An official website of the United States government
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.
more »
« less
Advancing the Understanding and Evaluation of AR-Generated Scenes: When Vision-Language Models Shine and Stumble
Augmented Reality (AR) enhances the real world by integrating virtual content, yet ensuring the quality, usability, and safety of AR experiences presents significant challenges. Could Vision-Language Models (VLMs) offer a solution for the automated evaluation of AR-generated scenes? Could Vision-Language Models (VLMs) offer a solution for the automated evaluation of AR-generated scenes? In this study, we evaluate the capabilities of three state-of-the-art commercial VLMs -- GPT, Gemini, and Claude -- in identifying and describing AR scenes. For this purpose, we use DiverseAR, the first AR dataset specifically designed to assess VLMs' ability to analyze virtual content across a wide range of AR scene complexities. Our findings demonstrate that VLMs are generally capable of perceiving and describing AR scenes, achieving a True Positive Rate (TPR) of up to 93% for perception and 71% for description. While they excel at identifying obvious virtual objects, such as a glowing apple, they struggle when faced with seamlessly integrated content, such as a virtual pot with realistic shadows. Our results highlight both the strengths and the limitations of VLMs in understanding AR scenarios. We identify key factors affecting VLM performance, including virtual content placement, rendering quality, and physical plausibility. This study underscores the potential of VLMs as tools for evaluating the quality of AR experiences.
more »
« less
- PAR ID:
- 10590353
- Publisher / Repository:
- IEEE GenAI-XR Workshop
- Date Published:
- Journal Name:
- Applied Electronics
- ISSN:
- 1803-7232
- Format(s):
- Medium: X
- Location:
- Saint Malo, France
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Modern safety-critical systems increasingly rely on human–robot interaction to reduce disaster risk and support decision-making during emergencies. Vision–Language Models (VLMs) are promising for these settings because they can interpret complex scenes and communicate safety-relevant information, but they still require careful evaluation to ensure reliable safety reasoning. In particular, current evaluations often frame danger recognition as a binary decision (Safe/Unsafe), making it unclear whether a model is identifying true physical hazards or merely reacting to unusual scene elements. We address this limitation by introducing an explicit distinction between hazard and anomaly, and by separately recognizing hazardous and anomalous states. We evaluate several state-of-the-art VLMs across two datasets and multiple prompting strategies to test whether this distinction changes model behavior. Our results show that VLMs frequently misinterpret anomalousness as hazardousness, revealing an over-reliance on contextual irregularity as a proxy for danger. We further show that explicitly separating anomaly from hazard provides a more informative evaluation of VLM safety reasoning and exposes failure modes that binary safety judgments can obscure.more » « less
-
Modern safety-critical systems increasingly rely on human-robot interaction to reduce disaster risk and support decision-making during emergencies. Vision-Language Models (VLMs) are promising for these settings because they can interpret complex scenes and communicate safety-relevant information, but they still require careful evaluation to ensure reliable safety reasoning. In particular, current evaluations often frame danger recognition as a binary decision (Safe/Unsafe), making it unclear whether a model is identifying true physical hazards or merely reacting to unusual scene elements. We address this limitation by introducing an explicit distinction between hazard and anomaly, and by separately recognizing hazardous and anomalous states. We evaluate several state-of-the-art VLMs across two datasets and multiple prompting strategies to test whether this distinction changes model behavior. Our results show that VLMs frequently misinterpret anomalousness as hazardousness, revealing an over-reliance on contextual irregularity as a proxy for danger. We further show that explicitly separating anomaly from hazard provides a more informative evaluation of VLM safety reasoning and exposes failure modes that binary safety judgments can obscure. Our public dataset is available on Roboflow: https://app.roboflow.com/vlm-in-context-anomaly-and-hazard-detection/camera-ready-roman-ds.more » « less
-
In Augmented Reality (AR), improper virtual content placement can obstruct real-world elements, causing confusion and degrading the experience. To address this, we present LOBSTAR (Language model-based OBSTruction detection for Augmented Reality), the first system leveraging a vision language model (VLM) to detect key objects and prevent obstructions in AR. We evaluated LOBSTAR using both real-world and virtual-scene images and developed a mobile app for AR content obstruction detection. Our results demonstrate that LOBSTAR effectively understands scenes and detects obstructive content with well-designed VLM prompts, achieving up to 96% accuracy and a detection latency of 580ms on a mobile app.more » « less
-
Augmented Reality (AR) technology offers the possibility of experiencing virtual images with physical objects and provides high quality hands-on experiences in an engineering lab environment. However, students still need help navigating the educational content in AR environments due to a mismatch problem between computer-generated 3D images and actual physical objects. This limitation could significantly influence their learning processes and workload in AR learning. In addition, a lack of student awareness of their learning process in AR environments could negatively impact their performance improvement. To overcome those challenges, we introduced a virtual instructor in each AR module and asked a metacognitive question to improve students’ metacognitive skills. The results showed that student workload was significantly reduced when a virtual instructor guided students during AR learning. Also, there is a significant correlation between student learning performance and workload when they are overconfident. The outcome of this study will provide knowledge to improve the AR learning environment in higher education settings.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript
- Conference Proceeding:
- The DOI is not currently available.
