Smart and secure access to vehicles is a crucial aspect of the evolving automotive industry. This paper focuses on the development of an end-to-end multimodal biometric recognition framework that identifies people walking towards a vehicle from an RGB video feed. The framework is based on a deep-learning pipeline for person detection and tracking, face and gait feature extraction, and fusion of the two modalities at the score and feature level. Traditional face recognition systems can suffer from variations in lighting and occlusions. In order to deal with these issues, the proposed framework integrates face and gait features with the aim to enhance accuracy. The pipeline is modular, enabling seamless integration of new models for each step of person identification without the need for additional training. Baseline face and gait recognition models, as well as score- and feature-level fusion techniques are evaluated on subsets of the CASIA-A and CASIA-B datasets. Experimental results show that weighted mean score-level fusion significantly improves both Rank-1 accuracy and verification accuracy (TAR@FAR=10−5) over unimodal baselines. Overall, reported work provides insights into current limitations and suggests directions for future research about secure identity verification in vehicles.

Zero-shot learning (ZSL) aims to train deep neural networks to recognize objects from unseen classes, starting from a semantic description of the concepts. Neuro-symbolic (NeSy) integration refers to a class of techniques that incorporate symbolic knowledge representation and reasoning with the learning capabilities of deep neural networks. However, to date, few studies have explored how to leverage NeSy techniques to inject prior knowledge during the training process to boost ZSL capabilities. Here, we present Fuzzy Logic Prototypical Network (FLPN) that formulates the classification task as prototype matching in a visual-semantic embedding space, which is trained by optimizing a NeSy loss. Specifically, FLPN exploits the Logic Tensor Network (LTN) framework to incorporate background knowledge in the form of logical axioms by grounding a first-order logic language as differentiable operations between real tensors. This prior knowledge includes class hierarchies (classes and macroclasses) along with robust high-level inductive biases. The latter allow, for instance, to handle exceptions in class-level attributes and to enforce similarity between images of the same class, preventing premature overfitting to seen classes and improving overall performance. Both class-level and attribute-level prototypes through an attention mechanism specialized for either convolutional- or transformer-based backbones. FLPN achieves state-of-the-art performance on the GZSL benchmarks AWA2 and SUN, matching or exceeding the performance of competing algorithms with minimal computational overhead. The code is available at https://github.com/FrancescoManigrass/FLPN .

The potential and promise of deep learning systems to provide an independent assessment and relieve radiologists' burden in screening mammography have been recognized in several studies. However, the low cancer prevalence, the need to process high-resolution images, and the need to combine information from multiple views and scales still pose technical challenges. Multi-view architectures that combine information from the four mammographic views to produce an exam-level classification score are a promising approach to the automated processing of screening mammography. However, training such architectures from exam-level labels, without relying on pixel-level supervision, requires very large datasets and may result in suboptimal accuracy. Emerging architectures such as Visual Transformers (ViT) and graph-based architectures can potentially integrate ipsi-lateral and contra-lateral breast views better than traditional convolutional neural networks, thanks to their stronger ability of modeling long-range dependencies. In this paper, we extensively evaluate novel transformer-based and graph-based architectures against state-of-the-art multi-view convolutional neural networks, trained in a weakly-supervised setting on a middle-scale dataset, both in terms of performance and interpretability. Extensive experiments on the CSAW dataset suggest that, while transformer-based architecture outperform other architectures, different inductive biases lead to complementary strengths and weaknesses, as each architecture is sensitive to different signs and mammographic features. Hence, an ensemble of different architectures should be preferred over a winner-takes-all approach to achieve more accurate and robust results. Overall, the findings highlight the potential of a wide range of multi-view architectures for breast cancer classification, even in datasets of relatively modest size, although the detection of small lesions remains challenging without pixel-wise supervision or ad-hoc networks.

The growing availability of affordable Virtual Reality (VR) hardware and the increasing interest in the Metaverse are driving the expansion of Social VR (SVR) platforms. These platforms allow users to embody avatars in immersive social virtual environments, enabling real-time interactions using consumer devices. Beyond merely replicating real-life social dynamics, SVR platforms offer opportunities to surpass real-world constraints by augmenting these interactions. One example of such augmentation is Artificial Facial Mimicry (AFM), which holds significant potential to enhance social experiences. Mimicry, the unconscious imitation of verbal and non-verbal behaviors, has been shown to positively affect human-agent interactions, yet its role in avatar-mediated human-to-human communication remains under-explored. AFM presents various possibilities, such as amplifying emotional expressions, or substituting one emotion for another to better align with the context. Furthermore, AFM can address the limitations of current facial tracking technologies in fully capturing users' emotions. To investigate the potential benefits of AFM in SVR, an automated AM system was developed. This system provides AFM, along with other kinds of head mimicry (nodding and eye contact), and it is compatible with consumer VR devices equipped with facial tracking. This system was deployed within a test-bench immersive SVR application. A between-dyads user study was conducted to assess the potential benefits of AFM for interpersonal communication while maintaining avatar behavioral naturalness, comparing the experiences of pairs of participants communicating with AFM enabled against a baseline condition. Subjective measures revealed that AFM improved interpersonal closeness, aspects of social attraction, interpersonal trust, social presence, and naturalness compared to the baseline condition. These findings demonstrate AFM's positive impact on key aspects of social interaction and highlight its potential applications across various SVR domains.

Background Over the last few years, the rapid advancement of technology has led to the development of many approaches to digitalization. In this respect, metaverse provides 3D persistent virtual environments that can be used to access digital content, meet virtually, and perform several professional and leisure tasks. Among the numerous technologies supporting the metaverse, immersive Virtual Reality (VR) plays a primary role and offers highly interactive social experiences. Despite growing interest in this area, there are no clear design guidelines for creating environments tailored to the metaverse. Methods This study seeks to advance research in this area by moving from state-of-the-art studies on the design of immersive virtual environments in the context of metaverse and proposing how to integrate cutting-edge technologies within this context. Specifically, the best practices were identified by i) analyzing literature studies focused on human behavior in immersive virtual environments, ii) extracting common features of existing social VR platforms, and iii) conducting interviews with experts in a specific application domain. Specifically, this study considered the creation of a new virtual environment for MetaLibrary, a VR-based social platform aimed at integrating public libraries into metaverse. Several implementation challenges and additional requirements have been identified for the development of virtual environments (VEs). These elements were considered in the selection of specific cutting-edge technologies and their integration into the development process. A user study was also conducted to investigate some design aspects (namely lighting conditions and richness of the scene layout) for which deriving clear indications from the above analysis was not possible because different alternative configurations could be chosen. Results The work reported in this paper seeks to bridge the gap between existing VR platforms and related literature in the field, on the one hand, and requirements regarding immersive virtual environments for the metaverse, on the other hand, by reporting a set of best practices which were used to build a social virtual environment that meets users' expectations and needs. Conclusions Results suggest that carefully designed virtual environments can positively affect user experience and interaction within metaverse. The insights gained from this study offer valuable cues for developing immersive virtual environments for the metaverse to deliver more effective and engaging experiences.

Scene Graph Generation (SGG) is a powerful tool for autonomous vehicles to understand their environment. In this paper, a novel one-stage neuro-symbolic architecture called nEuro-Symbolic Relation trAnsformer (ESRA) is proposed and its applications to SGG in the field of autonomous driving are investigated. This one-stage architecture can perform both object and relation recognition in a single step, attempting to incorporate prior knowledge in the form of logical propositions grounded by a Logic Tensor Network (LTN). To the best of our knowledge, this is the first attempt to combine a transformer-based architecture with an LTN for SGG. The results show that the integration of LTN increases mean recall (mR) by up to 21% in the best configuration, with mAP achieving an increase of up to 19%.

Technologies such as chroma-key, LED walls, motion capture (mocap), 3D visual storyboards, and simulcams are revolutionizing how films featuring visual effects are produced. Despite their popularity, these technologies have introduced new challenges for actors. An increased workload is faced when digital characters are animated via mocap, since actors are requested to use their imagination to envision what characters see and do on set. This work investigates how Mixed Reality (MR) technology can support actors during mocap sessions by presenting a collaborative MR system named CoMR-MoCap, which allows actors to rehearse scenes by overlaying digital contents onto the real set. Using a Video See-Through Head Mounted Display (VST-HMD), actors can see digital representations of performers in mocap suits and digital scene contents in real time. The system supports collaboration, enabling multiple actors to wear both mocap suits to animate digital characters and VST-HMDs to visualize the digital contents. A user study involving 24 participants compared CoMR-MoCap to the traditional method using physical props and visual cues. The results showed that CoMR-MoCap significantly improved actors’ ability to position themselves and direct their gaze, and it offered advantages in terms of usability, spatial and social presence, embodiment, and perceived effectiveness over the traditional method. • Technologies like motion capture and visual effects are revolutionizing filmmaking. • New technologies have increased actors’ workload when animating via motion capture. • Actors are required to imagine what digital characters see and do on set. • This work presents an MR system designed to assist actors during mocap sessions. • The system lets actors use mocap suits and headsets to animate and view characters.

Social networks are creating a digital world in which the cognitive, emotional, and pragmatic value of the imagery of human faces and bodies is arguably changing. However, researchers in the digital humanities are often ill-equipped to study these phenomena at scale. This work presents FRESCO (Face Representation in E-Societies through Computational Observation), a framework designed to explore the socio-cultural implications of images on social media platforms at scale. FRESCO deconstructs images into numerical and categorical variables using state-of-the-art computer vision techniques, aligning with the principles of visual semiotics. The framework analyzes images across three levels: the plastic level, encompassing fundamental visual features like lines and colors; the figurative level, representing specific entities or concepts; and the enunciation level, which focuses particularly on constructing the point of view of the spectator and observer. These levels are analyzed to discern deeper narrative layers within the imagery. Experimental validation confirms the reliability and utility of FRESCO, and we assess its consistency and precision across two public datasets. Subsequently, we introduce the FRESCO score, a metric derived from the framework’s output that serves as a reliable measure of similarity in image content. • FRESCO applies structural visual semiotics to analyze social media image meaning. • Validated FRESCO via experiments on human-centered datasets. • FRESCO-score computes a semiotic-aligned similarity assessment. • Converts unstructured images into structured data for analytics.

Creating facial animations using 3D computer graphics represents a very laborious and time-consuming task. Among the numberless approaches for animating faces, the use of blendshapes remains the most common solution because of their simplicity and the ability to produce high-quality results. This approach, however, is also characterized by important drawbacks. With the traditional animation suites, to select the blendshapes to be activated animators are generally requested to memorize the mapping between the blendshapes and the influenced mesh vertices; alternatively, they need to adopt a trial-and-error search within the whole library of available blendshapes. Moreover, the level of expressiveness that can be reached may be lower than expected; this is due to the fact that the possibility to apply transformations to vertices different than just linear translations and mechanisms for adding, e.g., exaggerations, are typically not integrated into the same animation environment. To tackle these issues, this article proposes an immersive virtual reality-based interface that leverages sketches for the direct manipulation of blendshapes. Animators can draw both linear and curved strokes, which are used to automatically extract information about the blendshape to be activated and its weight, the trajectories that associated vertices have to follow, as well as the timing of the overall animation. A user study was carried out with the aim of evaluating the proposed approach on several representative animations tasks. Both objective and subjective measurements were collected. Experimental results showed the benefits of the devised interface in terms of task completion time, animation accuracy, and usability.

Cybersickness (CS) represents one of the oldest problems affecting Virtual Reality (VR) technology. In an attempt to resolve or at least limit this form of discomfort, an increasing number of mitigation techniques have been proposed by academic and industrial researchers. However, the validation of such techniques is often carried out without grounding on a common methodology, making the comparison between the various works in the state of the art difficult. To address this issue, the present article proposes a novel testbed for studying CS in immersive VR and, in particular, methods to mitigate it. The testbed consists of four virtual scenarios, which have been designed to elicit CS in a targeted and predictable manner. The scenarios, grounded on available literature, support the extraction of objective metrics about user's performance. The testbed additionally integrates an experimental protocol that employs standard questionnaires as well as measurements typically adopted in state-of-the-art practice to assess levels of CS and other subjective aspects regarding User Experience. The article shows a possible use case of the testbed, concerning the evaluation of a CS mitigation technique that is compared with the absence of mitigation as baseline condition.

Deep neural networks are increasingly used in a wide range of technologies and services, but remain highly susceptible to out-of-distribution (OOD) samples, that is, drawn from a different distribution than the original training set. A common approach to address this issue is to endow deep neural networks with the ability to detect OOD samples. Several benchmarks have been proposed to design and validate OOD detection techniques. However, many of them are based on farOOD samples drawn from very different distributions, and thus lack the complexity needed to capture the nuances of real-world scenarios. In this work, we introduce a comprehensive benchmark for OOD detection, based on ImageNet and Places365, that assigns individual classes as in-distribution or out-of-distribution depending on the semantic similarity with the training set. Several techniques can be used to determine which classes should be considered in-distribution, yielding benchmarks with varying properties. Experimental results on different OOD detection techniques show how their measured efficacy depends on the selected benchmark and how confidence-based techniques may outperform classifier-based ones on near-OOD samples.

This work presents four possible designs for the handshake interaction in a Social VR-like virtual environment in which the user operates using hand-held controllers: a first design based on a graphics user interface (GUI), a second design leveraging a physical button on hand-held controllers, and two designs based on recreating the handshaking gesture by grabbing the other party’s hand and shaking it. The four designs were evaluated and compared through a user study which involved 24 participants, analyzing factors pertaining to embodiment, presence and social presence, usability, and handshake quality of experience. Results indicated that the gesture-based design was preferred, overall.

Upper Limb (UL) Rehabilitation in Multiple Scle- rosis (MS) is an open research field due to the complex interplay between cognitive and physical dysfunctions. Virtual Reality (VR) can face such an issue by enriching physical training with engaging features, including biofeedback strategies to self- regulate autonomic functions according to the visualisation of indices like heart rate variability (HRV). In the present work, HRV biofeedback is introduced in a VR-based exergame (a game designed to promote exercising), tailored to rehabilitation of the dominant upper limb in Persons with MS (PwMS). The exergame is based on a dual-task paradigm, integrating a UL motor rehabilitative task with a breathing task. The aim is to investigate how the design developed for the HRV biofeedback affects engagement and performance during the exergame session. As a preliminary study, sixteen able-bodied subjects are tested in a within-subjects design, to assess the quality of the game features and design, before approaching MS patients. Two conditions are presented, with and without biofeedback. The proposed HRV biofeedback has two possible levels, depending on whether or not the desired respiratory rate of six breaths/min is successfully maintained. It is used to control game elements and change difficulty of the session. The main finding of this study is that biofeedback improves both user performance and experience in healthy subjects. These results underline the great potential of this technique to promote engagement. Thus, they point to fostering the rehabilitative effectiveness of repetitive motor tasks and encouraging adherence to the long- term training. Future studies will encompass fine tuning of the experimental setup and include PwMS to further adjust the game to patients' needs and observe the setup compliance to rehabilitation settings.

Technology is disrupting the way films involving visual effects are produced. Chroma-key, LED walls, motion capture (mocap), 3D visual storyboards, and simulcams are only a few examples of the many changes introduced in the cinema industry over the last years. Although these technologies are getting commonplace, they are presenting new, unexplored challenges to the actors. In particular, when mocap is used to record the actors' movements with the aim of animating digital character models, an increase in the workload can be easily expected for people on stage. In fact, actors have to largely rely on their imagination to understand what the digitally created characters will be actually seeing and feeling. This paper focuses on this specific domain, and aims to demonstrate how Augmented Reality (AR) can be helpful for actors when shooting mocap scenes. To this purpose, we devised a system named AR-MoCap that can be used by actors for rehearsing the scene in AR on the real set before actually shooting it. Through an Optical See-Through Head-Mounted Display (OST-HMD), an actor can see, e.g., the digital characters of other actors wearing mocap suits overlapped in real-time to their bodies. Experimental results showed that, compared to the traditional approach based on physical props and other cues, the devised system can help the actors to position themselves and direct their gaze while shooting the scene, while also improving spatial and social presence, as well as perceived effectiveness.

Instance-level object re-identification is a fundamental computer vision task, with applications from image retrieval to intelligent monitoring and fraud detection. In this work, we propose the novel task of damaged object re-identification, which aims at distinguishing changes in visual appearance due to deformations or missing parts from subtle intra-class variations. To explore this task, we leverage the power of computer-generated imagery to create, in a semi-automatic fashion, high-quality synthetic images of the same bike before and after a damage occurs. The resulting dataset, Bent & Broken Bicycles (BB-Bicycles), contains 39,200 images and 2,800 unique bike instances spanning 20 different bike models. As a baseline for this task, we propose TransReI3D, a multi-task, transformer-based deep network unifying damage detection (framed as a multi-label classification task) with object re-identification. The BBBicycles dataset is available at https://tinyurl.com/37tepf7m

The use of tracking data in the field of sport analytics has increased in the last years as a starting point for in-depth tactical analyses. This work investigates the use of Temporal Convolutional Networks (TCNs), a powerful architecture for sequential data analysis, to extract ball possession information from tracking data. This task is a crucial step for many tactical analyses and is nowadays carried out manually by a human operator in the stadium, which is costly, difficult to implement, and prone to errors. In this work, several classification approaches are explored to classify the game state as dead, ball owned by the home team, or by the away team: as a single-branch, ternary prediction, or as two binary predictions, first detecting whether the game is dead or alive and then which team owns the ball. TCNs are exploited to create independent trajectory embeddings from tracking data of each object; since there is no semantic ordering among the tracked objects, we investigate different permutation-invariant layers to combine the embeddings, namely, an element-wise sum over the embeddings, a self-attention module, and the use of 2D convolutions. Performance evaluation on tracking data from professional soccer games shows that the proposed method outperforms state-of-the-art rule-based methods, achieving 86.2% accuracy in possession estimation (＋7.3% compared to the state of the art) and 89.2% accuracy in dead-alive classification (＋33.2% compared to the state of the art). Extensive ablation studies were conducted to investigate how different input data concur to the final prediction.

Abstract With the development of metaverse(s), industry and academia are searching for the best ways to represent users' avatars in shared virtual environments (VEs), where real‐time communication between users is required. The expressiveness of avatars is crucial for transmitting emotions that are key for social presence and user experience, and are conveyed via verbal and non‐verbal facial and body signals. In this paper, two real‐time modalities for conveying expressions in virtual reality (VR) via realistic, full‐body avatars are compared by means of a user study. The first modality uses dedicated hardware (i.e., eye and facial trackers) to allow a mapping between the user's facial expressions/eye movements and the avatar model. The second modality relies on an algorithm that, starting from an audio clip, approximates the facial motion by generating plausible lip and eye movements. The participants were requested to observe, for both the modalities, the avatar of an actor performing six scenes involving as many basic emotions. The evaluation considered mainly social presence and emotion conveyance. Results showed a clear superiority of facial tracking when compared to lip sync in conveying sadness and disgust. The same was less evident for happiness and fear. No differences were observed for anger and surprise.

Semantic image interpretation can vastly benefit from approaches that combine sub-symbolic distributed representation learning with the capability to reason at a higher level of abstraction. Logic Tensor Networks (LTNs) are a class of neuro-symbolic systems based on a differentiable, first-order logic grounded into a deep neural network. LTNs replace the classical concept of training set with a knowledge base of fuzzy logical axioms. By defining a set of differentiable operators to approximate the role of connectives, predicates, functions and quantifiers, a loss function is automatically specified so that LTNs can learn to satisfy the knowledge base. We focus here on the subsumption or isOfClass predicate, which is fundamental to encode most semantic image interpretation tasks. Unlike conventional LTNs, which rely on a separate predicate for each class (e.g., dog, cat), each with its own set of learnable weights, we propose a common isOfClass predicate, whose level of truth is a function of the distance between an object embedding and the corresponding class prototype. The PROTOtypical Logic Tensor Networks (PROTO-LTN) extend the current formulation by grounding abstract concepts as parametrized class prototypes in a high-dimensional embedding space, while reducing the number of parameters required to ground the knowledge base.We show how this architecture can be effectively trained in the few and zero-shot learning scenarios. Experiments on Generalized Zero Shot Learning benchmarks validate the proposed implementation as a competitive alternative to traditional embedding-based approaches. The proposed formulation opens up new opportunities in zero shot learning settings, as the LTN formalism allows to integrate background knowledge in the form of logical axioms to compensate for the lack of labelled examples. PROTO-LTN was implemented in Tensorflow and is available at https://github.com/FrancescoManigrass/PROTO-LTN.git

First responders require adequate training to operate safely in dangerous situations. However, low-fidelity exercises used in courses are an approximation of real scenarios. This is particularly true for firefighters since practice sessions often do not include real fire. A way to overcome these limitations consists in using Virtual Reality (VR) to create experiences where trainees can face detailed simulations of actual risks. The system presented in this work-in-progress paper aims to support practice training on the use of the blower as a firefighting tool and offers two training modes to assist trainees in learning the procedure and assessing their knowledge.

In automated Visual GUI Testing (VGT) for Android devices, the available tools often suffer from low robustness to mobile fragmentation, leading to incorrect results when running the same tests on different devices. To soften these issues, we evaluate two feature matching-based approaches for widget detection in VGT scripts, which use, respectively, the complete full-screen snapshot of the application ( Fullscreen ) and the cropped images of its widgets ( Cropped ) as visual locators to match on emulated devices. Our analysis includes validating the portability of different feature-based visual locators over various apps and devices and evaluating their robustness in terms of cross-device portability and correctly executed interactions. We assessed our results through a comparison with two state-of-the-art tools, EyeAutomate and Sikuli. Despite a limited increase in the computational burden, our Fullscreen approach outperformed state-of-the-art tools in terms of correctly identified locators across a wide range of devices and led to a 30% increase in passing tests. Our work shows that VGT tools’ dependability can be improved by bridging the testing and computer vision communities. This connection enables the design of algorithms targeted to domain-specific needs and thus inherently more usable and robust.

3D sketching in virtual reality (VR) enables users to create 3D virtual objects intuitively and immersively. However, previous studies showed that mid-air drawing may lead to inaccurate sketches. To address this issue, we propose to use one hand as a canvas proxy and the index finger of the other hand as a 3D pen. To this end, we first perform a formative study to compare two-handed interaction with tablet-pen interaction for VR sketching. Based on the findings of this study, we design HandPainter, a VR sketching system which focuses on the direct use of two hands for 3D sketching without requesting any tablet, pen, or VR controller. Our implementation is based on a pair of VR gloves, which provide hand tracking and gesture capture. We devise a set of intuitive gestures to control various functionalities required during 3D sketching, such as canvas panning and drawing positioning. We show the effectiveness of HandPainter by presenting a number of sketching results and discussing the outcomes of a user study-based comparison with mid-air drawing and tablet-based sketching tools.

In recent years, Machine Learning and Deep Learning communities have devoted many efforts to studying ever better models and more efficient training strategies. Nonetheless, the fundamental role played by dataset bias in the final behaviour of the trained models calls for strong and principled methods to collect, structure and curate datasets prior to training. In this paper we provide an overview on the use of causal models to achieve a deeper understanding of the underlying structure beneath datasets and mitigate biases, supported by several real-life use cases from the medical and industrial domains.

In the domain of CBRN (Chemical, Biological, Radiological, Nuclear) hazards, first responders need a high-quality training to avoid fatal errors that can compromise the success of operations. Nevertheless, CBRN exercises are often expensive and require a complex management. Furthermore, for preserving trainees’ safety and for logistic constraints, trials may reproduce just an approximation of the real hazard scenario. In order to cope with these issues, a prototype of an Immersive Virtual Reality (VR) training platform was developed by Politecnico di Torino and LINKS Foundation in cooperation with CBRN experts from the Italian Air Force, particularly from Terzo Stormo – Aeronautica Militare di Villafranca di Verona. The platform aims at allowing CBRN operators to train, alone or in a team, testing their ability to carry out required operational procedures in a digital environment that exhibits the complexity of a real-life situation, but does not expose them to life-threatening dangers.

Interest is raising around eXtended Reality Training Systems (XRTSs), which started to be considered as a credible option to train companies' workforce. Even though there is a growing body of literature on best practices and techniques to be adopted for teaching individuals how to perform a variety of operations (e.g., for assembly and maintenance procedures), there are also training situations which have gone mostly unexplored yet. In this paper, we propose and evaluate three different metaphors to face the key challenges associated with training procedures involving parallax-dependent tasks, i.e., tasks in which the instructor needs to make the trainee reach a target observation point and guide his/her attention towards a given point of interest at the same time. Effects observed through a user study that was run in a testbed environment indicated that metaphors based on 3D avatars and frustum visualization can provide important advantages over video-based techniques.

A common operation performed in Virtual Reality (VR) environments is locomotion. Although real walking can represent a natural and intuitive way to manage displacements in such environments, its use is generally limited by the size of the area tracked by the VR system (typically, the size of a room) or requires expensive technologies to cover particularly extended settings. A number of approaches have been proposed to enable effective explorations in VR, each characterized by different hardware requirements and costs, and capable to provide different levels of usability and performance. However, the lack of a well-defined methodology for assessing and comparing available approaches makes it difficult to identify, among the various alternatives, the best solutions for selected application domains. To deal with this issue, this article introduces a novel evaluation testbed which, by building on the outcomes of many separate works reported in the literature, aims to support a comprehensive analysis of the considered design space. An experimental protocol for collecting objective and subjective measures is proposed, together with a scoring system able to rank locomotion approaches based on a weighted set of requirements. Testbed usage is illustrated in a use case requesting to select the technique to adopt in a given application scenario.

Deep learning has thrived by training on large-scale datasets. However, in many applications, as for medical image diagnosis, getting massive amount of data is still prohibitive due to privacy, lack of acquisition homogeneity and annotation cost. In this scenario, transfer learning from natural image collections is a standard practice that attempts to tackle shape, texture and color discrepancies all at once through pretrained model fine-tuning. In this work, we propose to design a dedicated network module that focuses on color adaptation, thus preprocessing the input into a form (RGB) that is closer to the domain the classification backbone was trained on. We combine learning from scratch of the color module with transfer learning of different classification backbones, obtaining an end-to-end, easy-to-train architecture for diagnostic image recognition on x-ray images. Extensive experiments showed how our approach is particularly efficient in case of data scarcity and provides a new path for further transferring the learned color information across multiple medical datasets.

Although passive haptic interfaces have been shown to be capable to enhance the user's sense of presence in Mixed Reality experiences, their use is still constrained by the need to rely on exact replicas of virtual objects or on custom-made devices mimicking the original ones. Unfortunately, the former are not flexible enough in terms of reconfigurability, whereas the latter may be difficult to reproduce. To tackle these issues, this paper explores the possibility to build passive haptic interfaces using off-the-shelf toy construction bricks. Bricks can be assembled to provide the intended feedback in more than one task. Moreover, they may be reassembled in another application to mimic completely new objects and support totally different tasks.

Virtual character animation is receiving an ever-growing attention by researchers, who proposed already many tools with the aim to improve the effectiveness of the production process. In particular, significant efforts are devoted to create animation systems suited also to non-skilled users, in order to let them benefit from a powerful communication instrument that can improve information sharing in many contexts like product design, education, marketing, etc. Apart from methods based on the traditional Windows-Icons-Menus-Pointer (WIMP) paradigms, solutions devised so far leverage approaches based on motion capture/retargeting (the so-called performance-based approaches), on non-conventional interfaces (voice inputs, sketches, tangible props, etc.), or on natural language processing (NLP) over text descriptions (e.g., to automatically trigger actions from a library). Each approach has its drawbacks, though. Performance-based methods are difficult to use for creating non-ordinary movements (flips, handstands, etc.); natural interfaces are often used for rough posing, but results need to be later refined; automatic techniques still produce poorly realistic animations. To deal with the above limitations, we propose a multimodal animation system that combines performance- and NLP-based methods. The system recognizes natural commands (gestures, voice inputs) issued by the performer, extracts scene data from a text description and creates live animations in which pre-recorded character actions can be blended with performer’s motion to increase naturalness.

Software for computer animation is generally characterized by a steep learning curve, due to the entanglement of both sophisticated techniques and interaction methods required to control 3D geometries. This paper proposes a tool designed to support computer animation production processes by leveraging the affordances offered by articulated tangible user interfaces and motion capture retargeting solutions. To this aim, orientations of an instrumented prop are recorded together with animator's motion in the 3D space and used to quickly pose characters in the virtual environment. High-level functionalities of the animation software are made accessible via a speech interface, thus letting the user control the animation pipeline via voice commands while focusing on his or her hands and body motion. The proposed solution exploits both off-the-shelf hardware components (like the Lego Mindstorms EV3 bricks and the Microsoft Kinect, used for building the tangible device and tracking animator's skeleton) and free open-source software (like the Blender animation tool), thus representing an interesting solution also for beginners approaching the world of digital animation for the first time. Experimental results in different usage scenarios show the benefits offered by the designed interaction strategy with respect to a mouse & keyboard-based interface both for expert and non-expert users.

Due to the continuous advancements made in robot technologies, the development of intuitive and effective user interfaces for human-robot interaction is getting increasingly important. This paper investigates how different types of interfaces can be used for allowing a single operator to remotely control a team of robots endowed with different capabilities. Attention is focused on three user interfaces based on a gamepad, on a mobile device and on hand tracking, respectively. To evaluate pros and cons of the above interfaces, a user study was conducted, in which participants had to combine the capabilities of a rover, a drone and a robotic arm in order to carry out a search and pick task. Based on the experiments, the fastest way to complete the task was to use the mobile device. However, results showed that some of the interfaces could provide better performances for selected robots and associated sub-tasks. It is worth observing that, despite evidences about efficiency, participants rated the gamepad as the preferred interface from the point of view of subjective usability.

This paper presents a framework for automatically generating speech-based interfaces for controlling virtual and augmented reality (AR) applications on wearable devices. Starting from a set of natural language descriptions of application functionalities and a catalog of general-purpose icons, annotated with possible implied meanings, the framework creates both vocabulary and grammar for the speech recognizer, as well as a graphic interface for the target application, where icons are expected to be capable of evoking available commands. To minimize user's cognitive load during interaction, a semantics-based optimization mechanism was used to find the best mapping between icons and functionalities and to expand the set of valid commands. The framework was evaluated by using it with see-through glasses for AR-based maintenance and repair operations. A set of experimental tests were designed to objectively and subjectively assess first-time user experience of the automatically generated interface in relation to that of a fully personalized interface. Moreover, intuitiveness of the automatically generated interface was studied by analyzing the results obtained through trained users on the same interface. Objective measurements (in terms of false positives, false negatives, task completion rate, and average number of attempts for activating functionalities) and subjective measurements (about system response accuracy, likeability, cognitive demand, annoyance, habitability, and speed) reveal that the results obtained by the first-time users and experienced users with the proposed framework's interface are very similar, and their performances are comparable with those of both the considered references.

The players of the digital industry look at network Big Data as an incredible source of revenues, which can allow them to design products, services, and market strategies ever more tailored to users' interests and needs. This is the case of data collected by Web analytics tools, which describe the way users interact with Web contents and where their attention focuses onto during navigation. Given the complexity of information to analyze, existing tools often make use of visualization strategies to represent data aggregated throughout separate sessions and multiple users. In particular, heat maps are often adopted to study the distribution of mouse activity and identify page regions that are more frequently reached during interaction. Unfortunately, since Web contents are accessed via ever more heterogeneous devices, region-based heat maps cannot be exploited anymore to aggregate data concerning user's attention, since the same Web content may move to another page location or exhibit a different aspect depending on the access device used or the user agent setup. This paper presents the design of a visual analytics framework capable to deal with the above limitation by adopting a data collection approach that combines information about regions displayed with information about page structure. This way, the well-known heat map-based visualization can be produced, where interactions can be aggregated on a per-element basis independently of the specific access configuration. Experimental results showed that the framework succeeds in accurately quantifying user's attention and replicating results obtained by manual processing.

The ability to plan a training path fulfilling working needs plays a key role, when it comes to finding the desired job. Nonetheless, in a dynamic scenario characterized by frequent technological innovations and economic changes, as the present one, getting a clear picture of the requirements of the world of work could become difficult, as it would imply performing a deep periodic evaluation of available job offers, and matching them with own previous experience. Intelligent systems able to automatically match résumés with job offers already exist, but they are mostly targeted to recruiters, instead of learners. This work aims at filling this gap, by presenting a system for comparing learners' acquirements - expressed in terms of learning outcomes - with companies requirements. The proposed system relies on semantics for processing natural language texts and exploits bar charts to visualize learning outcomes and their levels in order to quickly depict similarities and differences between résumés and job offers. An evaluation, on 50 volunteers, underlined the added value of the system.

Drone-based delivery of goods could become a reality in the near future, as witnessed by the increasing successful experiences in both research and commercial fields. In this paper, a prototype system exploiting a do it yourself quad copter drone for delivering products is proposed. On the one hand, the hardware choices made in order to limit risks arising from autonomous delivery are presented. On the other hand, a framework for orders placement and shipment is shown. The advantages of a system like the one described in this paper are mainly related to an increased delivery speed, especially in urban contexts with traffic, to the possibility to make deliveries in areas usually difficult to be reached, and to the drone's ability to autonomously carry out consignments. A practical use case, in which the proposed system is used for delivering drugs (an application in which the need to quickly receive the good might be particularly important) is shown. Nevertheless, the proposed prototype could be employed in other contexts, such as take-away deliveries, product shipments, registered mail consignments, etc.

The fast pace of change in high technology product development requires high flexibility and adaptation in product design and technology management. Within this context, appropriate learning environments are planned to shape future Information and Communication Technology staff in such a way they could get able to keep up with new trends and innovations. A constructivist educational model based on Active Learning represents a step towards this vision for staff training. This paper deals with a new organization of a 13 weeks course based on the inverted class model, where students are introduced to the problem-posing domain by a direct involvement of enterprises, which participate specifically to the problem definition and assessment phases. In the case study shown, students simulate a production business unit in all the phases that characterize the product/service development life cycle. A mixed traditional and company-based organization of the course is supposed to improve students competences by proposing solutions leveraging on appropriate tools taken from traditional experiences and also gaining benefits from a learning-by-doing approach that encourages the use of state-of-the-art solutions. A preliminary study is performed to see how expected professional competences are met during the course, with the additional goal of collecting indications that could help to improve the learning framework for future editions of the course.

Students and learners mobility is more and more becoming a praxis. Nonetheless, despite the legislative efforts carried out by the European Commission, devoted to the definition of standards for describing qualifications, the comparison of qualifications contents, for instance in the phases of identification of more suitable training to attend abroad, or during the recognition of prior learning, still requires a considerable amount of efforts and time. This difficulty could be mainly attributed to the fact that qualification contents are usually expressed heterogeneously. In this work, a way to tackle this heterogeneity, by relying on a semantic thesaurus for graphically displaying main similarities and dissimilarities among qualifications is presented. By means of such representation, users could quickly become aware of the main characteristics of a training course, with respect to the others, without having to read all the textual information related to them.

Existing tracking methods designed for interacting with projection-based displays generally require visible artifacts to be introduced in the environment in order to guarantee effective stability and accuracy. For instance, in optical-oriented approaches, either the camera sensor or the reference pattern used for tracking are often located within the user's sight (or interfere with it), thus occluding portions of the scene or altering the perception of the virtual environment. Several ways to tackle these issues have been recently explored. Proposed approaches basically aim at making the presence of tracking references in the virtual space transparent to the user. However, such solutions introduce possibly critical constraints on required hardware or environment configuration. In this work, a novel tracking approach based on imperceptible fiducial markers is proposed. The approach relies on a hiding technique that allows digital images to be embedded in (and retrieved from) a projected scene by exploiting the properties of light polarization and additive color mixing. In particular, the virtual scene is obtained by overlapping the light beams of two projectors and by dealing with markers’ hiding via color compensation. A prototype setup has been deployed, where interaction with a flat surface projection environment has been evaluated in terms of tracking accuracy and artifacts avoidance performance by using a consumer camera equipped with a polarizing filter. Although the performed tests presented in this article represent only a preliminary and a partial evaluation of the proposed approach, they provided encouraging results indicating that the proposed technique could be possibly applied in more complex interaction scenarios still with limited hardware requirements.

Citizens' mobility and employability are receiving ever more attention by the European legislation. Various instruments have been defined to overcome lexical and semantic differences in the descriptions of qualifications, résumés and job profiles. However, the above differences still represent a significant constraint when abilities of non-European people have to be validated either for education and training or occupation purposes. In this work, a web platform that exploits semantic technologies to address such heterogeneity issues is presented. The platform allows migrants to annotate their knowledge, skills and competences in a shared format based on the European tools. The resulting knowledge base is then used to enable the automatic matchmaking of job seekers' abilities with companies' needs. The platform can additionally be used to support students and workers in the identification of their competence gap with respect to a given education or occupation opportunity, so that to personalize their further training.

The modular exponentiation on large numbers is computationally intensive. An effective way for performing this operation consists in using Montgomery exponentiation in the Residue Number System (RNS). This paper presents an algorithmic and architectural study of such exponentiation approach. From the algorithmic point of view, new and state-of-the-art opportunities that come from the reorganization of operations and precomputations are considered. From the architectural perspective, the design opportunities offered by well-known computer arithmetic techniques are studied, with the aim of developing an efficient arithmetic cell architecture. Furthermore, since the use of efficient RNS bases with a low Hamming weight are being considered with ever more interest, four additional cell architectures specifically tailored to these bases are developed and the tradeoff between benefits and drawbacks is carefully explored. An overall comparison among all the considered algorithmic approaches and cell architectures is presented, with the aim of providing the reader with an extensive overview of the Montgomery exponentiation opportunities in RNS.

The hardware implementation of modular exponentiation for very large integers is a well-known topic in digital arithmetic. An effective approach for obtaining parallel and carry-free implementations consists in using the Montgomery exponentiation algorithm and executing the necessary operations in RNS. Two efficient methods for performing the RNS Montgomery exponentiation have been proposed by Kawamura et al. and by Bajard and Imbert. The above approaches mainly differ in the algorithm used for implementing the base extension. This paper presents a modified RNS Montgomery exponentiation algorithm, where several multiplications are moved outside the main execution loop and replaced by an effective pre-processing stage producing a significant saving on the overall delay with respect to state-of-the-art approaches. Since the proposed modification should be applied to both of the above algorithms, two versions are specifically discussed.

A wide variety of remotely controlled mobile robots have been developed in the past. Despite the evolution in robotics functionalities, not much effort has been made to customize control interfaces to meet user preferences and needs. General users can find difficulties in controlling mobile robots by standard interfaces. On the other hand, recent advances in multi-touch devices now allow researchers to design and implement intuitive and user friendly interfaces. This paper presents recent efforts to control mobile platforms by using customizable multi-touch gestures drawn on a commonly available hand-held device. In particular, a live stream video coming from an on board camera allows both to automatically and manually control the platform to accomplish target tracking and following tasks.

Abstract Shared visualization environments represent an effective means to enhance collaborative work in engineering and scientific design tasks. The availability of high‐speed networks allows researchers to work together from geographically distributed locations, and mobile devices are able to carry out ubiquitous 3D visualization tasks through wireless network connections. This paper presents a scalable architecture for the delivery of shared 3D visualization services to heterogeneous terminals ranging from powerful workstations to mobile devices such as PDAs and smart‐phones. The framework design allows both desktop and mobile clients to simultaneously visualize the same model by sharing a common view. Remote‐rendering servers support effective visualization on thin clients, and a load balancing mechanism provides efficient resource usage. Copyright © 2011 John Wiley & Sons, Ltd.

Two's complement multipliers are important for a wide range of applications. In this paper, we present a technique to reduce by one row the maximum height of the partial product array generated by a radix-4 Modified Booth Encoded multiplier, without any increase in the delay of the partial product generation stage. This reduction may allow for a faster compression of the partial product array and regular layouts. This technique is of particular interest in all multiplier designs, but especially in short bit-width two's complement multipliers for high-performance embedded cores. The proposed method is general and can be extended to higher radix encodings, as well as to any size square and m \times n rectangular multipliers. We evaluated the proposed approach by comparison with some other possible solutions; the results based on a rough theoretical analysis and on logic synthesis showed its efficiency in terms of both area and delay.

With the tremendous growth of information available to end users through the Web, search engines come to play ever a more critical role. Nevertheless, because of their general purpose approach, it is always less uncommon that obtained result sets provide a burden of useless pages. Next generation Web architecture, represented by Semantic Web, provides the layered architecture possibly allowing to overcome this limitation. Several search engines have been proposed, which allow to increase information retrieval accuracy by exploiting a key content of Semantic Web resources, that is relations. However, in order to rank results, most of the existing solutions need to work on the whole annotated knowledge base. In this paper we propose a relation-based page rank algorithm to be used in conjunction with Semantic Web search engines that simply relies on information which could be extracted from user query and annotated resource. Relevance is measured as the probability that retrieved resource actually contains those relations whose existence was assumed by the user at the time of query definition.

A radix-2 digit-recurrence algorithm and architecture for the computation of the cube root are presented in this paper. The original recurrence based on the concept of completing the cube is modified to allow an efficient implementation of the algorithm, and the cycle time and area cost of the resulting architecture are estimated as 7.5 times the delay of a full adder and around 9000 $nand2$ cells, respectively, for double-precision computations.

Mobile devices such as Personal Digital Assistants, Tablet PCs, and cellular phones have greatly enhanced user capability to connect to remote resources. Although a large set of applications are now available bridging the gap between desktop and mobile devices, visualization of complex 3D models is still a task hard to accomplish without specialized hardware. This paper proposes a system where a cluster of PCs, equipped with accelerated graphics cards managed by the Chromium software, is able to handle remote visualization sessions based on MPEG video streaming involving complex 3D models. The proposed framework allows mobile devices such as smart phones, Personal Digital Assistants (PDAs), and Tablet PCs to visualize objects consisting of millions of textured polygons and voxels at a frame rate of 30 fps or more depending on hardware resources at the server side and on multimedia capabilities at the client side. The server is able to concurrently manage multiple clients computing a video stream for each one; resolution and quality of each stream is tailored according to screen resolution and bandwidth of the client. The paper investigates in depth issues related to latency time, bit rate and quality of the generated stream, screen resolutions, as well as frames per second displayed.

The aim of this paper is to present the results of the SpaceGRID project, funded by the ESA, at the end of 18 months of activities, focusing on the major outcomes of the analysis and the prototyping activities based on the next generation Grid Services architectures, also as a proof of concept for a new idea of Earth Observation enlarged Ground Segment.