Augmented Reality (AR) systems that overlay digital instructions onto the physical world can enhance user performance in complex industrial tasks. Procedural instructions can often be derived from existing approved technical documentation, as reuse reduces authoring effort and ensures compliance with established standards. This paper examines the potential of incorporating a Large Language Model (LLM) to add an intelligent, conversational AR assistant to a static AR manual. We conducted an A/B user study (N = 36) evaluating the effects of adding a conversational assistance layer for a hands-on task on a juice mixer laboratory installation using a HoloLens 2. The baseline condition provided a PDF manual requiring manual step navigation, complemented by situated visual anchors. The second variant supplemented this interface with a conversational assistant that could interpret user queries, provide direct verbal guidance, and automatically jump to the relevant page in the manual. We evaluated these interfaces across three distinct scenarios: a linear task, a non-linear task requiring users to jump between pages for troubleshooting, and a task-handoff where users had to identify the current state of a partially completed procedure. Our findings indicate that despite longer total task completion times, participants using the AI assistant spent significantly less time actively working in the linear and non-linear scenarios, indicating improved task efficiency beyond system latency. Eye-tracking analysis supports the efficiency gain observation as the conversational interface allowed users to focus their visual attention on actively understanding and learning the procedural task. This study highlights the potential of LLM-powered agents for AR guidance, suggesting that overcoming system latency is the critical barrier to their practical deployment in industrial fields.

As virtual reality (VR) headsets become more comfortable and accessible, their growing use in high-stakes, time-critical settings raises concerns about fatigue. Fatigue impairs perceptual and cognitive functioning. It reduces oculomotor accuracy and visual focus, and can lead to an early decline in depth estimation performance. By augmenting the visual presentation with depth information cues, adaptive designs can reduce fatigue-related depth perception errors, enhancing safety and task effectiveness. Explicit cues present depth information directly through text or color, while subtle cues adjust scene properties, such as depth-dependent blur, to convey depth information implicitly without drawing overt attention. We examined how fatigue interacts with different cues in a 27-hour within-subject protocol. Across six overnight sessions (20:00– 07:00), twenty-three participants completed a VR depth perception task at varying fatigue levels and under four cue conditions: no cue (baseline), text, color, and blur. Over the night, vigilance declined, sleepiness and mental effort increased, and simulator sickness rose before stabilizing, independent of cue condition. All augmented cues reduced depth estimation error relative to baseline. Text yielded the largest and most consistent accuracy gains, especially for farther targets and later sessions. At peak fatigue, response probability dipped for text but remained stable for blur and color, indicating an accuracy versus responsiveness trade-off. These results support mixed adaptive designs that default to subtle cues to preserve responsiveness at low alertness and introduce explicit overlays when precise metric information is needed.

Navigation is a fundamental task supporting guided exploration and wayfinding as standalone or as part of other immersive applications. Previous research showed that navigational cues do not only impact wayfinding performance but can also affect perceptual and cognitive processes, e.g. divide attention and impair spatial memory. This study investigates whether varying the density of visual navigation cues can influence situational awareness. Additionally, we examine how cue density affects navigation usability and task performance. We compare three visual cue designs, ranging from high to low density: (1) PathLine, (2) ArrowTrail and (3) TurnMarker. We designed a user study augmenting a 3D scanned digital twin of a building with virtual machinery to simulate a factory floor maintenance task, where the cues guided participants to the next point of interest. A secondary task, the reporting of anomalies, was installed to assess situational awareness. Results showed that performance metrics remained unaffected by cue type, situational awareness and user experience results showed significant differences. Notably, the ArrowTrail cue, the medium-density design, was preferred by most participants and yielded the best overall results, e.g. in terms of anomaly detection and reaction time. These findings suggest that moderate cue density may offer an optimal balance between effective guidance and maintaining environmental awareness.

Human cognition involves intricate neural processes for error perception and correction. These processes are crucial in error-monitoring processes such as feedback, learning, control, and decision-making. We present a complete workflow using explainable artificial intelligence (XAI) to guide the feature extraction of electroencephalographic (EEG) signals in a classification task with error-related brain responses. The identification of relevant channels for classification problems has practical relevance, as a dense electrode setup reduces the usability of brain-computer interfaces (BCIs). Specialists can inspect and select seemingly important sensors based on knowledge of brain regions and neural patterns. However, reduced configurations often harm accurate model performance. The contribution of this work lies in demonstrating that XAI can be used to inform the extraction of relevant temporal and spatial information with a strong connection to machine learning model sensitivity. We employed a local and a global XAI method to i) evaluate consistency with expert knowledge, ii) identify relevant time points for asynchronous error decoding, and iii) systematically reduce the error decoding EEG setup. This advances the integration of XAI in neuroscience, thus contributing to the design of practical BCIs.

This paper presents a comprehensive study of virtual 3D object manipulation along 4DoF on real surfaces in mixed reality (MR), using hand-based and tangible interactions. A custom cylindrical tangible proxy leverages affordances of physical knobs and tabletop support for stable input. We evaluate both modalities across isolated tasks (2DoF translation, 1DoF rotation/scaling), semi-combined (3DoF translation+rotation), and full 4DoF compound manipulation. We offer analyses of hand interactions, tangible interactions, and their comparison in MR tasks. For hand interactions, compound tasks required repetitive corrections, increasing completion times—yet surprisingly, rotation errors were smaller in compound tasks than in rotation-only tasks. Tangible interactions exhibited significantly larger errors in translation, rotation, and scaling during compound tasks compared to isolated tasks. Crucially, tangible interactions outperformed hand interactions in precision, likely due to tabletop support and constrained 4DoF design. These findings inform designers opting for hand-only interaction (highlighting trade-offs in compound tasks) and those leveraging tangibles (emphasizing precision gains despite compound-task challenges).

Continual learning (CL) provides a framework for training models in ever-evolving environments. Although re-occurrence of previously seen objects or tasks is common in real-world problems, the concept of repetition in the data stream is not often considered in standard benchmarks for CL. Unlike with the rehearsal mechanism in buffer-based strategies, where sample repetition is controlled by the strategy, repetition in the data stream naturally stems from the environment. This report provides a summary of the CLVision challenge at CVPR 2023, which focused on the topic of repetition in class-incremental learning. The report initially outlines the challenge objective and then describes three solutions proposed by finalist teams that aim to effectively exploit the repetition in the stream to learn continually. The experimental results from the challenge highlight the effectiveness of ensemble-based solutions that employ multiple versions of similar modules, each trained on different but overlapping subsets of classes. This report underscores the transformative potential of taking a different perspective in CL by employing repetition in the data stream to foster innovative strategy design. • An overview of the continual learning challenge of the CLVision workshop at CVPR 2023. • Novel benchmarks focussing on the topic of repetition in continual learning. • Description and discussion of the strategies submitted by the winning teams. • The results highlight the remarkable effectiveness of ensemble-based solutions.

Many industrial sectors rely on well-trained employees that are able to operate complex machinery. In this work, we demonstrate an immersive assistance system powered by Artificial Intelligence (AI) that supports users in performing complex tasks in industrial environments. Our system leverages a Virtual Reality (VR) environment that resembles a juice mixer setup. This digital twin of a physical setup simulates complex industrial machinery used to mix preparations or liquids (e.g., similar to the pharmaceutical industry) and includes various containers, sensors, pumps, and flow controllers. This setup demonstrates our system’s capabilities in a controlled environment while acting as a proof-of-concept for broader industrial applications. The core components of our multimodal AI assistant are a large language model and a speech-to-text model that process a video and audio recording of an expert performing the task in a VR environment. The video and speech input extracted from the expert’s video enables it to provide step-by-step guidance to support users in executing complex tasks. This demonstration showcases the potential of our AI-powered assistant to reduce cognitive load, increase productivity, and enhance safety in industrial environments.

Abstract In recent years, the development of Optical Music Recognition (OMR) has progressed significantly. However, music cultures with smaller communities have only recently been considered in this process. This results in a lack of adequate ground truth datasets needed for the development and benchmarking of OMR systems. In this work, the KuiSCIMA (Jiang Kui Score Images for Musicological Analysis) dataset is introduced. KuiSCIMA is the first machine-readable dataset of the suzipu notations in Jiang Kui’s collection Baishidaoren Gequ from 1202. Collected from five different woodblock print editions, the dataset contains 21797 manually annotated instances on 153 pages in total, from which 14500 are text character annotations, and 7297 are suzipu notation symbols. The dataset comes with an open-source tool which allows editing, visualizing, and exporting the contents of the dataset files. In total, this contribution promotes the preservation and understanding of cultural heritage through digitization.

Understanding an environment relies on human sensory systems, with visual perception as the primary source for defining spatial relationships and estimating distances. The visual system uses natural cues, each offering partial information that can lead to bias and conflicts, especially when ambiguous. Virtual Reality (VR) environments challenge these natural depth cues with discrepancies in perspective and variations in light and shadow depiction, leading to potential confusion in depth perception. However, VR also allows for the isolation and study of these cues. This paper introduces artificial subtle cues (texture blur) to enhance natural depth information in VR. Our results show that augmenting natural depth cues with artificial ones improves depth prediction accuracy and spatial relationship awareness. Subtle blur cues enhance depth estimation without participants’ subjective awareness of the augmentation, suggesting that such subtle cueing can effectively enhance depth perception.

Despite the availability of international prize-money competitions, scaled vehicles, and simulation environments, research on autonomous racing and the control of sports cars operating close to the limit of handling has been limited by the high costs of vehicle acquisition and management, as well as the limited physics accuracy of open-source simulators. In this paper, we propose a racing simulation platform based on the simulator Assetto Corsa to test, validate, and benchmark autonomous driving algorithms, including reinforcement learning (RL) and classical Model Predictive Control (MPC), in realistic and challenging scenarios. Our contributions include the development of this simulation platform, several state-of-the-art algorithms tailored to the racing environment, and a comprehensive dataset collected from human drivers. Additionally, we evaluate algorithms in the offline RL setting. All the necessary code (including environment and benchmarks), working examples, and datasets are publicly released and can be found at: https://github.com/dasGringuen/assettocorsagym.

Abstract While humans can effortlessly pick a view from multiple streams, automatically choosing the best view is a challenge. Choosing the best view from multi-camera streams poses a problem regarding which objective metrics should be considered. Existing works on view selection lack consensus about which metrics should be considered to select the best view. The literature on view selection describes diverse possible metrics. And strategies such as information-theoretic, instructional design, or aesthetics-motivated fail to incorporate all approaches. In this work, we postulate a strategy incorporating information-theoretic and instructional design-based objective metrics to select the best view from a set of views. Traditionally, information-theoretic measures have been used to find the goodness of a view, such as in 3D rendering. We adapted a similar measure known as the viewpoint entropy for real-world 2D images. Additionally, we incorporated similarity penalization to get a more accurate measure of the entropy of a view, which is one of the metrics for the best view selection. Since the choice of the best view is domain-dependent, we chose demonstration-based training scenarios as our use case. The limitation of our chosen scenarios is that they do not include collaborative training and solely feature a single trainer. To incorporate instructional design considerations, we included the trainer’s body pose, face, face when instructing, and hands visibility as metrics. To incorporate domain knowledge we included predetermined regions’ visibility as another metric. All of those metrics are taken into account to produce a parameterized view recommendation approach for demonstration-based training. An online study using recorded multi-camera video streams from a simulation environment was used to validate those metrics. Furthermore, the responses from the online study were used to optimize the view recommendation performance with a normalized discounted cumulative gain (NDCG) value of 0.912, which shows good performance with respect to matching user choices.

Error perception is known to elicit distinct brain patterns, which can be used to improve the usability of systems facilitating human-computer interactions, such as brain-computer interfaces. This requires a high-accuracy detection of erroneous events, e.g., misinterpretations of the user’s intention by the interface, to allow for suitable reactions of the system. In this work, we concentrate on steering-based navigation tasks. We present a combined electroencephalography-virtual reality (VR) study investigating different approaches for error detection and simultaneously exploring the corrective human behavior to erroneous events in a VR flight simulation. We could classify different errors allowing us to analyze neural signatures of unexpected changes in the VR. Moreover, the presented models could detect errors faster than participants naturally responded to them. This work could contribute to developing adaptive VR applications that exclusively rely on the user’s physiological information.

This paper provides evidence indicating that the most commonly used metric for validating feature attribution methods in eXplainable AI (XAI) is misleading when applied to time series data. To evaluate whether an XAI method attributes importance to relevant features, these are systematically perturbed while measuring the impact on the performance of the classifier. The assumption is that a drastic performance reduction with increasing perturbation of relevant features indicates that these are indeed relevant. We demonstrate empirically that this assumption is incomplete without considering low relevance features in the used metrics. We introduce a novel metric, the Perturbation Effect Size, and demonstrate how it complements existing metrics to offer a more faithful assessment of importance attribution. Finally, we contribute a comprehensive evaluation of attribution methods on time series data, considering the influence of perturbation methods and region size selection.

When designing a smart glove for gesture recognition, the set of sensors available and their layout on the glove are crucial. However, once a computational model reaches acceptable recognition accuracy, it is often not clear which sensors are more important for the task. Nor whether some sensors can be strategically removed while retaining similar performance in order to save cost. Furthermore, when aiming for a personalized setup, there can be minor deviation in how gestures are performed by each participant, and so the importance of a sensor may vary between participants. In this paper, we use feature selection to explore whether a personalised glove can be produced, and whether the set of significant sensors persist between users. We present a deep learning algorithm which utilises a layer of weights to estimate the importance of each sensor in relation to each other. Besides estimating importance in relation to recognition accuracy, it is demonstrated how the importance estimates can be extended to take into account factors external to the computational model, such as costs. This allows for a cost effective elimination of sensors to reduce hardware redundancy whilst having a controlled impact on performance. We provide 2 methods: generic or specific. The generic method exploits the importance estimate from all participants to select a set of sensors for removal. Whereas the specific method estimates importance, and removes sensors based on individuals to provide a personalised setup.

In this paper, we analyze the benefits of adopting user interfaces that offer control and transparency for searching in contexts of learning activities. Concretely, we conducted a user study with pharmacy students performing a problem-solving task in the course of a university lecture. The task involved finding scientific papers containing relevant information to solve a clinical case. Students were split into two independent groups and assigned one search tool to perform the task. The baseline group worked with PubMed, a popular search engine in the life sciences domain, whereas the second half of the class was assigned an exploratory search system (ESS) designed for control and transparency. In the analysis, we cover the objective and subjective dimensions of the task outcomes. Firstly, the objective analysis addresses the inherent difficulty of the search task in a learning scenario and identifies certain improvements in performance for those students using the ESS, most notably when searching for primary-source content. The subjective analysis investigates the human factors side, providing evidence that the ESS effectively increases the perception of control and transparency and is able to produce a better user experience. Lastly, we report on perceived learning as a subjective dimension measured separately from user experience.

We present a visual interface for communicating the internal state of a coffee machine via a tree metaphor. Nature-inspired representations have a positive impact on human well-being. We also hypothesize that representing the coffee machine as a tree stimulates emotional connection to it, which leads to better maintenance performance.The first study assessed the understandability of the tree representation, comparing it with icon-based and chart-based representations. An online survey with 25 participants indicated no significant mean error difference between representations.A two-week field study assessed the maintenance performance of 12 participants, comparing the tree representation with the icon-based representation. Based on 240 interactions with the coffee machine, we concluded that participants understood the machine states significantly better in the tree representation. Their comments and behavior indicated that the tree representation encouraged an emotional engagement with the machine. Moreover, the participants performed significantly more optional maintenance tasks with the tree representation.

Information-seeking tasks with learning or investigative purposes are usually referred to as exploratory search. Exploratory search unfolds as a dynamic process where the user, amidst navigation, trial and error, and on-the-fly selections, gathers and organizes information (resources). A range of innovative interfaces with increased user control has been developed to support the exploratory search process. In this work, we present our attempt to increase the power of exploratory search interfaces by using ideas of social search—for instance, leveraging information left by past users of information systems. Social search technologies are highly popular today, especially for improving ranking. However, current approaches to social ranking do not allow users to decide to what extent social information should be taken into account for result ranking. This article presents an interface that integrates social search functionality into an exploratory search system in a user-controlled way that is consistent with the nature of exploratory search. The interface incorporates control features that allow the user to (i) express information needs by selecting keywords and (ii) to express preferences for incorporating social wisdom based on tag matching and user similarity. The interface promotes search transparency through color-coded stacked bars and rich tooltips. This work presents the full series of evaluations conducted to, first, assess the value of the social models in contexts independent to the user interface, in terms of objective and perceived accuracy. Then, in a study with the full-fledged system, we investigated system accuracy and subjective aspects with a structural model revealing that when users actively interacted with all of its control features, the hybrid system outperformed a baseline content-based–only tool and users were more satisfied.

This paper proposes methods of optimising alphabet encoding for skin reading in order to avoid perception errors. First, a user study with 16 participants using two body locations serves to identify issues in recognition of both individual letters and words. To avoid such issues, a two-step optimisation method of the symbol encoding is proposed and validated in a second user study with eight participants using the optimised encoding with a seven vibromotor wearable layout on the back of the hand. The results show significant improvements in the recognition accuracy of letters (97%) and words (97%) when compared to the non-optimised encoding.

Proficiency in any form of reading requires a considerable amount of practice. With exposure, people get better at recognising words, because they develop strategies that enable them to read faster. This paper describes a study investigating recognition of words encoded with a 6-channel vibrotactile display. We train 22 users to recognise ten letters of the English alphabet. Additionally, we repeatedly expose users to 12 words in the form of training and reinforcement testing. Then, we test participants on exposed and unexposed words to observe the effects of exposure to words. Our study shows that, with exposure to words, participants did significantly improve on recognition of exposed words. The findings suggest that such a word exposure technique could be used during the training of novice users in order to boost the word recognition of a particular dictionary of words.

Previous research has demonstrated the feasibility of conveying vibrotactile encoded information efficiently using wearable devices. Users can understand vibrotactile encoded symbols and complex messages combining such symbols. Such wearable devices can find applicability in many multitasking use cases. Nevertheless, for multitasking, it would be necessary for the perception and comprehension of vibrotactile information to be less attention demanding and not interfere with other parallel tasks. We present a user study which investigates whether high speed vibrotactile encoded messages can be perceived in the background while performing other concurrent attention-demanding primary tasks. The vibrotactile messages used in the study were limited to symbols representing letters of English Alphabet. We observed that users could very accurately comprehend vibrotactile such encoded messages in the background and other parallel tasks did not affect users performance. Additionally, the comprehension of such messages did also not affect the performance of the concurrent primary task as well. Our results promote the use of vibrotactile information transmission to facilitate multitasking.

Labeling a real network dataset is specially expensive in computer security, as an expert has to ponder several factors before assigning each label. This paper describes an interactive intelligent system to support the task of identifying hostile behaviors in network logs. The RiskID application uses visualizations to graphically encode features of network connections and promote visual comparison. In the background, two algorithms are used to actively organize connections and predict potential labels: a recommendation algorithm and a semi-supervised learning strategy. These algorithms together with interactive adaptions to the user interface constitute a behavior recommendation. A study is carried out to analyze how the algorithms for recommendation and prediction influence the workflow of labeling a dataset. The results of a study with 16 participants indicate that the behaviour recommendation significantly improves the quality of labels. Analyzing interaction patterns, we identify a more intuitive workflow used when behaviour recommendation is available.

Digital libraries and services enable users to access large amounts of data on demand. Yet, quality assessment of information encountered on the Internet remains an elusive open issue. For example, Wikipedia, one of the most visited platforms on the Web, hosts thousands of user-generated articles and undergoes 12 million edits/contributions per month. User-generated content is undoubtedly one of the keys to its success but also a hindrance to good quality. Although Wikipedia has established guidelines for the “perfect article,” authors find it difficult to assert whether their contributions comply with them and reviewers cannot cope with the ever-growing amount of articles pending review. Great efforts have been invested in algorithmic methods for automatic classification of Wikipedia articles (as featured or non-featured) and for quality flaw detection. Instead, our contribution is an interactive tool that combines automatic classification methods and human interaction in a toolkit, whereby experts can experiment with new quality metrics and share them with authors that need to identify weaknesses to improve a particular article. A design study shows that experts are able to effectively create complex quality metrics in a visual analytics environment. In turn, a user study evidences that regular users can identify flaws, as well as high-quality content based on the inspection of automatic quality scores.

We developed a new concept to improve the efficiency of visual analysis through visual recommendations. It uses a novel eye-gaze based recommendation model that aids users in identifying interesting time-series patterns. Our model combines time-series features and eye-gaze interests, captured via an eye-tracker. Mouse selections are also considered. The system provides an overlay visualization with recommended patterns, and an eye-history graph, that supports the users in the data exploration process. We conducted an experiment with 5 tasks where 30 participants explored sensor data of a wind turbine. This work presents results on pre-attentive features, and discusses the precision/recall of our model in comparison to final selections made by users. Our model helps users to efficiently identify interesting time-series patterns.

This paper investigates the effects of using passive haptic learning to train the skill of comprehending text from vibrotactile patterns. The method of transmitting messages, skin-reading, is effective at conveying rich information but its active training method requires full user attention, is demanding, time-consuming, and tedious. Passive haptic learning offers the possibility to learn in the background while performing another primary task. We present a study investigating the use of passive haptic learning to train for skin-reading.

Vibrotactile skin-reading uses wearable vibrotactile displays to convey dynamically generated textual information. Such wearable displays have potential to be used in a broad range of applications. Nevertheless, the reading process is passive, and users have no control over the reading flow. To compensate for such drawback, this paper investigates what kind of interactions are necessary for vibrotactile skin reading and the modalities of such interactions. An interaction concept for skin reading was designed by taking into account the reading as a process. We performed a formative study with 22 participants to assess reading behaviour in word and sentence reading using a six-channel wearable vibrotactile display. Our study shows that word based interactions in sentence reading are more often used and preferred by users compared to character-based interactions and that users prefer gesture-based interaction for skin reading. Finally, we discuss how such wearable vibrotactile displays could be extended with sensors that would enable recognition of such gesture-based interaction. This paper contributes a set of guidelines for the design of wearable haptic displays for text communication.

Information-seeking tasks with learning or investigative purposes are usually referred to as exploratory search. Exploratory search unfolds as a dynamic process where the user, amidst navigation, trial-and-error and on-the-fly selections, gathers and organizes information (resources). A range of innovative interfaces with increased user control have been developed to support exploratory search process. In this work we present our attempt to increase the power of exploratory search interfaces by using ideas of social search, i.e., leveraging information left by past users of information systems. Social search technologies are highly popular nowadays, especially for improving ranking. However, current approaches to social ranking do not allow users to decide to what extent social information should be taken into account for result ranking. This paper presents an interface that integrates social search functionality into an exploratory search system in a user-controlled way that is consistent with the nature of exploratory search. The interface incorporates control features that allow the user to (i) express information needs by selecting keywords and (ii) to express preferences for incorporating social wisdom based on tag matching and user similarity. The interface promotes search transparency through color-coded stacked bars and rich tooltips. In an online study investigating system accuracy and subjective aspects with a structural model we found that, when users actively interacted with all its control features, the hybrid system outperformed a baseline content-based-only tool and users were more satisfied.

This paper aims to identify self-regulation strategies from students' interactions with the learning management system (LMS). We used learning analytics techniques to identify metacognitive and cognitive strategies in the data. We define three research questions that guide our studies analyzing i) self-assessments of motivation and self regulation strategies using standard methods to draw a baseline, ii) interactions with the LMS to find traces of self regulation in observable indicators, and iii) self regulation behaviours over the course duration. The results show that the observable indicators can better explain self-regulatory behaviour and its influence in performance than preliminary subjective assessments.

A video tutorial effectively conveys complex motions, but may be hard to follow precisely because of its restriction to a predetermined viewpoint. Augmented reality (AR) tutorials have been demonstrated to be more effective. We bring the advantages of both together by interactively retargeting conventional, two-dimensional videos into three-dimensional AR tutorials. Unlike previous work, we do not simply overlay video, but synthesize 3D-registered motion from the video. Since the information in the resulting AR tutorial is registered to 3D objects, the user can freely change the viewpoint without degrading the experience. This approach applies to many styles of video tutorials. In this work, we concentrate on a class of tutorials which alter the surface of an object.

Digital libraries and services enable users to access large amounts of data on demand. Yet, quality assessment of information encountered on the Internet remains an elusive open issue. For example, Wikipedia, one of the most visited platforms on the Web, hosts thousands of user-generated articles and undergoes 12 million edits/contributions per month. User-generated content is undoubtedly one of the keys to its success, but also a hindrance to good quality: contributions can be of poor quality because anyone, even anonymous users, can participate. Though Wikipedia has defined guidelines as to what makes the perfect article, authors find it difficult to assert whether their contributions comply with them and reviewers cannot cope with the ever growing amount of articles pending review. Great efforts have been invested in algorithmic methods for automatic classification of Wikipedia articles (as featured or non-featured) and for quality flaw detection. However, little has been done to support quality assessment of user-generated content through interactive tools that combine automatic methods and human intelligence. We developed WikiLyzer, a Web toolkit comprising three interactive applications designed to assist (i) knowledge discovery experts in creating and testing metrics for quality measurement, (ii) Wikipedia users searching for good articles, and (iii) Wikipedia authors that need to identify weaknesses to improve a particular article. A design study sheds a light on how experts could create complex quality metrics with our tool, while a user study reports on its usefulness to identify high-quality content.

Whenever users engage in gathering and organizing new information, searching and browsing activities emerge at the core of the exploration process. As the process unfolds and new knowledge is acquired, interest drifts occur inevitably and need to be accounted for. Despite the advances in retrieval and recommender algorithms, real-world interfaces have remained largely unchanged: results are delivered in a relevance-ranked list. However, it quickly becomes cumbersome to reorganize resources along new interests, as any new search brings new results. We introduce an interactive user-driven tool that aims at supporting users in understanding, refining, and reorganizing documents on the fly as information needs evolve. Decisions regarding visual and interactive design aspects are tightly grounded on a conceptual model for exploratory search. In other words, the different views in the user interface address stages of awareness, exploration, and explanation unfolding along the discovery process, supported by a set of text-mining methods. A formal evaluation showed that gathering items relevant to a particular topic of interest with our tool incurs in a lower cognitive load compared to a traditional ranked list. A second study reports on usage patterns and usability of the various interaction techniques within a free, unsupervised setting.

Whenever users engage in gathering and organizing new information, searching and browsing activities emerge at the core of the exploration process. As the process unfolds and new knowledge is acquired, interest drifts occur inevitably and need to be accounted for. Despite the advances in retrieval and recommender algorithms, real-world interfaces have remained largely unchanged: results are delivered in a relevance-ranked list. However, it quickly becomes cumbersome to reorganize resources along new interests, as any new search brings new results. We introduce uRank and investigate interactive methods for understanding, refining and reorganizing documents on-the-fly as information needs evolve. uRank includes views summarizing the contents of a recommendation set and interactive methods conveying the role of users' interests through a recommendation ranking. A formal evaluation showed that gathering items relevant to a particular topic of interest with uRank incurs in lower cognitive load compared to a traditional ranked list. A second study consisting in an ecological validation reports on usage patterns and usability of the various interaction techniques within a free, more natural setting.

Visualizations have a distinctive advantage when dealing with the information overload problem: Because they are grounded in basic visual cognition, many people understand them. However, creating proper visualizations requires specific expertise of the domain and underlying data. Our quest in this article is to study methods to suggest appropriate visualizations autonomously. To be appropriate, a visualization has to follow known guidelines to find and distinguish patterns visually and encode data therein. A visualization tells a story of the underlying data; yet, to be appropriate, it has to clearly represent those aspects of the data the viewer is interested in. Which aspects of a visualization are important to the viewer? Can we capture and use those aspects to recommend visualizations? This article investigates strategies to recommend visualizations considering different aspects of user preferences. A multi-dimensional scale is used to estimate aspects of quality for visualizations for collaborative filtering. Alternatively, tag vectors describing visualizations are used to recommend potentially interesting visualizations based on content. Finally, a hybrid approach combines information on what a visualization is about (tags) and how good it is (ratings). We present the design principles behind VizRec , our visual recommender. We describe its architecture, the data acquisition approach with a crowd sourced study, and the analysis of strategies for visualization recommendation.

Graphs are widely used to represent relationships between entities. Indeed, their simplicity in depicting connectedness backed by a mathematical formalism, make graphs an ideal metaphor to convey relatedness between entities irrespective of the domain. However, graphs pose several challenges for visual analysis. A large number of entities or a densely connected set quickly render the graph unreadable due to clutter. Typed relationships leading to multigraphs cannot clearly be represented in hierarchical layout or edge bundling, common clutter reduction techniques. We propose a novel approach to visual analysis of complex graphs based on two metaphors: semantic blossom and selective expansion. Instead of showing the whole graph, we display only a small representative subset of nodes, each with a compressed summary of relations in a semantic blossom. Users apply selective expansion to traverse the graph and discover the subset of interest. A preliminary evaluation showed that our approach is intuitive and useful for graph exploration and provided insightful ideas for future improvements.

In Augmented Reality (AR), ghosted views allow a viewer to explore hidden structure within the real-world environment. A body of previous work has explored which features are suitable to support the structural interplay between occluding and occluded elements. However, the dynamics of AR environments pose serious challenges to the presentation of ghosted views. While a model of the real world may help determine distinctive structural features, changes in appearance or illumination detriment the composition of occluding and occluded structure. In this paper, we present an approach that considers the information value of the scene before and after generating the ghosted view. Hereby, a contrast adjustment of preserved occluding features is calculated, which adaptively varies their visual saliency within the ghosted view visualization. This allows us to not only preserve important features, but to also support their prominence after revealing occluded structure, thus achieving a positive effect on the perception of ghosted views.

Real-time three-dimensional acquisition of real-world scenes has many important applications in computer graphics, computer vision and human-computer interaction. Inexpensive depth sensors such as the Microsoft Kinect allow to leverage the development of such applications. However, this technology is still relatively recent, and no detailed studies on its scalability to dense and view-independent acquisition have been reported. This paper addresses the question of what can be done with a larger number of Kinects used simultaneously. We describe an interference-reducing physical setup, a calibration procedure and an extension to the KinectFusion algorithm, which allows to produce high quality volumetric reconstructions from multiple Kinects whilst overcoming systematic errors in the depth measurements. We also report on enhancing image based visual hull rendering by depth measurements, and compare the results to KinectFusion. Our system provides practical insight into achievable spatial and radial range and into bandwidth requirements for depth data acquisition. Finally, we present a number of practical applications of our system.

In this paper, we explore techniques that aim to improve site understanding for outdoor Augmented Reality (AR) applications. While the first person perspective in AR is a direct way of filtering and zooming on a portion of the data set, it severely narrows overview of the situation, particularly over large areas. We present two interactive techniques to overcome this problem: multi-view AR and variable perspective view. We describe in details the conceptual, visualization and interaction aspects of these techniques and their evaluation through a comparative user study. The results we have obtained strengthen the validity of our approach and the applicability of our methods to a large range of application domains.

In augmented reality, it is often necessary to draw the user's attention to particular objects in the real world without distracting her from her task. We explore the effectiveness of directing a user's attention by imperceptibly modifying existing features of a video. We present three user studies of the effects of applying a saliency modulation technique to video; evaluating modulation awareness, attention, and memory. Our results validate the saliency modulation technique as an alternative means to convey information to the user, suggesting attention shifts and influencing recall of selected regions without perceptible changes to visual input.

In this paper, we report on four generations of display-sensor platforms for handheld augmented reality. The paper is organized as a compendium of requirements that guided the design and construction of each generation of the handheld platforms. The first generation, reported in [17]), was a result of various studies on ergonomics and human factors. Thereafter, each following iteration in the design-production process was guided by experiences and evaluations that resulted in new guidelines for future versions. We describe the evolution of hardware for handheld augmented reality, the requirements and guidelines that motivated its construction.

This paper focuses on the design of devices for handheld spatial interaction. In particular, it addresses the requirements and construction of a new platform for interactive AR, described from an ergonomics stance, prioritizing human factors of spatial interaction. The result is a multi-configurable platform for spatial interaction, evaluated in two AR application scenarios. The user tests validate the design with regards to grip, weight balance and control allocation, and provide new insights on the human factors involved in handheld spatial interaction.

This article introduces a framework to generate three-dimensional models for augmented reality (AR) including semantics. The semantics of 3D models have been studied previously in the field of intelligent virtual environments, but the process of creating real- world models containing such information have received little attention. We introduce a method for the creation of semantic, 3D models of the environment using an AR application named InventAry. Assisted by an ontology, InventAry enforces the creation of combined geometric and semantic environment model.

This paper presents first results of an interaction design study performed for a novel handheld interaction system. Human factors of mid-size, self-containing wearable computer systems are explored.