The World Ocean Assessment is a comprehensive investigation of oceanography. The assessment contains over 3500 bibliographic references, however due to the size of the assessment it can be difficult for a human to effectively use this as a starting point for literature reviews. Therefore, we implemented a system that visualizes the relationships between the assessment’s chapters, keywords, and bibliographic metadata. Filters and a tagging system were also implemented to make finding papers easier. Finally, we demonstrate the system’s features through a case study.

Given the expanding use of 3D Objects in a variety of fields such as animation, gaming, virtual worlds, commerce, augmented reality and 3D printing, we present a novel system for object browsing and searching. Specifically, the system packs objects into an interactive 3D cloud for browsing and searching. It was designed with the aim of increasing search efficiency in a variety of active environments, while providing a visually engaging layout, and we evaluated this by conducting a comparative user study. We show that our system can significantly decrease search times compared to the classic grid-based layout, and it has been suggested by subjects that cloud-based searching is more interesting and visually-engaging.

to create an integrated space. We consider an MR configuration in which collocated collaborators work around a tabletop display, while remote collaborators wear an HMD to interact with a connected virtual environment that gives a 3D perspective, and consider the impact of varying degrees of view congruence with their collaborators. In a within-subjects study with 18 groups of 3, groups completed task scenarios involving 3D object manipulation around a physical-virtual mapped tabletop. We compare a synchronized Tabletop display baseline and two MR conditions with different levels of view congruence: Fishtank and Hover. Fishtank has a high degree of congruence as it shares a top-down perspective of the 3D objects with the tabletop collaborators. The Hover condition has less view congruence since 3D content hovers front of the remote collaborator above the table. The MR conditions yielded higher self-reported awareness and co-presence than the Tabletop condition for both collocated and remote participants. Remote collaborators significantly preferred the MR conditions for manipulating shared 3D models and communicating with their collaborators. Our findings illustrate strengths and weaknesses of both MR techniques but show that more participants preferred the less-congruent Hover condition overall. Reasons include that it facilitated interaction and viewing 3D objects.

The worldwide adoption of Twitter turned it into one of the most popular platforms for content analysis as it serves as a gauge of the public’s feeling and opinion on a variety of topics. This is particularly true of political discussions and lawmakers’ actions and initiatives. Yet, one common but unrealistic assumption is that the data of interest for analysis is readily available in a comprehensive and accurate form. Data need to be retrieved, but due to the brevity and noisy nature of Twitter content, it is difficult to formulate user queries that match relevant posts that use different terminology without introducing a considerable volume of unwanted content. This problem is aggravated when the analysis must contemplate multiple and related topics of interest, for which comments are being concurrently posted. This article presents Active Tweet Retrieval Visualization (ATR-Vis), a user-driven visual approach for the retrieval of Twitter content applicable to this scenario. The method proposes a set of active retrieval strategies to involve an analyst in such a way that a major improvement in retrieval coverage and precision is attained with minimal user effort. ATR-Vis enables non-technical users to benefit from the aforementioned active learning strategies by providing visual aids to facilitate the requested supervision. This supports the exploration of the space of potentially relevant tweets, and affords a better understanding of the retrieval results. We evaluate our approach in scenarios in which the task is to retrieve tweets related to multiple parliamentary debates within a specific time span. We collected two Twitter datasets, one associated with debates in the Canadian House of Commons during a particular week in May 2014, and another associated with debates in the Brazilian Federal Senate during a selected week in May 2015. The two use cases illustrate the effectiveness of ATR-Vis for the retrieval of relevant tweets, while quantitative results show that our approach achieves high retrieval quality with a modest amount of supervision. Finally, we evaluated our tool with three external users who perform searching in social media as part of their professional work.

With the development of high-dynamic-range images and tone mapping operators comes a need for image quality evaluation of tone mapped images. However, because of the significant difference in dynamic range between high-dynamic-range images and tone mapped images, conventional image quality assessment algorithms that predict distortion based on the magnitude of intensity or normalized contrast are not suitable for this task. In this article, we present a feature-based quality metric for tone mapped images that predicts the perceived quality by measuring the distortion in important image features that affect quality judgment. Our metric utilizes multi-exposed virtual photographs taken from the original high-dynamic-range images to bridge the gap between dynamic ranges in image feature analysis. By combining measures for brightness distortion, visual saliency distortion, and detail distortion in light and dark areas, the metric measures the overall perceptual distortion and assigns a score to a tone mapped image. Experiments on a subject-rated database indicate that the proposed metric is more consistent with subjective evaluation results than alternative approaches.

Twitter has become one of the most important platforms for gathering information, where users follow breaking news, track ongoing events and learn about their topics of interest. Considering the sheer volume of Twitter data and the ever-growing number of users, it is of great importance to have real-time systems that can monitor and recommend relevant and non-redundant tweets with respect to users' interests. In this paper, we propose a framework using language models as a basis for analyzing strategies and techniques for tweet recommendation based on user interest profiles. Results show that identifying named entities in profiles has a major impact on the accuracy of the recommender. We also performed a thorough comparison to investigate whether state-of-the-art semantic relatedness techniques have a positive impact on the precision of the recommended tweets. The TREC 2015 Microblog track dataset is used for comparison and evaluation throughout this paper.

Tone mapping is the process of transforming high dynamic range images for display on low dynamic range devices. While many tone mapping operators have been proposed, there is no single operator that generates optimal results under all conditions. Blending the results from multiple operators with varying weights allows for leveraging the strengths of each of the operators considered. Prior work has used interactive evolution as a tool for blended tone mapping. In this paper, we build on recent progress in the development of objective quality measures for tone mapped images that allows us to automate the process of evolving blended tone mapped images. The quality measure used assesses tone mapped images in terms of brightness, visual saliency, and detail reproduction in bright and dark regions and assigns an overall score to each image. The blended tone mapping problem can then be solved as an optimization problem, where the operators' parameters and the weights that determine the relative influence of each operator are tuned to generate images with optimal perceptual quality. We show that the optimization can be accomplished with an evolutionary algorithm. Experiments with high dynamic range images demonstrate the effectiveness of our approach.

We present a study exploring privacy behaviours in mixed reality collaborative environments. We consider two scenarios involving hiding and sharing blended physical-virtual documents around a tabletop, under two vertical display conditions: a solo display showing the remote collaborator as an avatar, and circumambient displays showing the rest of the connected virtual environment. We explore four types of cues (interactive, communication-based, ambient, and infrastructure) and their impact on how collaborators hide and share blended physical-virtual documents. Many participants did not obviously develop a sense that the physical and virtual surroundings were fused; as a result, certain physical privacy behaviours (e.g., orienting one's body to shield documents from the remote collaborator) were less apparent in the study. However, the circumambient displays generated curiosity about how the spaces were connected, and episodes where breaches were enacted or the spatial correlation was otherwise suggested led some participants to trust the environment less. On the table, the presence of fiducial markers and digital document "shadows" served to cue participants about the impact of hiding and sharing physical documents, however accidental breaches usually went unnoticed.

Users of microblogging social networks produce millions of short messages every day. Retrieving relevant information to a particular event from this sheer volume of data is not a trivial task. In this paper, we present a framework for the retrieval of Twitter posts that are relevant to a set of political debates. Our main contribution is the proposal of a set of strategies for involving the user in the retrieval process, so that by presenting to her meaningful posts to be labeled, the method achieves a noticeably higher accuracy. The correct retrieval or labeling could be provided by an external information source such as a domain expert, or simulated with an oracle. A key aspect of active retrieval methods is to request the labels of the instances that help improve the retrieval accuracy the most, while keeping the number of labeling requests to a minimum. The proposed strategies for selecting labeling requests make use of the textual content of tweets and their structural information. The experimental results show the advantages of the proposed methods and the effectiveness of the selection strategies for involving the user in the retrieval process.

Tone mapping is a computational task of significance in the context of displaying high dynamic range images on low dynamic range devices. While a number of tone mapping algorithms have been proposed and are in common use, there is no single operator that yields optimal results under all conditions. Moreover, obtaining satisfactory mappings often requires the manual tweaking of parameters. This paper proposes interactive evolution as a computational tool for tone mapping. An evolution strategy that blends the results from several tone mapping operators while at the same time adapting their parameters is found to yield promising results with little effort required of the user.

This paper presents an audio classification system that provides improved accuracy, robustness and flexibility over reported content-based audio classification methods. The system reads an input audio file, performs segmentation and classification of the composite sounds contained within the file and, for each sound clip, determines the most plausible matching class of audio in the database. Improvements in the accuracy of audio classification are largely due to the partitioning of the input audio file into homogeneous segments while the incorporation of new class detection offers greater flexibility of use.

This paper presents a model-free and training-free two-phase method for audio segmentation that separates monophonic heterogeneous audio files into acoustically homogeneous regions where each region contains a single sound. A rough segmentation separates audio input into audio clips based on silence detection in the time domain. Then a self-similarity matrix, based on selected audio features in the frequency domain to discover the level of similarity between frames in the audio clip, is calculated. Subsequently an edge detection method is used to find regions in the similarity image that determine plausible sounds in the audio clip. The results of the two phases are combined to form the final boundaries for the input audio. This two-phase method is evaluated using established methods and a standard non-musical database. The method reported here offers more accurate segmentation results than existing methods for audio segmentation. We propose that this approach could be adapted as an efficient preprocessing stage in other audio processing systems such as audio retrieval, classification, music analysis and summarization.

Many network visualizations make the assumption that an administrator has previously determined the subset of data that should be visualized. Yet the problem remains that if the visualization provides no insight into the network events that warrant further consideration, then the administrator must go back to the data to determine what should be visualized next. This is a critical issue given the amount of network data under consideration, only a small portion of which can be examined at any one time. In this paper we present a visualization that provides context for network visualizations by providing a high-level view of network events. Our visualization not only provides a starting point for network visualization, but also reduces the cognitive burden of the analyst by providing a visual paradigm for both the filtering of network data and the selection of network data to drill into and visualize with alternative representations. We demonstrate, through the use of a case study, that our visualization can provide motivation for further investigation into anomalous network activity.

SUMMARY: A phenotypic character has a convex mapping to a given phylogenetic tree if each character state can be assigned a single point of origin in the tree. However, phenomena such as convergent evolution and lateral genetic transfer can lead to intermingling of character states and a consequent non-convex mapping. The phylogenetic heterogeneity of different characters can identify subsets of states that are non-randomly associated or which may have been transferred from one lineage to another. We have developed Radié, an interactive 3D Java application for mapping character states to the leaves and internal edges of a radial phylogenetic tree. In Radié each state of a given character is associated with a unique color, and internal edges with many descendant character states can be represented in a number of different ways to illustrate the diversity within each group. AVAILABILITY: Radié is freely available for download at http://kiwi.cs.dal.ca/~beiko/software-and-data/radie; source code is available upon request from the authors.

Abstract We present a new method for the realistic real‐time simulation of rivers. Our solution includes a 2D fluid solver that simulates the flow of a river's surface, an efficient method for adaptively computing 3D flow information and an animated 3D procedural wave texture that is advected through the fluid via advection particles in order to mimic the highly detailed fluid surfaces that are characteristic of rivers. Our technique that couples animated texture advection with a pseudo‐3D fluid simulation produces stable results that are representative of large‐scale real‐world rivers and suitable for use in real‐time applications. Our system surpasses prior work on real‐time river rendering both with regards to efficiency and visual quality, which we establish through the rendering of rivers tens of kilometers long. Copyright © 2009 John Wiley & Sons, Ltd.

Event logs or log files form an essential part of any network management and administration setup. While log files are invaluable to a network administrator, the vast amount of data they sometimes contain can be overwhelming and can sometimes hinder rather than facilitate the tasks of a network administrator. For this reason several event clustering algorithms for log files have been proposed, one of which is the event clustering algorithm proposed by Risto Vaarandi, on which his simple log file clustering tool (SLCT) is based. The aim of this work is to develop a visualization tool that can be used to view log files based on the clusters produced by SLCT. The proposed visualization tool, which is called LogView, utilizes treemaps to visualize the hierarchical structure of the clusters produced by SLCT. Our results based on different application log files show that LogView can ease the summarization of vast amount of data contained in the log files. This in turn can help to speed up the analysis of event data in order to detect any security issues on a given application.

This paper presents a technique for mixed media non-photorealistic painting and portraiture. The goal of this work is to transform digital images into renderings that approximate the appearance of mixed media artwork, which incorporates two or more traditional visual media. We achieve this by first separating an input image into distinct regions based on the degree of local detail present in the image. Each region is then processed independently with a user-selected NPR filter. This allows the user to treat highly detailed regions differently from regions of low frequency content. The separately processed regions are then smoothly fused in the gradient domain. In addition, we extend our work to the rendering of mixed media portraits. Portraits pose unique challenges that we address with our method of segmentation, which is based on a composite of face detection and image detail. Our approach offers the user a great deal of flexibility over the end result, while at the same time requiring very little input. This input takes the form of a few simple and discrete choices. The results demonstrate an impressive array of transformational possibilities.

We present a method of restyling an image so that it approximates the visual appearance of a work of stained glass. To this end, we develop a novel approach which involves image warping, segmentation, querying, and colorization along with texture synthesis. In our method, a given input image is first segmented. Each segment is subsequently transformed to match real segments of stained glass queried from a database of image exemplars. By using real sources of stained glass, our method produces high quality results in this nascent area of nonphotorealistic rendering. The generation of the stained glass requires only modest amounts of user interaction. This interaction is facilitated with a unique region-merging tool.

We present a simple method of interactive texture editing that utilizes self-similarity to replicate intended operations globally over an image. Inspired by the recent successes of hierarchical approaches to texture synthesis, this method also uses multi-scale neighborhoods to assess the similarity of pixels within a texture. However, neighborhood matching is not employed to generate new instances of a texture. We instead locate similar neighborhoods for the purpose of replicating editing operations on the original texture itself, thereby creating a fundamentally new texture. This general approach is applied to texture painting, cloning and warping. These global operations are performed interactively, most often directed with just a single mouse movement.