Web3D Program – Web3D 2026

Web3D 2026 – Hosted by the College of Science and Engineering at Hamad Bin Khalifa University in a hybrid format (in-person and virtual)

Tue, Oct 13th

Room 01
08:30 – 09:00	Welcome Message
09:00 – 10:00	Keynote
10:00 – 11:30	Paper Session 1: Immersive XR Interaction Insomnia VR: A serious game pilot study exploring the effects of virtual reality on insomnia in individuals with spinal cord injury undergoing inpatient rehabilitation Mariona Baizhi Cos Garcia; Eloy Opisso; Alicia Romero; Elena Hernández; Alejandro Ríos; Eva Monclús; Pere-Pau Vázquez Alcocer Human-Centered Digital Twins: A Bidirectional WebXR-Based Architecture for Immersive Interaction Hayette Hadjar; Zineb Hadjadj; Matthias Hemmje; Abdelkrim Meziane Spatial Hyperlinking for Novices: Portal-Based Navigation of Multi-Room Volumetric Captures Wilhelm Herbrich; Ting Jiang; Aurelia Föllner; Christian Hansen; Danny Schott A Reality Modeling Language to Facilitate the Definition of WebXR-enabled Spatial Computing Experiences Mikel Salazar; Gabor Soros; Alina Kadlubsky
11:30 – 12:00	Coffee Break
12:00 – 13:30	Paper Session 2: Gaussian Splatting and Neural Rendering PanoGS: Foundation-Depth-Guided High-Resolution Gaussian Splatting from Sparse Panoramic Images for Interactive Web3D Exploration Uzair Shah; Giovanni Pintore; Alberto Jaspe Villanueva; Muhammad Tukur; Wala Elsharif; Mahmood Alzubaidi; Enrico Gobbetti; Marco Agus WebSVR: Joint Volume Measurement and Novel View Synthesis in Web-based 3D Environments Raphael Kunert; Rico Richter; Jürgen Döllner BONSAI: Browser-Optimized Neural Gaussian Avatar Streaming Sina Kitz; Simon Katz; Jan Spindler; Zhastay Yeltay; Reza Rastegari; Julian Kaltheuner; Patrick Stotko; Reinhard Klein From Viewing to Playing: Bring-Your-Own-Scene Interaction for 3D Gaussian Splatting on the Web Danci Shen
13:30 – 14:00	Poster Session 1
14:00 – 14:30	Sponsor Talk 1
14:30 – 16:00	Paper Session 3: Digital Twins and Scientific Visualization HGCalVis: Web-Based Importance-Ordered Visualization of CERN’s HGCal Particle Showers Waad F. Alsheri; Matevž Tadel; Alja M. Tadel; Marco Rovere; Markus Hadwiger; Alberto Jaspe-Villanueva From Signals to Geometry: Web-Rendered Digital Twins over the Asset Administration Shell Amaia Pérez; Javier Guerra; María del Puy Carretero; Xabier Oregui; Bruno Simoes Real-time Web-Based Hemodynamic Visualization and Computation for 4D Flow MRI Martin Cartajena; Javier Bóbeda; Karen López-Linares; Nerea Arrarte; Jesús Ruiz-Cabello; Wilmer Lainez-Oyuela; Amaia Perez; Ander Arbelaiz Scalable WebGPU-based Feature-Space Visualization Using the Particle-Magnet Metaphor Jürgen Döllner; Willy Scheibel; Jose Carvajal
16:30 – 17:00	Poster Session 2
17:00 – 18:30	Paper Session 4: AI-Assisted X3D and Web3D Authoring Decentralized Storage, Distribution and Verifiable Retrieval of X3D Spatial Audio Scenes Eftychia Lakka; Chrysoula Tzermia; Don Brutzman; Athanasios G. Malamos; Evangelos K. Markakis x3d_mcp: A Model Context Protocol Server for AI-Assisted X3D Authoring Nikhil Narra; Rick Lentz; Don Brutzman; Nicholas Polys Technē: A Deterministic Craft Layer for Model Context Protocol Servers, Grounded in LLM-Authored X3D Alexander Hoffman

Room 02
10:00 – 11:30	Workshop 1: Beyond Observation: Designing 3D Games for Neurodiversity Assessment and Intervention Presenters: Mohamad Hassan Fadi Hijab, Yasmin Abdelaal, Marwa Qaraqe, Bilikis Banire, Kamran Khawaja, Dena Al-Thani Interactive 3D games and WebXR environments are typically framed as entertainment, yet they also constitute structured, naturalistic settings in which behavior can be observed and measured. This workshop explores how such environments can be designed as tools for assessment and intervention in autistic and other neurodiverse users, moving beyond passive observation toward the systematic quantification of behavior during gameplay. It examines how multimodal gameplay data can characterize attention, engagement, decision-making, and interaction, drawing on signals such as gaze, movement, navigation, object interaction, response timing, task choices, collaboration, and performance. The workshop also considers how behavioral analytics and AI can help translate these signals into interpretable behavioral indicators and potential digital biomarkers. These measures are intended to complement, rather than replace, clinician- and caregiver-reported assessments, and the workshop emphasizes the importance of avoiding deficit-based interpretations. This 90-minute hybrid session features an invited keynote, a showcase of A-Sense projects, research and industry talks, and a structured discussion, and aims to create a community roadmap and foster future collaboration.
11:30 – 12:00	Coffee Break
12:00 – 13:30	Tutorial 1: When Worlds ‘Coalide’: Building Sustainable Digital Dioramas and Virtual Worlds with IIIF 3D Presenter: Ronald Haynes How do we reduce digital divides and creatively engage more people in immersive worlds? If we deliver material on more platforms, including ones which a majority of people can regularly use, the divide is greatly reduced. WebXR-based environments are expanding to enable more people to enjoy engagement with some XR-based environments and virtual worlds, using any of the billions of devices which can run a modern web browser, including smartphones, tablets and laptops, as well as XR glasses and headsets. WebXR-enabled systems such as VRSpace.org, the Virtual Worlds Museum‘s WebXR Exhibit and Arrival.Space, ClassVR’s Mozilla Hubs-powered Eduverse, and HTC’s Viverse demonstrate some of the great growth and potential of a more sustainable approach to broadly accessible virtual environments. IIIF 3D, launched this year, will complement that move toward universality even further, by enabling enhanced information presentation, sharing, preservation, and preservation capabilities. This tutorial will share latest updates and applications for IIIF-enabled 3D data, along with other media in easily sharable scenes, from the IIIF 3D Technical Specification Group (TSG) and the enhanced capabilities provided by the International Image Interoperability Framework (IIIF) Presentation API 4.0 specifications: extensive interoperability with 2D, audio, video, 3D data in shared scenes easier incorporation of materials from key digital collections around the world coherent building of enriched 3D scenes – e.g. digital dioramas, virtual worlds scenes incorporating data combining local, remote, public, or authenticated sources essential management of lights, cameras, object positioning, orientation and scaling initial and customised views, enabling guided tours, exploration, and alternate views wide access to multiple file types and robust 3D viewers ensuring interoperability enriched editable annotations and associated linked metadata enabling portability sustainability through enhanced practical details and simplified portability of assets: solid foundations based on well-supported open web technologies HTTP / HTTPS HTML + CSS + JS (HTML5) JSON-LD (JavaScript Object Notation for Linked Data) W3C Web Annotation Data Model RDF (Resource Description Framework) Three.js, Babylon.js The collaboratively developed standards for IIIF Presentation API 4.0, with enhanced details and data portability provided by the JSON-LD-based manifest file, enable display of 3D resources using IIIF-enabled tools, in standard websites, in apps, and increasingly embedded in various digital asset collection and online systems. The large community actively supporting this IIIF 3D technical development, sharing collection content and development and testing expertise, include the British Library, Courtauld Gallery, Deutsches Museum, Digirati, Institute of Information Science and Technologies (Italian National Research Council – Visual Computing Lab), MorphoSource, Mnemoscene, Sketchfab (via specialists), Smithsonian, Web3D Consortium (X3D standards group), and universities of Basel, Brighton, California (San Diego), Cambridge, Cyprus (University of Technology – Digital Heritage Research Lab), Duke, Edinburgh, Harvard (Library), Oxford, Stanford, Yale. IIIF is also working closely with the Europeana Foundation, Internet Archive, Wikimedia Foundation, all of whom are IIIF 2D and/or Audio/Video adopters, and expected to be IIIF 3D adopters.
13:30 – 14:00	Poster Session 1
14:30 – 16:30	Workshop 2: The Future of Immersive Content: Generation, Modeling, Standards, and Interaction from an Application Perspective Presenters: Anita Havele, Pere Pau Vazquez, Alberto Jaspe Villanueva, Dena Al Thani, Giovanni Pintore, Jens Schneider and Marco Agus Immersive 3D systems are entering a new phase in which content generation, scene modeling, data analysis, interoperability, and interaction are increasingly shaped by generative AI, foundation models, web technologies, and application-specific requirements. However, progress is uneven across domains: methods that produce visually convincing assets may not preserve geometry or semantics; rich representations may be difficult to transmit and render through the web; standards may lag behind emerging neural content; and interaction techniques must adapt to increasingly dynamic, AI-generated environments. This workshop brings together experts representing complementary perspectives on this transition: 3D standards and interoperability, biomedical applications, indoor architecture and design, cultural heritage, data and visual analytics, and human-centered interaction. The format combines a sequence of short position presentations with a moderated round table based on predefined cross-cutting questions. Rather than surveying technologies in isolation, the workshop adopts an application-first perspective and asks which representations, authoring tools, standards, and interaction paradigms will be needed to make future immersive content useful, trustworthy, interoperable, and sustainable.
16:30 – 17:00	Poster Session 2
17:00 – 18:30	Workshop 3: AI with X3D: Use Cases, Generative Pipelines, and the x3d_mcp Framework Presenters: Anita Havele, Nicholas Polys, Don Brutzman and Len Bullard This workshop explores how Artificial Intelligence enhances X3D based spatial computing across medical visualization, avatar LLM interaction, and generative 3D pipelines. We introduce the x3d_mcp framework, demonstrate AI assisted spatial reasoning (shortest path, best path), and evaluate generative X3D methods for both model level and scene level workloads. Participants will gain practical insight into integrating AI with X3D’s expressive semantics and will help define best practices for next generation intelligent 3D web systems. Each presentation and demonstration will also provide recommendations for how new capabilities can be shared for best progress. Topics AI for Scientific visualization and training Spatial reasoning using X3D metadata Avatars interacting with LLMs: Multi Channel Coherence Field (MCCF) x3d_mcp framework for modular AI pipelines Evaluating generative X3D (models vs. scenes, prompts, reasoning) Format Presentations (x3d_mcp framework, Generative X3D, AI Use cases) Demos (medical AI, avatar LLM, MCP modules) Q and A Roundtable discussion and roadmap creation (30 Minutes) Expected Outcomes Shared evaluation criteria for generative X3D Demonstrations of AI-driven X3D workflows Draft best practice guidelines for AI + X3D Recommendations for future X3D specification enhancements

Hall
08:30 – 11:30	Registration
11:30 – 12:00	Coffee Break
13:30 – 14:30	Lunch
16:30 – 17:00	Coffee Break

Wed, Oct 14th

Room 01
08:30 – 09:30	Keynote 2
09:30 – 11:30	Industrial Cases 1: From Terrestrial Laser Scanning to Web3D Digital Twins: Georeferenced Reconstruction of Industrial Facilities Mohammed Walid Karoui CellMap: WebGPU Visualization of Multi-Scale AI-Generated Cell Biology Data Ondrej Strnad, Deng Luo, Venkatesh Nair and Le Song NiceChunk: Compact Deterministic 3D State for Persistent Voxel Worlds on the Web Mars Kiw Agentic 3D: Transforming Industrial Workflows through AI-Driven Digital Twins Johannes Behr, Christian Stein and Hannes Krug
11:30 – 12:00	Coffee Break
12:00 – 14:00	Tutorial 3: Modeling Human Skeletons and Bodies using the Humanoid Animation (HAnim) International Standard Presenters: Don Brutzman, Joe D. Williams and Aaron Bergstrom The Humanoid Animation (HAnim) standard is an abstract architecture defining skeletons, skin, motion animation with anatomically correct models, with current work focused on aligning apparel and clothing, motion capture (MOCAP) markers, and facial expressions. HAnim design principles pay close attention to strict naming and scene-graph structures in order to create validatable interoperable humanoid models. As a tandem capability, the Extensible 3D (X3D) Graphics standard defines networked interoperable interactive animated 3D models that can be composed for anyone on the World Wide Web, also providing multiple implementations that support HAnim model presentation and animation. This tutorial provides an in-depth exploration on creation of the open-source Humanoid Animation X3D Example Archive models demonstrating full anatomical coverage of an interactive animated human skeleton. Careful naming, metadata, scene-graph structure and specification compliance were combined with direct visualization of incremental results to confirm correctness throughout a repeatably consistent (but highly detailed) process. Multiple open-source tools included Sunrize and X_ITE, Castle Model Viewer, and the X3D-Edit authoring tool. Multiple animation techniques are available for HAnim by using validatable key-frame interpolator sets, BVH-style frame sequences in HAnimMotion nodes, and pose snapshots using the new HAnimPose node. The tutorial also includes recent dramatic progress by the open-source RawKee exporter for Maya, which maps the sophisticated capabilities of an industry-leading modeling tool for production of Web-compatible models. Exported models take advantage of advanced Physically Based Rendering (PBR) and advanced texturing capabilities associated with the glTF standard for interoperable presentation as HAnim X3D models. Looking further ahead, interesting current efforts on future capabilities include cross-technology Avatar use cases from the Metaverse Standards Forum (MSF); addition of apparel, footwear, and wearables via an HAnim Coveroid; motion capture (MOCAP) marker standardization for broader availability of MOCAP streams; and repeatable expressions using the Facial Animation Coding System (FACS). Agenda outline (10 minutes) Context: HAnim and X3D humanoid design using named hierarchies (30 minutes) Creating the Bones archive, how HAnim modelers can get started (30 minutes) RawKee exporter for Maya (20 minutes) Emerging HAnim work in progress, final summaries, additional questions
14:00 – 14:30	Sponsor Talk 1
14:30 – 16:00	Paper Session 5: Web3D Systems, Performance and Engineering Explore-and-Scrutinize: A Spatio-Temporal Decoupling Paradigm for Interactive Analysis in Panoramic Video on the Web Zhang Yanxiang; Yi Qiu Interactive Web-Based Visualization for Aircraft Multidisciplinary Design, Analysis, and Exploration Sanjana Anem; Eden Enwiya; Ben Sandbrook; Darshan Sarojini; Nicholas Polys Improve Rendering of Dynamic 3D Meshes in WebVR Environments with QoE-aware Level of Detail Adaptation Nguyen Thi Quynh Ly; Duc V. Nguyen; Truong Thu Huong; Pham Ngoc Nam From Physical Objects to Web-Ready 3D Assets: An Affordable AI-Assisted RGB-D Scanning Pipeline Marwan Humaid; Ahmad Al-Moslemani; Hamad Aldous; Ejmen Al-Ubejdij; Umair Gavankar; Abdullatif Albaseer; Jens Schneider; Marco Agus
16:30 – 17:00	Sponsor Talk 3
17:00 – 18:30	Visit HBKU Lab

Room 02
09:30 – 11:30	Tutorial 2: AI-Augmented Panoramic WebXR: Foundation Models and Generative Pipelines for Immersive Indoor Experiences Presenters: Marco Agus, Giovanni Pintore and Enrico Gobbetti WebXR provides a standards-based path for delivering immersive applications through the browser, but the creation of interactive, spatially coherent, and lightweight content remains a major bottleneck. At the same time, foundation models and generative AI are changing how 3D and panoramic content can be understood, reconstructed, edited, and authored. This tutorial presents a web-first workflow that connects these two developments. It covers the WebXR runtime and interaction stack; web-native scene representations and delivery; foundation models for depth, segmentation, layout, semantics, and vision–language understanding; and structure-guided generative methods for scene completion, restyling, object insertion, and novel-view synthesis. A running indoor-tour example shows how 360° captures can be transformed into structured assets, enriched through generative editing, optimized for network delivery, and explored on desktop, mobile, and head-mounted devices. The tutorial emphasizes practical integration choices, performance and comfort constraints, hybrid client–cloud execution, and the limitations of current AI models when spatial consistency and real-time interaction are required.
11:30 – 12:00	Coffee Break
12:00 – 14:00	Paper Session 6: Standards, Rendering and Spatial Computing Implementing X3DCPPSAIL with AI-Assistance Ryker Ellingworth; Aaron Bergstrom; Don Brutzman Rendering Explicit and Implicit Complex Functions in 4D: A WebGPU-Based Approach Admir Huseini; Art Saiti OGC 3D Tiles implementation for X3DOM Andreas Plesch; Anita Havele LEGS: Laplacian-Enhanced Gaussian Splatting with a Nonlinear Weighted Loss Yongfei Guo; Qizhou Huo; Xuan Sun; Yuanhao Gong
14:30 – 16:00	Standards Sessions

Hall
08:30 – 11:30	Registration
11:30 – 12:00	Coffee Break
13:30 – 14:30	Lunch
16:30 – 17:00	Coffee Break

Thu, Oct 15th

Room 01
08:30 – 09:30	Keynote 3
09:30 – 10:00	Web3D Consortium Townhall
10:00 – 11:30	Panel Session
11:30 – 12:00	Coffee Break
12:00 – 12:30	Fast Forward: Lighting Demos
12:30 – 13:00	Competitions: HANIM, AI and 3D Content Contest
13:00 – 14:00	Awards Ceremony
15:00 – 17:00	Metaverse Workshop

Hall
08:30 – 11:30	Registration
11:30 – 12:00	Coffee Break
14:00 – 15:00	Lunch

Laboratory Manager, College of Science and Engineering, Hamad Bin Khalifa University (HBKU), Qatar

Holds a PhD in Computer Science and Engineering (HBKU, 2024), an MSc in Data Analytics (HBKU, 2021), a BS in Computer Science (AUCE, 2018), and a BS in Mathematics (LIU, 2020).
Research focuses on human–computer interaction, AI and machine learning, and computer vision for autism, including eye-tracking, pose estimation, and object detection in play-based and learning contexts.
Leads and contributes to projects on joint engagement, preference assessment, and inclusive assessment for autistic learners, often combining sensing technologies, VR, and learning analytics.
Has published in venues such as CoDesign, Frontiers in Education, International Journal of Human–Computer Interaction, JMIR, BCS HCI, and LNCS (HCII/INTERACT).
Recipient of the Mada Innovation Award (2022) for the MAAN accessibility-focused application.
Experienced educator and tutor in mathematics and computer science, and teaching assistant for courses on ICT accessibility and statistics for science and engineering.
Actively serves as a reviewer and collaborator on interdisciplinary projects spanning autism, inclusive technology, AI, and education.