George Papagiannakis

Manos Kamarianakis, Antonis Protopsaltis, Dimitris Angelis et al.

This work presents the introduction of UniSG^GA, a novel integrated scenegraph structure, that to incorporates behavior and geometry data on a 3D scene. It is specifically designed to seamlessly integrate Graph Neural Networks (GNNs) and address the challenges associated with transforming a 3D scenegraph (3D-SG) during generative tasks. To effectively capture and preserve the topological relationships between objects in a simplified way, within the graph representation, we propose UniSG^GA, that seamlessly integrates Geometric Algebra (GA) forms. This novel approach enhances the overall performance and capability of GNNs in handling generative and predictive tasks, opening up new possibilities and aiming to lay the foundation for further exploration and development of graph-based generative AI models that can effectively incorporate behavior data for enhanced scene generation and synthesis.

Ilias Chrysovergis, Manos Kamarianakis, Mike Kentros et al.

We present a Convolutional Neural Network (CNN) suitable to assess unconstrained surgical cuttings, trained on a dataset created with a data augmentation technique.

Prodromos Kolyvakis, Manos Kamarianakis, George Papagiannakis

This paper introduces a novel integration of Large Language Models (LLMs) with Conformal Geometric Algebra (CGA) to revolutionize controllable 3D scene editing, particularly for object repositioning tasks, which traditionally requires intricate manual processes and specialized expertise. These conventional methods typically suffer from reliance on large training datasets or lack a formalized language for precise edits. Utilizing CGA as a robust formal language, our system, Shenlong, precisely models spatial transformations necessary for accurate object repositioning. Leveraging the zero-shot learning capabilities of pre-trained LLMs, Shenlong translates natural language instructions into CGA operations which are then applied to the scene, facilitating exact spatial transformations within 3D scenes without the need for specialized pre-training. Implemented in a realistic simulation environment, Shenlong ensures compatibility with existing graphics pipelines. To accurately assess the impact of CGA, we benchmark against robust Euclidean Space baselines, evaluating both latency and accuracy. Comparative performance evaluations indicate that Shenlong significantly reduces LLM response times by 16% and boosts success rates by 9.6% on average compared to the traditional methods. Notably, Shenlong achieves a 100% perfect success rate in common practical queries, a benchmark where other systems fall short. These advancements underscore Shenlong's potential to democratize 3D scene editing, enhancing accessibility and fostering innovation across sectors such as education, digital entertainment, and virtual reality.

Manos Kamarianakis, Pandelis Sofianos, George Papagiannakis

What symbolic format should an LLM emit for reliable 3D scene editing from natural language, and does algebraic structure help beyond compact syntax? We evaluate Conformal Geometric Algebra (CGA) as a compact symbolic interface against a verbose Euclidean 4$\times$4 matrix baseline and a non-CGA Compact SE3 control in a natural-language 3D editing pipeline with controlled prompting and deterministic geometric execution. Our primary result is compositional fidelity under sequential instruction chains. In a sequence-stress protocol (20 templates, 6 trials each; $\texttt{n=120}$ outputs per method), Simple CGA and Compact SE3 both achieve 100% parse validity, but Simple CGA preserves exact ordered operation chains more reliably (97.5% vs 90.0%, two-proportion $\texttt{p=0.016}$) with lower completion-token cost (112.6 vs 133.6 tokens). This pattern is consistent with algebraic expression form supporting compositional faithfulness beyond compactness alone. A second result is confirmatory in the powered hard semantic suite ($\texttt{n=100}$ per method): compact representations (Simple CGA 45.0%, Compact SE3 42.0%, Shenlong 44.0%) all exceed the Euclidean 4$\times$4 baseline (24.0%). Simple CGA vs Euclidean is +21 pp ($\texttt{p=0.0028}$) and Compact SE3 vs Euclidean is +18 pp ($\texttt{p=0.0103}$), while Simple CGA vs Compact SE3 is statistically close ($\texttt{p=0.7755}$). Separating parse validity from geometric correctness reveals substantial optimization headroom invisible to syntax-only metrics. Overall, compact symbolic interfaces appear to drive reliability-cost gains, with CGA motor composition providing an additional advantage on ordered instruction chains. These findings inform real-time natural-language editing in immersive and interactive 3D environments.

Manos Kamarianakis, Nick Lydatakis, Antonis Protopsaltis et al.

In this work, we present an integrated geometric framework: "deep- cut" that enables for the first time a user to geometrically and algorithmically cut, tear and drill the surface of a skinned model without prior constraints, layered on top of a custom soft body mesh deformation algorithm. Both layered algorithms in this frame- work yield real-time results and are amenable for mobile Virtual Reality, in order to be utilized in a variety of interactive application scenarios. Our framework dramatically improves real-time user experience and task performance in VR, without pre-calculated or artificially designed cuts, tears, drills or surface deformations via predefined rigged animations, which is the current state-of-the-art in mobile VR. Thus our framework improves user experience on one hand, on the other hand saves both time and costs from expensive, manual, labour-intensive design pre-calculation stages.

George Papagiannakis, Walter Greenleaf, Michael Cole et al.

Computational Medical Extended Reality (CMXR), brings together life sciences and neuroscience with mathematics, engineering and computer science. It unifies computational science (scientific computing) with intelligent extended reality and spatial computing for the medical field. It significantly differs from previous "Clinical XR" or "Medical XR" terms, as it is focusing on how to integrate computational methods from neural simulation to computational geometry, computational vision and computer graphics with deep learning models to solve specific hard problems in medicine and neuroscience: from low/no-code/genAI authoring platforms to deep learning XR systems for training, planning, operative navigation, therapy and rehabilitation.

Manos Kamarianakis, Nick Lydatakis, George Papagiannakis

In this work, we propose two algorithms that can be applied in the context of a networked virtual environment to efficiently handle the interpolation of displacement data for hand-based VR HMDs. Our algorithms, based on the use of dual-quaternions and multivectors respectively, impact the network consumption rate and are highly effective in scenarios involving multiple users. We illustrate convincing results in a modern game engine and a medical VR collaborative training scenario.

Evropi Stefanidi, Maria Korozi, Asterios Leonidis et al.

While the body of research focusing on Intelligent Environments (IEs) programming by adults is steadily growing, informed insights about children as programmers of such environments are limited. Previous work already established that young children can learn programming basics. Yet, there is still a need to investigate whether this capability can be transferred in the context of IEs, since encouraging children to participate in the management of their intelligent surroundings can enhance responsibility, independence, and the spirit of cooperation. We performed a user study (N=15) with children aged 7-12, using a block-based, gamified AR spatial coding prototype allowing to manipulate smart artifacts in an Intelligent Living room. Our results validated that children understand and can indeed program IEs. Based on our findings, we contribute preliminary implications regarding the use of specific technologies and paradigms (e.g. AR, trigger-action programming) to inspire future systems that enable children to create enriching experiences in IEs.

Efstratios Geronikolakis, George Papagiannakis

Applications of the Extended Reality (XR) spectrum, a superset of Mixed, Augmented and Virtual Reality, are gaining prominence and can be employed in a variety of areas, such as virtual museums. Examples can be found in the areas of education, cultural heritage, health/treatment, entertainment, marketing, and more. The majority of computer graphics applications nowadays are used to operate only in one of the above realities. The lack of applications across the XR spectrum is a real shortcoming. There are many advantages resulting from this problem's solution. Firstly, releasing an application across the XR spectrum could contribute in discovering its most suitable reality. Moreover, an application could be more immersive within a particular reality, depending on its context. Furthermore, its availability increases to a broader range of users. For instance, if an application is released both in Virtual and Augmented Reality, it is accessible to users that may lack the possession of a VR headset, but not of a mobile AR device. The question that arises at this point, would be "Is it possible for a full s/w application stack to be converted across XR without sacrificing UI/UX in a semi-automatic way?". It may be quite difficult, depending on the architecture and application implementation. Most companies nowadays support only one reality, due to their lack of UI/UX software architecture or resources to support the complete XR spectrum. In this work, we present an "automatic reality transition" in the context of virtual museum applications. We propose a development framework, which will automatically allow this XR transition. This framework transforms any XR project into different realities such as Augmented or Virtual. It also reduces the development time while increasing the XR availability of 3D applications, encouraging developers to release applications across the XR spectrum.

George Papagiannakis, Paul Zikas, Nick Lydatakis et al.

In this work, we present MAGES 3.0, a novel Virtual Reality (VR)-based authoring SDK platform for accelerated surgical training and assessment. The MAGES Software Development Kit (SDK) allows code-free prototyping of any VR psychomotor simulation of medical operations by medical professionals, who urgently need a tool to solve the issue of outdated medical training. Our platform encapsulates the following novel algorithmic techniques: a) collaborative networking layer with Geometric Algebra (GA) interpolation engine b) supervised machine learning analytics module for real-time recommendations and user profiling c) GA deformable cutting and tearing algorithm d) on-the-go configurable soft body simulation for deformable surfaces.

Stéphanie Bertrand, Martha Vassiliadi, Paul Zikas et al.

This chapter will extend its preliminary scope by examining how literary transportation further amplifies presence and affects user response vis-á-vis virtual heritage by focusing on embodiment and aesthetic experience. To do so, it will draw on recent findings emerging from the fields of applied psychology, neuroaesthetics and cognitive literary studies; and consider a case study advancing the use of literary travel narratives in the design of DCH applications for Antiquities - in this case the well-known ancient Greek monument of Acropolis. Subsequently, the chapter will discuss how Literary-based MR Presence shifts public reception from an education-entertainment-touristic consumption paradigm to a response predicated on valuation. It will show that this type of public engagement is more closely aligned both with MR applications' default mode of usership, and with newly emerging conceptions of a user-centered museum (e.g., the Museum 3.0), thus providing a Virtual Museum model expressly suited to cultural heritage.

Efstratios Geronikolakis, Paul Zikas, Steve Kateros et al.

In this work, a new and innovative way of spatial computing that appeared recently in the bibliography called True Augmented Reality (AR), is employed in cultural heritage preservation. This innovation could be adapted by the Virtual Museums of the future to enhance the quality of experience. It emphasises, the fact that a visitor will not be able to tell, at a first glance, if the artefact that he/she is looking at is real or not and it is expected to draw the visitors' interest. True AR is not limited to artefacts but extends even to buildings or life-sized character simulations of statues. It provides the best visual quality possible so that the users will not be able to tell the real objects from the augmented ones. Such applications can be beneficial for future museums, as with True AR, 3D models of various exhibits, monuments, statues, characters and buildings can be reconstructed and presented to the visitors in a realistic and innovative way. We also propose our Virtual Reality Sample application, a True AR playground featuring basic components and tools for generating interactive Virtual Museum applications, alongside a 3D reconstructed character (the priest of Asinou church) facilitating the storyteller of the augmented experience.

Paul Zikas, George Papagiannakis, Nick Lydatakis et al.

Virtual reality (VR) has re-emerged as a low-cost, highly accessible consumer product, and training on simulators is rapidly becoming standard in many industrial sectors. However, the available systems are either focusing on gaming context, featuring limited capabilities or they support only content creation of virtual environments without any rapid prototyping and modification. In this project, we propose a code-free, visual scripting platform to replicate gamified training scenarios through rapid prototyping and VR software design patterns. We implemented and compared two authoring tools: a) visual scripting and b) VR editor for the rapid reconstruction of VR training scenarios. Our visual scripting module is capable to generate training applications utilizing a node-based scripting system whereas the VR editor gives user/developer the ability to customize and populate new VR training scenarios directly from the virtual environment. We also introduce action prototypes, a new software design pattern suitable to replicate behavioral tasks for VR experiences. In addition, we present the training scenegraph architecture as the main model to represent training scenarios on a modular, dynamic and highly adaptive acyclic graph based on a structured educational curriculum. Finally, a user-based evaluation of the proposed solution indicated that users - regardless of their programming expertise - can effectively use the tools to create and modify training scenarios in VR.