading eventually to a sophisticated and highly effective regularizer for inverse problems in imaging. In this work we further study the patch-permutation concept, and harness it to propose a new simple yet effective regularization for image restoration problems. Our approach builds on the classic Maximum A'posteriori probability (MAP), with a penalty function consisting of a regular log-likelihood term and a novel permutation-based regularization term. Using a plain 1D Laplacian, the proposed regularization forces robust smoothness (L1) on the permuted pixels. Since the permutation originates from patch-ordering, we propose to accumulate the smoothness terms over all the patches' pixels. Furthermore, we take into account the found distances between adjacent patches in the ordering, by weighting the Laplacian outcome. We demonstrate the proposed scheme on a diverse set of problems: (i) severe Poisson image denoising, (ii) Gaussian image denoising, (iii) image deblurring, and (iv) single image super-resolution. In all these cases, we use recent methods that handle these problems as initialization to our scheme. This is followed by an L-BFGS optimization of the above-described penalty function, leading to state-of-the-art results, and especially so for highly ill-posed cases.
Inverse problems in image and audio, and super-resolution in particular, can be seen as high-dimensional structured prediction problems, where the goal is to characterize the conditional distribution of a high-resolution output given its low-resolution corrupted observation. When the scaling ratio is small, point estimates achieve impressive performance, but soon they suffer from the regression-to-the-mean problem, result of their inability to capture the multi-modality of this conditional distribution. Modeling high-dimensional image and audio distributions is a hard task, requiring both the ability to model complex geometrical structures and textured regions. In this paper, we propose to use as conditional model a Gibbs distribution, where its sufficient statistics are given by deep convolutional neural networks. The features computed by the network are stable to local deformation, and have reduced variance when the input is a stationary texture. These properties imply that the resulting sufficient statistics minimize the uncertainty of the target signals given the degraded observations, while being highly informative. The filters of the CNN are initialized by multiscale complex wavelets, and then we propose an algorithm to fine-tune them by estimating the gradient of the conditional log-likelihood, which bears some similarities with Generative Adversarial Networks. We evaluate experimentally the proposed approach in the image super-resolution task, but the approach is general and could be used in other challenging ill-posed problems such as audio bandwidth extension.
In this paper we aim to tackle the problem of reconstructing a high-resolution image from a single low-resolution input image, known as single image super-resolution. In the literature, sparse representation has been used to address this problem, where it is assumed that both low-resolution and high-resolution images share the same sparse representation over a pair of coupled jointly trained dictionaries. This assumption enables us to use the compressed sensing theory to find the jointly sparse representation via the low-resolution image and then use it to recover the high-resolution image. However, sparse representation of a signal over a known dictionary is an ill-posed, combinatorial optimization problem. Here we propose an algorithm that adopts the smoothed $l_0$-norm (SL0) approach to find the jointly sparse representation. Improved quality of the reconstructed image is obtained for most images in terms of both peak signal-to-noise-ratio (PSNR) and structural similarity (SSIM) measures.
We consider image transformation problems, where an input image is transformed into an output image. Recent methods for such problems typically train feed-forward convolutional neural networks using a \emph{per-pixel} loss between the output and ground-truth images. Parallel work has shown that high-quality images can be generated by defining and optimizing \emph{perceptual} loss functions based on high-level features extracted from pretrained networks. We combine the benefits of both approaches, and propose the use of perceptual loss functions for training feed-forward networks for image transformation tasks. We show results on image style transfer, where a feed-forward network is trained to solve the optimization problem proposed by Gatys et al in real-time. Compared to the optimization-based method, our network gives similar qualitative results but is three orders of magnitude faster. We also experiment with single-image super-resolution, where replacing a per-pixel loss with a perceptual loss gives visually pleasing results.
We introduce a method to recover a continuous domain representation of a piecewise constant two-dimensional image from few low-pass Fourier samples. Assuming the edge set of the image is localized to the zero set of a trigonometric polynomial, we show the Fourier coefficients of the partial derivatives of the image satisfy a linear annihilation relation. We present necessary and sufficient conditions for unique recovery of the image from finite low-pass Fourier samples using the annihilation relation. We also propose a practical two-stage recovery algorithm which is robust to model-mismatch and noise. In the first stage we estimate a continuous domain representation of the edge set of the image. In the second stage we perform an extrapolation in Fourier domain by a least squares two-dimensional linear prediction, which recovers the exact Fourier coefficients of the underlying image. We demonstrate our algorithm on the super-resolution recovery of MRI phantoms and real MRI data from low-pass Fourier samples, which shows benefits over standard approaches for single-image super-resolution MRI.