Exemplar-based Algorithm Research Articles

Fast exemplar-based fringe extrapolation technology

When using the Fourier transform method to demodulate the phase from an interferogram, the Gibbs effect often occurs at the edge of the phase picture. Fringe extrapolation is a straightforward method to suppress the Gibbs effect. Currently, many algorithms are used in fringe extrapolation, and exemplar-based methods are one such alternative. However, the traditional exemplar-based algorithm is based on the Criminisi algorithm, which is extremely time-consuming. In this study, the structure-guided image-completion method was applied to interferogram extrapolation. Simulations and experiments show that the new method reduces computer memory requirements and saves computing time. Compared with extrapolation using the Criminisi algorithm, the speed can be increased by tens or even hundreds of times.

Improved image inpainting exemplar-based algorithms by boundary priori-knowledge

Image inpainting plays an important role in restoration of cultural relics, pictures beautification. Criminisi algorithm creates good results in large-area inpainting. However, it does still have some deficiencies such as over-extending. In this paper, two improved algorithms based on prior knowledge of the boundary had been proposed by simulating the idea of manual repairing. An algorithm, by simulating the strategy that the next inpainted pixel will be near to the prior one, named nearer neighbor first algorithm, can void the random bounding of the to-be-inpainted pixle. Another algorithm, by simulating the strategy that the inpainting process, named no-inpainted first algorithm, will be in multiple directions, can void the inpainting process in a single direction. The results reveal that the neighborhood-first algorithm performs better than Criminsi algorithm in repairing the missing structure while the unrepaired-first algorithm performs better than Criminsi algorithm in repairing the missing texture.

The digital reconstruction of degraded ancient temple murals using dynamic mask generation and an extended exemplar-based region-filling algorithm

A mural is any piece of artwork sculpted or applied directly on a wall, ceiling or other permanent surface. This artwork symbolizes various culture’s, traditions, historical events, spiritual stories, and civilizations of respective societies of ancient times. But these mural paintings are subjected to degradation either by various natural causes as well as pollution or by human beings without knowing their value. Restoring these paintings requires skilled artisans who are hard to find these days. Consequently, an efficient image restoration technique is required to meet the particular needs of the paintings. Existing in-painting algorithms largely use pixel-based textural reconstruction. The technique, however, does not work well for images with large, degraded portions. and also fails in the restoration of the structure. To resolve these drawbacks, we propose a combined technique for the textural and structural reconstruction of ancient murals. The proposed Extended Exemplar-based Region-Filling Algorithm uses a patch-based reconstruction procedure and masked images are created automatically using the Dynamic Mask Generation Algorithm. The deteriorated portions are identified by creating masks, and masks are created in such a way that degraded portions have a pixel intensity value of one and the remaining part has a value of zero, and filling is done by analyzing the surrounding pixel values of the degraded pixel. The algorithm reconstructs the structure of the paintings efficiently by generating sketches. The proposed technique reconstructs both the structure and textural information, and ensures efficient reconstructed results, compared to existing in-painting techniques. Performance is evaluated by metrics such as the Mean Square Error (MSE), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Index (SSIM).

An objective method to evaluate exemplar‐based inpainted images quality using Jaccard index

Objective evaluation of images is one of the most essential and practical aspects of image inpainting. The existing objective evaluation methods of image inpainting are functional only on an individual basis and do not provide an accurate and useful objective evaluation of inpainted images. Currently, there is no objective measure for evaluating inpainted images. In this study, an objective evaluation method was developed for image inpainting. In the proposed method, first, 100 images were inpainted using an exemplar-based algorithm. Then, the saliency map and its complementary region in the original image were obtained and a new objective measure was proposed for the evaluation of inpainted images based on the saliency map features. To make the assessment more realistic and comparable to human judgments, two terms, namely penalty and compensation, were taken into account. To assess the performance of our proposed objective measure, the inpainted images were also evaluated using a subjective test. The experiments demonstrates that the proposed objective measure correlated with the qualitative opinion in a human observer study. Finally, the objective measure was compared against three other measures, and the results showed that our proposed objective measure performed better than the other evaluation measures.

An improved image quality algorithm for exemplar-based image inpainting

Image inpainting is a common technique for repairing image regions that are scratched or damaged. This process involves reconstructing damaged parts and filling-in regions in which data/colour information is missing. There are many potential applications for image inpainting, such as repairing old images, repairing scratched images, removing unwanted objects, and filling-in missing areas. This paper develops an exemplar-based algorithm, one of the most important and popular image inpainting techniques, to fill-in missing regions caused by removing unwanted objects, image compression, scratches, or image transformation via the Internet. The proposed algorithm includes two phases of searching to select the best-matching information. In the first phase, the searching mechanism uses the entire image to find and select the most similar patches using the Euclidean distance. The second phase measures the distance between the location of the selected patches and the location of the patch to be filled. The performance of the proposed approach is evaluated through comprehensive experiments on several well-known images used in this area of research. The experimental results demonstrate the superior performance of the proposed approach over some state-of-the-art approaches in terms of quality in terms of both objective (using the peak signal-to-noise ratio (PSNR) as well as the structural similarity index method (SSIM)) and subjective (i.e., visual) measures.

Remote sensing data retouching based on image inpainting algorithms in the forgery generation problem

We investigate image retouching algorithms for generating forgery Earth remote sensing data. We provide an overview of existing neural network solutions in the field of generation and inpainting of remote sensing images. To retouch Earth remote sensing data, we use image-inpainting algorithms based on convolutional neural networks and generative-adversarial neural networks. We pay special attention to a generative neural network with a separate contour prediction block that includes two series-connected generative-adversarial subnets. The first subnet inpaints contours of the image within the retouched area. The second subnet uses the inpainted contours to generate the resulting retouch area. As a basis for comparison, we use exemplar-based algorithms of image inpainting. We carry out computational experiments to study the effectiveness of these algorithms when retouching natural data of remote sensing of various types. We perform a comparative analysis of the quality of the algorithms considered, depending on the type, shape and size of the retouched objects and areas. We give qualitative and quantitative characteristics of the efficiency of the studied image inpainting algorithms when retouching Earth remote sensing data. We experimentally prove the advantage of generative-competitive neural networks in the construction of forgery remote sensing data.

Use of Generative Adversarial Networks to Altering Remote Sensing Data

The paper investigates the use of generative adversarial networks (GAN) for intentional modification of Earth remote sensing data. A generative neural network that includes a special subnet for object boundary inpainting is considered. The network comprises two GAN: the first completes the object boundary, and the second repaints blank areas. Actual remote sensing data are used to test the generative network under consideration. The exemplar-based Patch-Match algorithm is taken as a reference for comparison purposes. The experimental results allow the conclusion that the approach is an effective tool for the intentional modification of large terrestrial area images in falsification of Earth remote sensing data.

3D human pose estimation from a single image via exemplar augmentation

3D human pose estimation from a single image is a challenging problem due to occlusion, viewpoint variance, and the ill-posed nature of back projection. We follow a standard two-step pipeline which first detects 2D joint locations and uses them to infer 3D pose. For the first step, we use a recent deep learning-based detector. For the second step, we propose a novel exemplar-based algorithm to implicitly augment the exemplar set for 3D human pose estimation. The motivation of this algorithm is to well represent various poses in the real world with finite real exemplars. We achieve it by a strategy of synthesizing virtual candidate poses which ensures that the augmented exemplar set has much more variety. Moreover, we also present an effective approach to select the best exemplar from candidate set to well match the detected 2D pose. Experimental results show that our method achieves competitive performance on Human3.6M dataset.

Concept detection using text exemplars aligned with a specialized ontology

Knowledge extraction from text documents requires identifying and classifying semantic content. Utilizing an appropriate domain ontology can facilitate this process if words and phrases can be linked to the classes and relationships within the ontology. This paper presents an exemplar-based algorithm to link text to semantically similar classes within an ontology constructed for the chronic pain medicine domain. Human annotators linked classes to text segments within a random document set for construction of an exemplar dictionary, which we examined for completeness using Zipf plot analysis. An algorithm was created to use this dictionary on previously unseen text to form a map between sentence text and probable class assignments. We performed a 5×5 cross-validation between human and algorithm annotations and examined both ROC and precision versus recall curves to show that the algorithm can identify the many medical and biopsychosocial components from the texts. We briefly describe a use case for detecting pain relief from various interventions utilizing the word-by-class maps. We conclude that an exemplar-based method can be a valuable tool in knowledge extraction from texts that share similar construction, such as medical progress notes.

Light Field Coding With Field-of-View Scalability and Exemplar-Based Interlayer Prediction

Light field imaging based on microlens arrays-a.k.a. holoscopic, plenoptic, and integral imaging-has currently risen up as a feasible and prospective technology for future image and video applications. However, deploying actual light field applications will require identifying more powerful representations and coding solutions that support arising new manipulation and interaction functionalities. In this context, this paper proposes a novel scalable coding solution that supports a new type of scalability, referred to as field-of-view scalability. The proposed scalable coding solution comprises a base layer compliant with the High Efficiency Video Coding (HEVC) standard, complemented by one or more enhancement layers that progressively allow richer versions of the same light field content in terms of content manipulation and interaction possibilities. In addition, to achieve high-compression performance in the enhancement layers, novel exemplar-based interlayer coding tools are also proposed, namely: 1) a direct prediction based on exemplar texture samples from lower layers and 2) an interlayer compensated prediction using a reference picture that is built relying on an exemplar-based algorithm for texture synthesis. Experimental results demonstrate the advantages of the proposed scalable coding solution to cater to users with different preferences/requirements in terms of interaction functionalities, while providing better rate-distortion performance (independently of the optical setup used for acquisition) compared to HEVC and other scalable light field coding solutions in the literature.

Image completion using multispectral imaging

Here, the authors explore the potential of multispectral imaging applied to image completion. Snapshot multispectral cameras correspond to breakthrough technologies that are suitable for everyday use. Therefore, they correspond to an interesting alternative to digital cameras. In their experiments, multispectral images are acquired using an ultracompact snapshot camera-recorder that senses 16 different spectral channels in the visible spectrum. Direct exploitation of completion algorithms by extension of the spectral channels exhibits only minimum enhancement. A dedicated method that consists in a prior segmentation of the scene has been developed to address this issue. The segmentation derives from an analysis of the spectral data and is employed to constrain research area of exemplar-based completion algorithms. The full processing chain takes benefit from standard methods that were developed by both hyperspectral imaging and computer vision communities. Results indicate that image completion constrained by spectral presegmentation ensures better consideration of the surrounding materials and simultaneously improves rendering consistency, in particular for completion of flat regions that present no clear gradients and little structure variance. The authors validate their method with a perceptual evaluation based on 20 volunteers. This study shows for the first time the potential of multispectral imaging applied to image completion.

Compound risk judgment in tasks with both idiosyncratic and systematic risk: The “Robust Beauty” of additive probability integration

In this study, we explore how people integrate risks of assets in a simulated financial market into a judgment of the conjunctive risk that all assets decrease in value, both when assets are independent and when there is a systematic risk present affecting all assets. Simulations indicate that while mental calculation according to naïve application of probability theory is best when the assets are independent, additive or exemplar-based algorithms perform better when systematic risk is high. Considering that people tend to intuitively approach compound probability tasks using additive heuristics, we expected the participants to find it easiest to master tasks with high systematic risk – the most complex tasks from the standpoint of probability theory – while they should shift to probability theory or exemplar memory with independence between the assets. The results from 3 experiments confirm that participants shift between strategies depending on the task, starting off with the default of additive integration. In contrast to results in similar multiple cue judgment tasks, there is little evidence for use of exemplar memory. The additive heuristics also appear to be surprisingly context-sensitive, with limited generalization across formally very similar tasks.

Incremental enhanced α-expansion move for large data: a probability regularization perspective

To deal with large data clustering tasks, an incremental version of exemplar-based clustering algorithm is proposed in this paper. The novel clustering algorithm, called Incremental Enhanced α-Expansion Move (IEEM), processes large data chunk by chunk. The work here includes two aspects. First, in terms of the maximum a posteriori principle, a unified target function is developed to unify two typical exemplar-based clustering algorithms, namely Affinity Propagation (AP) and Enhanced α-Expansion Move (EEM). Secondly, with the proposed target function, the probability based regularization term is proposed and accordingly the proposed target function is extended to make IEEM have the ability to improve clustering performance of the entire dataset by leveraging the clustering result of previous chunks. Another outstanding characteristic of IEEM is that only by modifying the definitions of several variables used in EEM, the minimization procedure of EEM and its theoretical spirit can be easily kept in IEEM, and hence no more efforts are needed to develop a new optimization algorithm for IEEM. In contrast to AP, EEM and the existing incremental clustering algorithm IMMFC, our experimental results of synthetic and real-world datasets indicate the effectiveness of IEEM.

A Novel Exemplar-Based Non-Local Algorithm for Image Inpainting

A Novel Exemplar-Based Non-Local Algorithm for Image Inpainting

Exemplar-Based Image Inpainting Using a Modified Priority Definition.

Exemplar-based algorithms are a popular technique for image inpainting. They mainly have two important phases: deciding the filling-in order and selecting good exemplars. Traditional exemplar-based algorithms are to search suitable patches from source regions to fill in the missing parts, but they have to face a problem: improper selection of exemplars. To improve the problem, we introduce an independent strategy through investigating the process of patches propagation in this paper. We first define a new separated priority definition to propagate geometry and then synthesize image textures, aiming to well recover image geometry and textures. In addition, an automatic algorithm is designed to estimate steps for the new separated priority definition. Comparing with some competitive approaches, the new priority definition can recover image geometry and textures well.

Object removal by depth-wise image inpainting

The problem posed in this paper was to fill-up the hole in the digital image left behind after the removal of an object from it. The real challenge is to deal with images of cluttered scenes of objects with complex and infrequent textures. In the past, this problem has been addressed by two classes of algorithms: (i) exemplar-based algorithms to fill the hole using the patches available in a given example and (ii) depth-based algorithms to differentiate the foreground and background before inpainting. This paper presents a novel method that combines advantages of these two approaches. We use depth map of an image to find the order of all the objects in the target image and a database of multi-views of objects to fill the holes. To reduce the size of the database, we propose a method called Keyview Extraction. An object retrieval followed by geometric and photometric registration algorithms is employed to make every object an exact match for inpainting. A number of experiments on both real and synthetic images demonstrate the advantage of depth-wise image inpainting compared with other inpainting methods. In these experiments, we compare three different types of image inpainting methods with our method quantitatively by computing their SSIM values using a ground-truth.

Image Inpainting Using Patch Sparsity

In this paper we discuss a modified exemplar-based inpainting method through investigating the sparsity of natural image patches. In the exemplar-based algorithms, with the help of available information the unknown blocks of target region are inpainted by the most similar blocks extracted from the source region. To decide the filling order of missing pixels ensures the connectivity of object boundaries. The priority term shold defined. In the exemplar-based patch sparsity approaches, a sparse representation of missing pixels was considered to define a new priority term. Here, we discuss modification of the priority term and take measures to compute the similarities between fill-front and candidate patches. Image inpainting by patch propogation using patch sparsity shows the effeteness over traditional exemplar based inpainting.

Clustering based on enhanced α-expansion move

The exemplar-based data clustering problem can be formulated as minimizing an energy function defined on a Markov random field (MRF). However, most algorithms for optimizing the MRF energy function cannot be directly applied to the task of clustering, as the problem has a high-order energy function. In this paper, we first show that the high-order energy function for the clustering problem can be simplified as a pairwise energy function with the metric property, and consequently it can be optimized by the α-expansion move algorithm based on graph cut. Then, the original expansion move algorithm is improved in the following two aspects: 1) Instead of solving a minimal s-t graph cut problem, we show that there is an explicit and interpretable solution for minimizing the energy function in the clustering problem. Based on this interpretation, a fast α-expansion move algorithm is proposed, which is much more efficient than the graph-cut-based algorithm. 2) The fast α-expansion move algorithm is further improved by extending its move space so that a larger energy value reduction can be achieved in each iteration. Experiments on benchmark data sets show that the enhanced expansion move algorithm has a better performance, compared to other state-of-the-art exemplar-based clustering algorithms.

Lexical Disambiguation of Arabic Language: An Experimental Study

In this paper we test some supervised algorithms that most of the existing related works of word sen se disambiguation have cited. Due to the lack of lingu istic data for the Arabic language, we work on non-annotated corpus and with the help of four annotators; we were able to annota te the different samples containing the ambiguous words. Since that, we test the Naive Bayes algorithm, the decision lists and the exemplar based algorithm. During the experimental study, we test the influence of the window size on the disamb iguation quality, the derivation and the technique of smooth ing for the (2n+1)-grams. For these tests the exemplar based al gorithm achieves the best rate of precision. Index Terms —Supervised algorithms, training data, Naive Bayes, decision list, exemplar based algorithm, win dow size.

Towards an exemplar-based model of stress in English noun–noun compounds

It is well known that stress assignment in English noun–noun compounds is non-uniform (compare e.g. left-prominent ópera glasses and right-prominent steel brídge), and recent corpus-based studies (e.g. Plag et al. 2007, 2008) have shown that categorical, rule-based approaches that make use of argument structure (e.g. Giegerich 2004) or semantics (e.g. Fudge 1984) are not able to account satisfactorily for the existing variability. Using data from the corpus studies by Plag and collegues, I argue in this paper that an exemplar-based approach is better-suited to accounting for stress assignment in English noun–noun compounds than a traditional, rule-based paradigm. Specifically, it is shown that two current implementations of exemplar-based algorithms, TiMBL (Daelemans et al. 2007) and AM::Parallel (Skousen & Stanford 2007), clearly outperform comparable rule models in terms of how well they predict stress assignment in the corpora. Furthermore, systematic testing reveals that the reasons for the differences between exemplar and rule models mainly lie in their ability to incorporate detailed, non-abstract information (specifically, constituent family information). The present study therefore adds to the growing evidence in favour of the importance of constituent family information in compounding (e.g. Gagné 2001, Krott, Schreuder & Baayen 2002).