This paper introduces an enhanced Point Cloud Quality Assessment (PCQA) metric, termed PointPCA+, as an extension of PointPCA, with a focus on computational simplicity and feature richness. PointPCA+ refines the original PCA-based descriptors by employing Principal Component Analysis (PCA) solely on geometry data; additionally, the texture descriptors are refined through a direct application of the function on YCbCr values, enhancing the efficiency of computation. The metric combines geometry and texture features, capturing local shape and appearance properties, through a learning-based fusion to generate a total quality score. Prior to fusion, a feature selection module is incorporated to identify the most effective features from a proposed super-set. Experimental results demonstrate the high predictive performance of PointPCA+ against subjective ground truth scores obtained from four publicly available datasets. The metric consistently outperforms state-of-the-art solutions, offering valuable insights into the design of similarity measurements and the effectiveness of handcrafted features across various distortion types.@en

Point clouds (PCs) have attracted researchers and developers due to their ability to provide immersive experiences with six degrees of freedom (6DoF). However, there are still several open issues in understanding the Quality of Experience (QoE) and visual attention of end users while experiencing 6DoF volumetric videos. First, encoding and decoding point clouds require a significant amount of both time and computational resources. Second, QoE prediction models for dynamic point clouds in 6DoF have not yet been developed due to the lack of visual quality databases. Third, visual attention in 6DoF is hardly explored, which impedes research into more sophisticated approaches for adaptive streaming of dynamic point clouds. In this work, we provide an open-source Compressed Point cloud dataset with Eye-tracking and Quality assessment in Mixed Reality (ComPEQ - MR). The dataset comprises four compressed dynamic point clouds processed by Moving Picture Experts Group (MPEG) reference tools (i.e., VPCC and GPCC), each with 12 distortion levels. We also conducted subjective tests to assess the quality of the compressed point clouds with different levels of distortion. The rating scores are attached to ComPEQ - MR so that they can be used to develop QoE prediction models in the context of MR environments. Additionally, eye-tracking data for visual saliency is included in this dataset, which is necessary to predict where people look when watching 3D videos in MR experiences. We collected opinion scores and eye-tracking data from 41 participants, resulting in 2132 responses and 164 visual attention maps in total. The dataset is available at https://ftp.itec.aau.at/datasets/ComPEQ-MR/.

@en

Extended Reality (XR) has emerged as a transformative and immersive technology with versatile applications in content creation and consumption. As XR gains popularity, companies eager to adopt it often possess a surface-level understanding, investing significant resources without effectively addressing the genuine needs of end-users. This study explores the current workflows of XR production companies, and the potential of social XR in mitigating challenges throughout the XR production workflow. We present the outcomes of three respective focus group workshops conducted with three XR production companies and their experts (N=17). The results indicate that at every stage of the production, namely pre-production, production, post-production, and post-release, there are communication challenges between producers and clients, as well as different production and post-production specialists. We discuss various aspects of XR concerning the problem and propose novel opportunities offered by social XR to ameliorate those challenges, improving communication and making development more agile.

@en

The Internet of Multisensory, Multimedia and Musical Things (Io3MT) is a new concept that arises from the confluence of several areas of computer science, arts, and humanities, with the objective of grouping in a single place devices and data that explore the five human senses, besides multimedia aspects and music content. In the context of this brave new idea paper, we advance the proposition of a theoretical alignment between the emerging domain in question and the field of Artificial Intelligence (AI). The main goal of this endeavor is to tentatively delineate the inceptive trends and conceivable consequences stemming from the fusion of these domains within the sphere of artistic presentations. Our comprehensive analysis spans a spectrum of dimensions, encompassing the automated generation of multimedia content, the real-time extraction of sensory effects, and post-performance analytical strategies. In this manner, artists are equipped with quantitative metrics that can be employed to enhance future artistic performances. We assert that this cooperative amalgamation has the potential to serve as a conduit for optimizing the creative capabilities of stakeholders.

@en

Immersive technologies like eXtended Reality (XR) are the next step in videoconferencing. In this context, understanding the effect of delay on communication is crucial. This article presents the first study on the impact of delay on collaborative tasks using a realistic Social XR system. Specifically, we design an experiment and evaluate the impact of end-to-end delays of 300, 600, 900, 1,200, and 1,500 ms on the execution of a standardized task involving the collaboration of two remote users that meet in a virtual space and construct block-based shapes. To measure the impact of the delay in this communication scenario, objective and subjective data were collected. As objective data, we measured the time required to execute the tasks and computed conversational characteristics by analyzing the recorded audio signals. As subjective data, a questionnaire was prepared and completed by every user to evaluate different factors such as overall quality, perception of delay, annoyance using the system, level of presence, cybersickness, and other subjective factors associated with social interaction. The results show a clear influence of the delay on the perceived quality and a significant negative effect as the delay increases. Specifically, the results indicate that the acceptable threshold for end-to-end delay should not exceed 900 ms. This article additionally provides guidelines for developing standardized XR tasks for assessing interaction in Social XR environments.@en

The latest social VR technologies have enabled users to attend traditional media and arts performances together while being geographically removed, making such experiences accessible despite budget, distance, and other restrictions. In this work, we aim at improving the way remote performances are shared by designing and evaluating a VR theatre lobby which serves as a space for users to gather, interact, and relive the common experience of watching a virtual opera. We conducted an initial test with experts ($\mathrm{N}=10$, i.e., designers and opera enthusiasts) in pairs using our VR lobby prototype, developed based on the theoretical lobby design concept. A unique aspect of our experience is its highly realistic representation of users in the virtual space. The test results guided refinements to the VR lobby structure and implementation, aiming to improve the user experience and align it more closely with the social VR lobby's intended purpose. With the enhanced prototype, we ran a between-subject controlled study ($\mathrm{N}=40$) to compare the user experience in the social VR lobby between individuals and paired participants. To do so, we designed and validated a questionnaire to measure the user experience in the VR lobby. Results of our mixed-methods analysis, including interviews, questionnaire results, and user behavior, reveal the strength of our social VR lobby in connecting with other users, consuming the opera in a deeper manner, and exploring new possibilities beyond what is common in real life. All supplemental materials are available at https://github.com/cwi-dis/IEEEVR2024-VRLobby.

@en

Point clouds represent one of the prevalent formats for 3D content. Distortions introduced at various stages in the point cloud processing pipeline affect the visual quality, altering their geometric composition, texture information, or both. Understanding and quantifying the impact of the distortion domain on visual quality is vital to driving rate optimization and guiding post-processing steps to improve the quality of experience. In this paper, we propose a multi-task guided multi-modality no reference metric (M3-Unity), which utilizes 4 types of modalities across attributes and dimensionalities to represent point clouds. An attention mechanism establishes inter/intra associations among 3D/2D patches, which can complement each other, yielding local and global features, to fit the highly nonlinear property of the human vision system. A multi-task decoder involving distortion type classification selects the best association among 4 modalities, aiding the regression task and enabling the in-depth analysis of the interplay between geometrical and textural distortions. Furthermore, our framework design and attention strategy enable us to measure the impact of individual attributes and their combinations, providing insights into how these associations contribute particularly in relation to distortion type. Extensive experimental results on 4 datasets consistently outperform the state-of-the-art metrics by a large margin.@en

Advances in Generative Artificial Intelligence (AI) are resulting in AI-generated media output that is (nearly) indistinguishable from human-created content. This can drastically impact users and the media sector, especially given global risks of misinformation. While the currently discussed European AI Act aims at addressing these risks through Article 52's AI transparency obligations, its interpretation and implications remain unclear. In this early work, we adopt a participatory AI approach to derive key questions based on Article 52's disclosure obligations. We ran two workshops with researchers, designers, and engineers across disciplines (N=16), where participants deconstructed Article 52's relevant clauses using the 5W1H framework. We contribute a set of 149 questions clustered into five themes and 18 sub-themes. We believe these can not only help inform future legal developments and interpretations of Article 52, but also provide a starting point for Human-Computer Interaction research to (re-)examine disclosure transparency from a human-centered AI lens.@en

Affective computing has experienced substantial advancements in recognizing emotions through image and facial expression analysis. However, the incorporation of physiological data remains constrained. Emotion recognition with physiological data shows promising results in controlled experiments but lacks generalization to real-world settings. To address this, we present G-REx, a dataset for real-world affective computing. We collected physiological data (photoplethysmography and electrodermal activity) using a wrist-worn device during long-duration movie sessions. Emotion annotations were retrospectively performed on segments with elevated physiological responses. The dataset includes over 31 movie sessions, totaling 380 h+ of data from 190+ subjects. The data were collected in a group setting, which can give further context to emotion recognition systems. Our setup aims to be easily replicable in any real-life scenario, facilitating the collection of large datasets for novel affective computing systems.

@en

As 3D immersive media continues to gain prominence, Point Cloud Quality Assessment (PCQA) is essential for ensuring high-quality user experiences. This paper introduces ViSam-PCQA, a no-reference PCQA metric guided by visual saliency information across three modalities, which facilitates the performance of the quality prediction. Firstly, we project the 3D point cloud to acquire 2D texture, depth, and normal maps. Secondly, we extract the saliency map based on the texture map and refine it with the corresponding depth map. This refined saliency map is used to weight low-level feature maps to highlight perceptually important areas in the texture channel. Thirdly, high-level features from the texture, normal, and depth maps are then processed by a Transformer to capture global and local point cloud representations across the three modalities. Lastly, saliency along with global and local embeddings, are concatenated and processed through a multi-task decoder to derive the final quality scores. Our experiments on the SJTU, WPC, and BASICS datasets show high Spearman rank order correlation coefficients/Pearson linear correlation coefficients of 0.953/0.962, 0.920/0.920 and 0.887/0.936 respectively, demonstrating superior performance compared to current state-of-the-art methods.@en

In recent years, a large variety of online communication tools have emerged, including social Virtual Reality (VR) platforms for interacting in a virtual world with participants being represented as virtual avatars. Given their popularity, an active area of research focuses on improving the user experience in these virtual experiences. To enable experimentation at large scale on online platforms, it is however essential to collect behavioural data (e.g. movements and audio information). In this work, we present a toolchain that enables the running of experiments using a modified version of the social VR platform Mozilla Hubs. Specifically, our toolkit enables collection and tracking or user positions and movements at a central location, enabling fine-grained analysis of user behaviour during a social VR experience.

@en

Emotion recognition systems are typically trained to classify a given psychophysiological state into emotion categories. Current platforms for emotion ground-truth collection show limitations for real-world scenarios of long-duration content (e.g. >10 minutes), namely: 1) Real-time annotation tools are distracting and become exhausting; 2) Perform retrospective annotation of the whole content in bulk (providing highly coarse annotations); or 3) Are used by external experts (depending on the number of annotators and their subjective experience). We explore a novel approach, the EmotiphAI Annotator, that allows undisturbed content visualisation and simplifies the annotation process by using segmentation algorithms that select brief clips for emotional annotation retrospectively. We compare three methods for content segmentation based on physiological data (Electrodermal Activity (EDA), emotion-based), scene (time-based), and random (control) selection. The EmotiphAI Annotator attained a B+ System Usability Scale score and low-average mental workload as per the NASA Task Load Index (40%). The reliability of the self-report was analysed by the inter-rater agreement (STD < 0.75), coherence across time segmentation methods (STD < 0.17), comparison against the state-of-the-art ground truth (STD < 0.7), and correlation to EDA (>0.3 to 0.8), where the EDA-based method obtained the overall best performance.

@en

Within our Distributed and Interactive Systems research group, we focus on affective haptics, where we design and develop systems that can enhance human emotional states through the sense of touch. Such artificial haptic sensations can potentially augment and enhance our mind, body, and (virtual) social connections. In three works—voice communication, news consumption, and virtual embodiment—we explore the effects of enriching media experiences with thermal and vibrotactile affective haptics, and how such stimulation influences our affective perception.@en

Instead of predicting just one emotion for one activity (e.g., video watching), fine-grained emotion recognition enables more temporally precise recognition. Previous works on fine-grained emotion recognition require segment-by-segment, fine-grained emotion labels to train the recognition algorithm. However, experiments to collect these labels are costly and time-consuming compared with only collecting one emotion label after the user watched that stimulus (i.e., the post-stimuli emotion labels). To recognize emotions at a finer granularity level when trained with only post-stimuli labels, we propose an emotion recognition algorithm based on Deep Multiple Instance Learning (EDMIL) using physiological signals. EDMIL recognizes fine-grained valence and arousal (V-A) labels by identifying which instances represent the post-stimuli V-A annotated by users after watching the videos. Instead of fully-supervised training, the instances are weakly-supervised by the post-stimuli labels in the training stage. The V-A of instances are estimated by the instance gains, which indicate the probability of instances to predict the post-stimuli labels. We tested EDMIL on three different datasets, CASE, MERCA and CEAP-360VR, collected in three different environments: desktop, mobile and HMD-based Virtual Reality, respectively. Recognition results validated with the fine-grained V-A self-reports show that for subject-independent 3-class classification (high/neutral/low), EDMIL obtains promising recognition accuracies: 75.63% and 79.73% for V-A on CASE, 70.51% and 67.62% for V-A on MERCA and 65.04% and 67.05% for V-A on CEAP-360VR. Our ablation study shows that all components of EDMIL contribute to both the classification and regression tasks. Our experiments also show that (1) compared with fully-supervised learning, weakly-supervised learning can reduce the problem of overfitting caused by the temporal mismatch between fine-grained annotations and physiological signals, (2) instance segment lengths between 1-2 s result in the highest recognition accuracies and (3) EDMIL performs best if post-stimuli annotations consist of less than 30% or more than 60% of the entire video watching.

@en

Eliciting and capturing drivers' affective responses in a realistic outdoor setting with pedestrians poses a challenge when designing in-vehicle, empathic interfaces. To address this, we designed a controlled, outdoor car driving circuit where drivers (N=27) drove and encountered pedestrian confederates who performed non-verbal positive or non-positive road crossing actions towards them. Our findings reveal that drivers reported higher valence upon observing positive, non-verbal crossing actions, and higher arousal upon observing non-positive crossing actions. Drivers' heart signals (BVP, IBI and BPM), skin conductance and facial expressions (brow lowering, eyelid tightening, nose wrinkling, and lip stretching) all varied significantly when observing positive and non-positive actions. Our car driving study, by drawing on realistic driving conditions, further contributes to the development of in-vehicle empathic interfaces that leverage behavioural and physiological sensing. Through automatic inference of driver affect resulting from pedestrian actions, our work can enable novel empathic interfaces for supporting driver emotion self-regulation.

@en

Emotion plays a key role in the emerging wave of immersive, multi-sensory audience news engagement experiences. Since emotions can be triggered by somatosensory feedback, in this work we explore how augmenting news video watching with haptics can influence affective perceptions of news. Using a mixed-methods approach, we design and evaluate FeelTheNews, a prototype that combines vibrotactile and thermal stimulation (Matching, 70Hz/20° C, 200Hz/40° C) during news video watching. In a within-subjects study (N=20), we investigate the effects of haptic stimulation and video valence on perceived valence, emotion intensity, comfort, and overall haptic experiences. Findings showed: (a) news valence and emotion intensity ratings were not affected by haptics, (b) no stimulation was more comfortable than including stimulation, (c) attention and engagement with the news can override haptic sensations, and (d) users' perceived agency over their reactions is critical to avoid distrust. We contribute cautionary insights for haptic augmentation of the news watching experience.

@en

Capturing drivers’ affective responses given driving context and driver-pedestrian interactions remains a challenge for designing in-vehicle, empathic interfaces. To address this, we conducted two lab-based studies using camera and physiological sensors. Our first study collected participants’ (N = 21) emotion self-reports and physiological signals (including facial temperatures) toward non-verbal, pedestrian crossing videos from the Joint Attention for Autonomous Driving dataset. Our second study increased realism by employing a hybrid driving simulator setup to capture participants’ affective responses (N = 24) toward enacted, non-verbal pedestrian crossing actions. Key findings showed: (a) non-positive actions in videos elicited higher arousal ratings, whereas different in-video pedestrian crossing actions significantly influenced participants’ physiological signals. (b) Non-verbal pedestrian interactions in the hybrid simulator setup significantly influenced participants’ facial expressions, but not their physiological signals. We contribute to the development of in-vehicle empathic interfaces that draw on behavioral and physiological sensing to in-situ infer driver affective responses during non-verbal pedestrian interactions.

@en

The development and widespread adoption of immersive XR applications has led to a renewed interest in representations that are capable of reproducing real-world objects and scenes with high fidelity. Among such representations, point clouds have attracted the interest of industry and academia alike, and new compression solutions have been developed to facilitate their adoption in mainstream applications. To ensure the best quality of experience for the end-user in limited bandwidth scenarios, new full-reference objective quality metrics have been proposed, promoting features designed specifically for point cloud contents. However, the performance of such features to predict the quality of point cloud contents when the reference is not available is largely unexplored. In this paper, we evaluate the performance of features commonly used to model point cloud distortions in a no-reference framework. The obtained features are integrated into a quality value through a support vector regression model. Results demonstrate the potential of full-reference features for no-reference assessment.

@en

Virtual reality (VR) is the experience in a simulated interactive virtual space. It provides synthetic sensory feedback to users' actions that can physically and mentally immerse the users. Social VR is one type of VR system that allows multiple users to join a collaborative virtual environment and communicate with each other, usually by means of visual and audio cues. The virtual space can be a computer-generated 3D scene or a 360° scene captured by an omnidirectional camera. Each user is represented as a computer-generated avatar or using a virtual representation based on live capture. Social VR has shown to be a promising solution to enrich remote communication experiences in the face of increasing pressures to reduce travel and work remotely. In this chapter, we will showcase the design, implementation, and real-world deployment of a series of social VR applications, which are developed for immersive interactive communication and collaboration for multiple domains (e.g., healthcare, cultural heritage). Moreover, we discuss the experimental protocols for evaluating user experiences in this new medium.

@en

Sharing breathing signals has the capacity to provide insights into hidden experiences and enhance interpersonal communication. However, it remains unclear how the modality of breath signals (visual, haptic) is socially interpreted during collaborative tasks. In this mixed-methods study, we design and evaluate BreatheWithMe, a prototype for real-time sharing and receiving of breathing signals through visual, vibrotactile, or visual-vibrotactile modalities. In a within-subjects study (15 pairs), we investigated the effects of modality on breathing synchrony, social presence, and overall user experience. Key findings showed: (a) there were no significant effects of visualization modality on breathing synchrony, only on deliberate music-driven synchronization; (b) visual modality was preferred over vibrotactile feedback, despite no differences across social presence dimensions; (c) BreatheWithMe was perceived to be an insightful window into others, however included data exposure and social acceptability concerns. We contribute insights into the design of multi-modal real-time breathing visualization systems for colocated, collaborative tasks.

@en