arXiv subjects

Georg Regal

Publications and source records attributed to Georg Regal.

4 recordsLinked to original sources

arXiv2026

Ground-Truth Depth in Vision Language Models: Spatial Context Understanding in Conversational AI for XR-Robotic Support in Emergency First Response

Large language models (LLMs) are increasingly used in emergency first response (EFR) applications to support situational awareness (SA) and decision-making, yet most operate on text or 2D imagery and offer little support for core EFR SA competencies like spatial reasoning. We address this gap by evaluating a prototype that fuses robot-mounted depth sensing and YOLO detection with a vision language model (VLM) capable of verbalizing metrically-grounded distances of detected objects (e.g., the chair is 3.02 meters away). In a mixed-reality toxic-smoke scenario, participants estimated distances to a victim and an exit window under three conditions: video-only, depth-agnostic VLM, and depth-augmented VLM. Depth-augmentation improved objective accuracy and stability, e.g., the victim and window distance estimation error dropped, while raising situational awareness without increasing workload. Conversely, depth- agnostic assistance increased workload and slightly worsened accuracy. We contribute to human SA augmentation by demonstrating that metrically grounded, object-centric verbal information supports spatial reasoning in EFR and improves decision-relevant judgments under time pressure.

cs.HC↗

arXiv2023

MED1stMR: Mixed Reality to Enhance Training of Medical First Responder]{MED1stMR: Mixed Reality to Enhance the Training of Medical First Responders for Challenging Contexts

Mass-casualty incidents with a large number of injured persons caused by human-made or by natural disasters are increasing globally. In such situations, medical first responders (MFRs) need to perform diagnosis, basic life support, or other first aid to help stabilize victims and keep them alive to wait for the arrival of further support. Situational awareness and effective coping with acute stressors is essential to enable first responders to take appropriate action that saves lives. Virtual Reality (VR) has been demonstrated in several domains to be a serious alternative, and in some areas also a significant improvement to conventional learning and training. Especially for the challenges in the training of MFRs, it can be highly useful for practicing and learning domains where the context of the training is not easily available. VR training offers controlled, easy-to-create environments that can be created and trained repeatedly under the same conditions. As an advanced alternative to VR, Mixed Reality (MR) environments have the potential to augment current VR training by providing a dynamic simulation of an environment and hands-on practice on injured victims. Building on this interpretation of MR, the main aim of MED1stMR is to develop a new generation of MR training with haptic feedback for enhanced realism. in this workshop paper, we will present the vision of the project and suggest questions for discussion.

cs.CY↗

arXiv2019

POINTS -- Playful objects for inclusive, personalized movement games

Promoting exercise and promoting fun in sport and activity is a common goal of schools. However, children and adolescents do not exercise enough, which can favor a number of chronic illnesses. Exercise and sports often require coordination of visual perception and reaction, which is an additional barrier for visually impaired (blind and partially sighted) people. Due to their highly motivating appeal, games promoting physical activity (exertion games) have become increasingly popular. Although accessible exertion games have been developed, they do not consider the different abilities of players. Especially in team sports player roles that consider individual abilities can foster inclusion. To personalize roles and assign certain abilities to players, wearable technology can play an important role. In this position paper we present ideas how digital objects can be used to design exertion games for visually impaired students and we reflect how wearable technology can be used for personalized player roles.

cs.HC↗

arXiv2016

Using cognitive agent-based simulation for the evaluation of indoor wayfinding systems

This paper presents a novel approach to simulate human wayfinding behaviour incorporating visual cognition into a software agent for a computer aided evaluation of wayfinding systems in large infrastructures. The proposed approach follows the Sense-Plan-Act paradigm comprised of a model for visual attention, navigation behaviour and pedestrian movement. Stochastic features of perception are incorporated to enhance generality and diversity of the developed wayfinding simulation to reflect a variety of behaviours. The validity of the proposed approach was evaluated based on empirical data collected through wayfinding experiments with 20 participants in an immersive virtual reality environment using a life-sized 3D replica of Vienna's new central railway station. The results show that the developed cognitive agent-based simulation provides a further contribution to the simulation of human wayfinding and subsequently a further step to an effective evaluation tool for the planning of wayfinding and signage.

cs.HC↗