In Augmented Reality (AR), virtual content enhances user experience by providing additional information. However, improperly positioned or designed virtual content can be detrimental to task performance, as it can impair users' ability to accurately interpret real-world information. In this paper we examine two types of task-detrimental virtual content: obstruction attacks, in which virtual content prevents users from seeing real-world objects, and information manipulation attacks, in which virtual content interferes with users' ability to accurately interpret real-world information. We provide a mathematical framework to characterize these attacks and create a custom open-source dataset for attack evaluation. To address these attacks, we introduce ViDDAR (Vision language model-based Task-Detrimental content Detector for Augmented Reality), a comprehensive full-reference system that leverages Vision Language Models (VLMs) and advanced deep learning techniques to monitor and evaluate virtual content in AR environments, employing a user-edge-cloud architecture to balance performance with low latency. To the best of our knowledge, ViDDAR is the first system to employ VLMs for detecting task-detrimental content in AR settings. Our evaluation results demonstrate that ViDDAR effectively understands complex scenes and detects task-detrimental content, achieving up to 92.15% obstruction detection accuracy with a detection latency of 533 ms, and an 82.46% information manipulation content detection accuracy with a latency of 9.62 s.

3D object detection (OD) is a crucial element in scene understanding. However, most existing 3D OD models have been tailored to work with light detection and ranging (LiDAR) and RGB-D point cloud data, leaving their performance on commonly available visual-inertial simultaneous localization and mapping (VI-SLAM) point clouds unexamined. In this paper, we create and release two datasets: VIP500, 4772 VI-SLAM point clouds covering 500 different object and environment configurations, and VIP500-D, an accompanying set of 20 RGB-D point clouds for the object classes and shapes in VIP500. We then use these datasets to quantify the differences between VI-SLAM point clouds and dense RGB-D point clouds, as well as the discrepancies between VI-SLAM point clouds generated with different object and environment characteristics. Finally, we evaluate the performance of three leading OD models on the diverse data in our VIP500 dataset, revealing the promise of OD models trained on VI-SLAM data; we examine the extent to which both object and environment characteristics impact performance, along with the underlying causes.

Inaccurate spatial tracking in extended reality (XR) headsets can cause virtual object jitter, misalignment, and user discomfort, limiting the headsets' potential for immersive content and natural interactions. We develop a modular testbed to evaluate the tracking performance of commercial XR headsets, incorporating system calibration, tracking data acquisition, and result analysis, and allowing the integration of external cameras and IMU sensors for comparison with open-source VI-SLAM algorithms. Using this testbed, we quantitatively assessed spatial tracking accuracy under various user movements and environmental conditions for the latest XR headsets, Apple Vision Pro and Meta Quest 3. The Apple Vision Pro outperformed the Meta Quest 3, reducing relative pose error (RPE) and absolute pose error (APE) by 33.9% and 14.6%, respectively. While both headsets achieved subcentimeter APE in most cases, they exhibited APE exceeding 10 cm in challenging scenarios, highlighting the need for further improvements in reliability and accuracy.

In this work, we introduce SEESys, the first system to provide online pose error estimation for Simultaneous Localization and Mapping (SLAM). Unlike prior offline error estimation approaches, the SEESys framework efficiently collects real-time system features and delivers accurate pose error magnitude estimates with low latency. This enables real-time quality-of-service information for downstream applications. To achieve this goal, we develop a SLAM system run-time status monitor (RTS monitor) that performs feature collection with minimal overhead, along with a multi-modality attention-based Deep SLAM Error Estimator (DeepSEE) for error estimation. We train and evaluate SEESys using both public SLAM benchmarks and a diverse set of synthetic datasets, achieving an RMSE of 0.235 cm of pose error estimation, which is 15.8% lower than the baseline. Additionally, we conduct a case study showcasing SEESys in a real-world scenario, where it is applied to a real-time audio error advisory system for human operators of a SLAM-enabled device. The results demonstrate that SEESys provides error estimates with an average end-to-end latency of 37.3 ms, and the audio error advisory reduces pose tracking error by 25%.

Virtual content instability caused by device pose tracking error remains a prevalent issue in markerless augmented reality (AR), especially on smartphones and tablets. However, when examining environments which will host AR experiences, it is challenging to determine where those instability artifacts will occur; we rarely have access to ground truth pose to measure pose error, and even if pose error is available, traditional visualizations do not connect that data with the real environment, limiting their usefulness. To address these issues we present SiTAR (Situated Trajectory Analysis for Augmented Reality), the first situated trajectory analysis system for AR that incorporates estimates of pose tracking error. We start by developing the first uncertainty-based pose error estimation method for visual-inertial simultaneous localization and mapping (VI-SLAM), which allows us to obtain pose error estimates without ground truth; we achieve an average accuracy of up to 96.1% and an average FI score of up to 0.77 in our evaluations on four VI-SLAM datasets. Next, we present our SiTAR system, implemented for ARCore devices, combining a backend that supplies uncertainty-based pose error estimates with a frontend that generates situated trajectory visualizations. Finally, we evaluate the efficacy of SiTAR in realistic conditions by testing three visualization techniques in an in-the-wild study with 15 users and 13 diverse environments; this study reveals the impact both environment scale and the properties of surfaces present can have on user experience and task performance.

Simultaneous localization and mapping (SLAM) takes in sensor data, e.g., camera frames, and estimates the user's trajectory while creating a map of the surrounding environment. However, existing SLAM evaluation methods are not reference-free, requiring ground-truth trajectories collected from external systems that are infeasible for most scenarios. In this demo, we present Deep SLAM Error Estimator (DeepSEE), a framework that collects features from a standard visual SLAM pipeline as multivariate time series and uses an attention-based neural network to estimate the tracking error at run time. We evaluate DeepSEE in a game engine-based virtual environment, which generates the visual input for DeepSEE and provides the ground-truth trajectory. Demo participants can navigate the virtual environment to create their own trajectories and view the online pose error estimation. This demo showcases how DeepSEE can act as a quality-of-service indicator for downstream applications.

Recent advances in eye tracking have given birth to a new genre of gaze-based context sensing applications, ranging from cognitive load estimation to emotion recognition. To achieve state-of-the-art recognition accuracy, a large-scale, labeled eye movement dataset is needed to train deep learning-based classifiers. However, due to the heterogeneity in human visual behavior, as well as the labor-intensive and privacy-compromising data collection process, datasets for gaze-based activity recognition are scarce and hard to collect. To alleviate the sparse gaze data problem, we present EyeSyn, a novel suite of psychology-inspired generative models that leverages only publicly available images and videos to synthesize a realistic and arbitrarily large eye movement dataset. Taking gaze-based museum activity recognition as a case study, our evaluation demonstrates that EyeSyn can not only replicate the distinct pat-terns in the actual gaze signals that are captured by an eye tracking device, but also simulate the signal diversity that results from dif-ferent measurement setups and subject heterogeneity. Moreover, in the few-shot learning scenario, EyeSyn can be readily incorpo-rated with either transfer learning or meta-learning to achieve 90% accuracy, without the need for a large-scale dataset for training.

The personalization of augmented reality (AR) experiences based on environmental and user context is key to unlocking their full potential. The recent addition of eye tracking to AR headsets provides a convenient method for detecting user context, but complex analysis of raw gaze data is required to detect where a user's attention and thoughts truly lie. In this demo we present Catch My Eye, the first system to incorporate deep neural network (DNN)-based activity recognition from user gaze into a realistic mobile AR app. We develop an edge computing-based architecture to offload context computation from resource-constrained AR devices, and present a working example of content adaptation based on user context, for the scenario of a virtual art gallery. It shows that user activities can be accurately recognized and employed with sufficiently low latency for practical AR applications.

Demand is growing for markerless augmented reality (AR) experiences, but designers of the real-world spaces that host them still have to rely on inexact, qualitative guidelines on the visual environment to try and facilitate accurate pose tracking. Furthermore, the need for visual texture to support markerless AR is often at odds with human aesthetic preferences, and understanding how to balance these competing requirements is challenging due to the siloed nature of the relevant research areas. To address this, we present an integrated design methodology for AR spaces, that incorporates both tracking and human factors into the design process. On the tracking side, we develop the first VI-SLAM evaluation technique that combines the flexibility and control of virtual environments with real inertial data. We use it to perform systematic, quantitative experiments on the effect of visual texture on pose estimation accuracy; through 2000 trials in 20 environments, we reveal the impact of both texture complexity and edge strength. On the human side, we show how virtual reality (VR) can be used to evaluate user satisfaction with environments, and highlight how this can be tailored to AR research and use cases. Finally, we demonstrate our integrated design methodology with a case study on AR museum design, in which we conduct both VI-SLAM evaluations and a VR-based user study of four different museum environments.

Mobile Augmented Reality (AR), which overlays digital content on the real-world scenes surrounding a user, is bringing immersive interactive experiences where the real and virtual worlds are tightly coupled. To enable seamless and precise AR experiences, an image recognition system that can accurately recognize the object in the camera view with low system latency is required. However, due to the pervasiveness and severity of image distortions, an effective and robust image recognition solution for “in the wild” mobile AR is still elusive. In this article, we present CollabAR, an edge-assisted system that provides distortion-tolerant image recognition for mobile AR with imperceptible system latency . CollabAR incorporates both distortion-tolerant and collaborative image recognition modules in its design. The former enables distortion-adaptive image recognition to improve the robustness against image distortions, while the latter exploits the spatial-temporal correlation among mobile AR users to improve recognition accuracy. Moreover, as it is difficult to collect a large-scale image distortion dataset, we propose a Cycle-Consistent Generative Adversarial Network-based data augmentation method to synthesize realistic image distortion. Our evaluation demonstrates that CollabAR achieves over 85% recognition accuracy for “in the wild” images with severe distortions, while reducing the end-to-end system latency to as low as 18.2 ms.

In this work, we present GazeGraph, a system that leverages human gazes as the sensing modality for cognitive context sensing. GazeGraph is a generalized framework that is compatible with different eye trackers and supports various gaze-based sensing applications. It ensures high sensing performance in the presence of heterogeneity of human visual behavior, and enables quick system adaptation to unseen sensing scenarios with few-shot instances. To achieve these capabilities, we introduce the spatial-temporal gaze graphs and the deep learning-based representation learning method to extract powerful and generalized features from the eye movements for context sensing. Furthermore, we develop a few-shot gaze graph learning module that adapts the `learning to learn' concept from meta-learning to enable quick system adaptation in a data-efficient manner. Our evaluation demonstrates that GazeGraph outperforms the existing solutions in recognition accuracy by 45% on average over three datasets. Moreover, in few-shot learning scenarios, GazeGraph outperforms the transfer learning-based approach by 19% to 30%, while reducing the system adaptation time by 80%.