Image Matting: A Comprehensive Survey on Techniques, Comparative Analysis, Applications and Future Scope

In the era of rapid growth of technologies, image matting plays a key role in image and video editing along with image composition. In many significant real-world applications such as film production, it has been widely used for visual effects, virtual zoom, image translation, image editing and video editing. With recent advancements in digital cameras, both professionals and consumers have become increasingly involved in matting techniques to facilitate image editing activities. Image matting plays an important role to estimate alpha matte in the unknown region to distinguish foreground from the background region of an image using an input image and the corresponding trimap of an image which represents a foreground and unknown region. Numerous image matting techniques have been proposed recently to extract high-quality matte from image and video sequences. This paper illustrates a systematic overview of the current image and video matting techniques mostly emphasis on the current and advanced algorithms proposed recently. In general, image matting techniques have been categorized according to their underlying approaches, namely, sampling-based, propagation-based, combination of sampling and propagation-based and deep learning-based algorithms. The traditional image matting algorithms depend primarily on color information to predict alpha matte such as sampling-based, propagation-based or combination of sampling and propagation-based algorithms. However, these techniques mostly use low-level features and suffer from high-level background which tends to produce unwanted artifacts when color is same or semi-transparent in the foreground object. Image matting techniques based on deep learning have recently introduced to address the shortcomings of traditional algorithms. Rather than simply depending on the color information, it uses deep learning mechanism to estimate the alpha matte using an input image and the trimap of an image. A comprehensive survey on recent image matting algorithms and in-depth comparative analysis of these algorithms has been thoroughly discussed in this paper.

Regularization of the Factorization Method applied to diffuse optical tomography

In this paper, we develop a new regularized version of the Factorization Method for positive operators mapping a complex Hilbert Space into it’s dual space. The Factorization Method uses Picard’s Criteria to define an indicator function to image an unknown region. In most applications the data operator is compact which gives that the singular values can tend to zero rapidly which can cause numerical instabilities. The regularization of the Factorization Method presented here seeks to avoid the numerical instabilities in applying Picard’s Criteria. This method allows one to image the interior structure of an object with little a priori information in a computationally simple and analytically rigorous way. Here we will focus on an application of this method to diffuse optical tomography where will prove that this method can be used to recover an unknown subregion from the Dirichlet-to-Neumann mapping. Numerical examples will be presented in two dimensions.

Two Amphioxus ApeC-Containing Proteins Bind to Microbes and Inhibit the TRAF6 Pathway

The apextrin C-terminal (ApeC) domain is a class of newly discovered protein domains with an origin dating back to prokaryotes. ApeC-containing proteins (ACPs) have been found in various marine and aquatic invertebrates, but their functions and the underlying mechanisms are largely unknown. Early studies suggested that amphioxus ACP1 and ACP2 bind to bacterial cell walls and have a role in immunity. Here we identified another two amphioxus ACPs (ACP3 and ACP5), which belong to the same phylogenetic clade with ACP1/2, but show distinct expression patterns and sequence divergence (40-50% sequence identities). Both ACP3 and ACP5 were mainly expressed in the intestine and hepatic cecum, and could be up-regulated after bacterial challenge. Both prokaryotic-expressed recombinant ACP3 and ACP5 could bind with several species of bacteria and yeasts, showing agglutinating activity but no microbicidal activity. ELISA assays suggested that their ApeC domains could interact with peptidoglycan (PGN), but not with lipoteichoic acid (LTA), lipopolysaccharides (LPS) and zymosan A. Furthermore, they can only bind to Lys-type PGN from Staphylococcus aureus, but not to DAP-type PGN from Bacillus subtilis and not to moieties of PGN such as MDPs, NAMs and NAGs. This recognition spectrum is different from that of ACP1/2. We also found that when expressed in mammalian cells, ACP3 could interact with TRAF6 via a conserved non-ApeC region, which inhibited the ubiquitination of TRAF6 and hence suppressed downstream NF-κB activation. This work helped define a novel subfamily of ACPs, which have conserved structures, and have related yet diversified molecular functions. Its members have dual roles, with ApeC as a lectin and a conserved unknown region as a signal transduction regulator. These findings expand our understanding of the ACP functions and may guide future research on the role of ACPs in different animal clades.

Trust-based partner switching among partitioned regions promotes cooperation in public goods game

In this paper, the coevolution mechanism of trust-based partner switching among partitioned regions on an adaptive network is studied. We investigate a low-information approach to building trust and cooperation in public goods games. Unlike reputation, trust scores are only given to players by those with whom they have a relationship in the game, depending on the game they play together. A player’s trust score for a certain neighbor is given and known by that player only. Players can adjust their connections to neighbors with low trust scores by switching their partners to other players. When switching partners, players divide other nodes in the network into three regions: immediate neighbors as the known region, indirectly connected second-order neighbors as the intermediate region, and other nodes as the unknown region. Such choices and compartmentalization often occur in global and regional economies. Our results show that preference for switching to partners in the intermediate region is not conducive to spreading cooperation, while random selection has the disadvantage of protecting the cooperator. However, selecting new partners in the remaining two regions based on the average trust score of the known region performs well in both protecting partners and finding potential cooperators. Meanwhile, by analyzing the parameters, we find that the influence of vigilance increasing against unsatisfactory behavior on evolution direction depends on the level of cooperation reward.

A Novel Image Inpainting Framework Using Regression

In this article, a blockwise regression-based image inpainting framework is proposed. The core idea is to fill the unknown region in two stages: Extrapolate the edges to the unknown region and then fill the unknown pixels values in each sub-region demarcated by the extended edges. Canny edge detection and linear edge extension are used to respectively identify and extend edges to the unknown region followed by regression within each sub-region to predict the unknown pixel values. Two different regression models based on K-nearest neighbours and support vectors machine are used to predict the unknown pixel values. The proposed framework has the advantage of inpainting without requiring prior training on any image dataset. The extensive experiments on different images with contrasting distortions demonstrate the robustness of the proposed framework and a detailed comparative analysis shows that the proposed technique outperforms existing state-of-the-art image inpainting methods. Finally, the proposed techniques are applied to MRI images suffering from susceptibility artifacts to illustrate the practical usage of the proposed work.

Adaptive Depth Guided Image Completion for Structure and Texture Synthesis

The problem of synthesis of missing image parts represents an interesting and challenging area of image processing and computer vision with significant potential. This thesis, focuses on an adaptive depth-guided image completion method that addresses the image completion problem using information contained in the rest of the image. The completion process is separated into structure and texture synthesis. A method is first introduced for completing the respective depth map through the use of a diffusion-based operation, preserving global image structure within the unknown region. Building upon the state of the art exemplar based inpainting technique of Barnes et al., we complete the target (unknown) region by matching to and blending source patches drawn from the rest of the image, using the reconstructed depth information to guide the completion process. Secondly, for each target patch, we formulate an adaptive patch size determination as an optimization problem that minimizes an objective function involving local image gradient magnitude and orientations. An extension to the coherence- based objective function introduced by Wexler et al. is then introduced, which not only encourages coherence of the respective target region with respect to the source region in colour but also in depth. We further consider the variance between patches in the SSD criteria for preventing error accumulation and propagation. Experimental results show that our method can provide a significant improvement to patch-based image completion algorithms shown by PSNR and SSIM calculations as well as a qualitative subjective study.

Adaptive Depth Guided Image Completion for Structure and Texture Synthesis

The problem of synthesis of missing image parts represents an interesting and challenging area of image processing and computer vision with significant potential. This thesis, focuses on an adaptive depth-guided image completion method that addresses the image completion problem using information contained in the rest of the image. The completion process is separated into structure and texture synthesis. A method is first introduced for completing the respective depth map through the use of a diffusion-based operation, preserving global image structure within the unknown region. Building upon the state of the art exemplar based inpainting technique of Barnes et al., we complete the target (unknown) region by matching to and blending source patches drawn from the rest of the image, using the reconstructed depth information to guide the completion process. Secondly, for each target patch, we formulate an adaptive patch size determination as an optimization problem that minimizes an objective function involving local image gradient magnitude and orientations. An extension to the coherence- based objective function introduced by Wexler et al. is then introduced, which not only encourages coherence of the respective target region with respect to the source region in colour but also in depth. We further consider the variance between patches in the SSD criteria for preventing error accumulation and propagation. Experimental results show that our method can provide a significant improvement to patch-based image completion algorithms shown by PSNR and SSIM calculations as well as a qualitative subjective study.

Annelids of the eastern Australian abyss collected by the 2017 RV ‘Investigator’ voyage

In Australia, the deep-water (bathyal and abyssal) benthic invertebrate fauna is poorly known in comparison with that of shallow (subtidal and shelf) habitats. Benthic fauna from the deep eastern Australian margin was sampled systematically for the first time during 2017 RV ‘Investigator’ voyage ‘Sampling the Abyss’. Box core, Brenke sledge, and beam trawl samples were collected at one-degree intervals from Tasmania, 42°S, to southern Queensland, 24°S, from 900 to 4800 m depth. Annelids collected were identified by taxonomic experts on individual families around the world. A complete list of all identified species is presented, accompanied with brief morphological diagnoses, taxonomic remarks, and colour images. A total of more than 6000 annelid specimens consisting of 50 families (47 Polychaeta, one Echiura, two Sipuncula) and 214 species were recovered. Twenty-seven species were given valid names, 45 were assigned the qualifier cf., 87 the qualifier sp., and 55 species were considered new to science. Geographical ranges of 16 morphospecies extended along the eastern Australian margin to the Great Australian Bight, South Australia; however, these ranges need to be confirmed with genetic data. This work providing critical baseline biodiversity data on an important group of benthic invertebrates from a virtually unknown region of the world’s ocean will act as a springboard for future taxonomic and biogeographic studies in the area.

3D Shared Matting Method for Directly Extracting Standard Organ Models from Human Body Color Volume Image

Background: In some medical applications (e.g., virtual surgery), standard human organ models are very important and useful. Now that real human body slice image sets have been collected by several countries, it is possible to obtain real standard organ models. Introduction: Understanding how to abandon the traditional model construction method of Photoshop sketching slice by slice and directly extracting 3D models from volume images has been an interesting and challenging issue. In this paper, a 3D color volume image matting method has been proposed to segment human body organ models. Methods: First, the scope of the known area will be expanded by means of propagation. Next, neighborhood sampling to find the best sampling for voxels in an unknown region will be performed and then the preliminary opacity using the sampling results will be calculated. Results: The final result will be obtained by applying local smoothing to the image. Conclusion: From the experimental results, it has been observed that our method is effective for real standard organ model extraction.

AN AUTOMATIC SEMANTIC MAP GENERATION METHOD USING TRAJECTORY DATA

It’s easily to obtain the geometric information of terrain features in a timely manner using advanced surveying and mapping methods, but it is impossible to obtain their semantic information with low latency due to the rapid development of cities. The popularity of GPS-enabled devices and technologies provide us a large number of personal location information. Moreover, it is possible to extract the personal or group behavior pattern due to the regularity of human behavior. Those conditions make it possible to extract and identify human behavior patterns from their trajectory data. In this paper, we present an automatic semantic map generation method that extract semantic patterns and take advantage of them to tagging spatial objects in an unknown region based on known semantic patterns. We study the regularity of trajectory data and build the semantic pattern based on the regularity of human behavior. Most importantly, we use known semantic patterns to identify the semantics of the stay points in the unknown region, and use this method to realize the semantic recognition of the stay points. Results of the experiments show the effectiveness of our proposed method.