AbstractÐNearly all super-resolution algorithms are based on the fundamental constraints that the super-resolution image should generate the low resolution input images when appropriately warped and down-sampled to model the image formation process. �These reconstruction constraints are normally combined with some form of smoothness prior to regularize their solution.) In the first part of this paper, we derive a sequence of analytical results which show that the reconstruction constraints provide less and less useful information as the magnification factor increases. We also validate these results empirically and show that, for large enough magnification factors, any smoothness prior leads to overly smooth results with very little high-frequency content �however, many low resolution input images are used). In the second part of this paper, we propose a super-resolution algorithm that uses a different kind of constraint, in addition to the reconstruction constraints. The algorithm attempts to recognize local features in the low-resolution images and then enhances their resolution in an appropriate manner. We call such a super-resolution algorithm a hallucination or recogstruction algorithm. We tried our hallucination algorithm on two different data sets, frontal images of faces and printed Roman text. We obtained significantly better results than existing reconstruction-based algorithms, both qualitatively and in terms of RMS pixel error. Index TermsÐSuper-resolution, analysis of reconstruction constraints, learning, faces, text, hallucination, recogstruction. 1

Superresolution is a technique that can produce images of a higher resolution than that of the originally captured ones. Nevertheless, improvement in resolution using such a technique is very limited in practice. This makes it significant to study the problem: “Do fundamental limits exist for superresolution?” In this paper, we focus on a major class of superresolution algorithms, called the reconstruction-based algorithms, which compute high-resolution images by simulating the image formation process. Assuming local translation among low-resolution images, this paper is the first attempt to determine the explicit limits of reconstruction-based algorithms, under both real and synthetic conditions. Based on the perturbation theory of linear systems, we obtain the superresolution limits from the conditioning analysis of the coefficient matrix. Moreover, we determine the number of low-resolution images that are sufficient to achieve the limit. Both real and synthetic experiments are carried out to verify our analysis.

Recent years have seen growing interest in the problem of super-resolution restoration of video sequences. Whereas in the traditional single image restoration problem only a single input image is available for processing, the task of reconstructing super-resolution images from multiple undersampled and degraded images can take advantage of the additional spatiotemporal data available in the image sequence. In particular, camera and scene motion lead to frames in the source video sequence containing similar, but not identical information. The additional information available in these frames make possible reconstruction of visually superior frames at higher resolution than that of the original data. In this paper we review the current state of the art and identify promising directions for future research.

In most surveillance scenarios there is a large distance between the camera and the objects of interest in the scene. Surveillance cameras are also usually set up with wide #elds of view in order to image as much of the scene as possible. The end result is that the objects in the scene normally appear very small in surveillance imagery. It is generally possible to detect and track the objects in the scene, however, for tasks such as automatic face recognition and license plate reading, resolution enhancement techniques are often needed. Although numerous

Superresolution reconstruction produces a high-resolution image from a set of low-resolution images. Previous iterative methods for superresolution had not adequately addressed the computational and numerical issues for this ill-conditioned and typically underdetermined large scale problem. We propose efficient block circulant preconditioners for solving the Tikhonov-regularized superresolution problem by the conjugate gradient method. We also extend to underdetermined systems the derivation of the generalized cross-validation method for automatic calculation of regularization parameters. Effectiveness of our preconditioners and regularization techniques is demonstrated with superresolution results for a simulated sequence and a forward looking infrared (FLIR) camera image sequence.

Growing interest in super-resolution (SR) restoration of video sequences and the closely related problem of construction of SR still images from image sequences has led to the emergence of several competing methodologies. We review the state of the art of SR techniques using a taxonomy of existing techniques. We critique these methods and identify areas which promise performance improvements. 1. Introduction The problem of spatial resolution enhancement of video sequences has been an area of active research since the seminal work by Tsai and Huang [20] which considers the problem of resolution enhanced stills from a sequence of lowresolution (LR) images of a translated scene. Whereas in the traditional single image restoration problem only a single input image is available, the task of obtaining a superresolved image from an undersampled and degraded image sequence can take advantage of the additional spatiotemporal data available in the image sequence. In particular, camera and scen...

Abstract—In many image restoration/resolution enhancement applications, the blurring process, i.e., point spread function (PSF) of the imaging system, is not known or is known only to within a set of parameters. We estimate these PSF parameters for this ill-posed class of inverse problem from raw data, along with the regularization parameters required to stabilize the solution, using the generalized cross-validation method (GCV). We propose efficient approximation techniques based on the Lanczos algorithm and Gauss quadrature theory, reducing the computational complexity of the GCV. Data-driven PSF and regularization parameter estimation experiments with synthetic and real image sequences are presented to demonstrate the effectiveness and robustness of our method. Index Terms—Blind restoration, blur identification, generalized cross-validation, quadrature rules, superresolution. I.

Recently, there has been much work developing super-resolution algorithms for combining a set of low quality images to produce a set of higher quality images. In most cases, such algorithms must first register the collection of images to a common sampling grid and then reconstruct the high resolution image. While many such algorithms have been proposed to address each one of these subproblems, no work has addressed the overall performance limits for this joint estimation problem. In this paper, we analyze the performance limits from statistical first principles using the Cramer-Rao bound. We offer insight into the fundamental bottlenecks limiting the performance of multiframe image reconstruction algorithms and hence super-resolution.

The extraction of a single high-quality image from a set of lowresolution images is an important problem which arises in elds such as remote sensing, surveillance, medical imaging and the extraction of still images from video. Typical approaches are based on the use of cross-correlation to register the images followed by the inversion of the transformation from the unknown high resolution image to the observed low resolution images, using regularization to resolve the ill-posed nature of the inversion process. In this paper we develop a Bayesian treatment of the super-resolution problem in which the likelihood function for the image registration parameters is based on a marginalization over the unknown high-resolution image. This approach allows us to estimate the unknown point spread function, and is rendered tractable through the introduction of a Gaussian process prior over images. Results indicate a signi cant improvement over techniques based on MAP (maximum a-posteriori) point optimization of the high resolution image and associated registration parameters.

This paper addresses the problem of generating a superresolution (SR) image from a single low-resolution input image. We approach this problem from the perspective of compressed sensing. The low-resolution image is viewed as downsampled version of a high-resolution image, whose patches are assumed to have a sparse representation with respect to an over-complete dictionary of prototype signalatoms. The principle of compressed sensing ensures that under mild conditions, the sparse representation can be correctly recovered from the downsampled signal. We will demonstrate the effectiveness of sparsity as a prior for regularizing the otherwise ill-posed super-resolution problem. We further show that a small set of randomly chosen raw patches from training images of similar statistical nature to the input image generally serve as a good dictionary, in the sense that the computed representation is sparse and the recovered high-resolution image is competitive or even superior in quality to images produced by other SR methods. 1.