ECCCH Tools

Online Conservation Restoration Annotator

Acronym: OCRA

OCRA is a collaborative, web-based tool, primarily designed to help the workflows in conservation-restoration, for managing three-dimensional documentation of CH objects by annotating their high-resolution digital media representation. This platform exploits 3D models to support a 3D-centric documentation workflow where autoptic observations, information and scientific analysis are referenced in a 3D space to locations and areas on the very surface of digital representations of CH objects. Similarly, it is possible to spatially annotate digital 2D+ representations such as multi-layer images (e.g., from multi-spectral surveys) or Reflectance Transformation Images (RTI).
Using advanced multiresolution techniques, and specialized 3D and 2D+ visualization components, it is possible to work on complex, high-resolution digital representations of CH objects, enabling the users to annotate directly on detailed geometries and image layers from the browser, in a collaborative, multi-user environment. Editing function enables the users to trace spatial annotations with multiple primitives (points, polylines, areas), and then connect these annotations into semantic information. This semantic mapping links spatial annotations to the concepts used in the project and ensures interoperability across tools and institutions. The spatial nature of these annotations makes them metric, enabling an easy and accurate extraction of measurements and statistics. Annotations and mappings can be browsed and presented online interactively, exported as data structures to other tools, but also used to generate static, printable documentation output.

Virtual Transcription Laboratory

Acronym: VTL

Virtual Transcription Laboratory (VTL) empowers researchers, cultural heritage professionals, and data providers to unlock and enrich text-based cultural heritage (CH) materials at scale. By combining advanced automated processing with expert human input, it delivers flexible, reusable workflows for text recognition, and semantic enrichment. Whether working with historical newspapers, manuscripts, or ancient tablets, VTL supports the transformation of diverse CH objects into structured, searchable, and meaningful data.

Collection Ingestion Tool

Acronym: CIT

The Collection Ingestion Tool (CIT) is designed to support cultural heritage institutions in managing, enriching, and publishing their collections in an interoperable, standards-based, and future-ready way. CIT empowers GLAMs (Galleries, Libraries, Archives, and Museums), universities, and smaller cultural heritage institutions to catalogue their resources using domain-specific standards, connect them with European-level infrastructures, and foster thematic and semantic reuse of their datasets.
The Cultural Heritage Cloud Knowledge Base and the Heritage Digital Twins (HDT) develops CIT to help users seamlessly organize and share their cultural collections through intuitive tools, AI/NLP-assisted enrichment, and cross-institutional search and discovery. Its modular architecture ensures flexibility, scalability, and alignment with FAIR data principles.

Calibrated artefacts and Standard Assessment Methodology for on-site Multiscale Scan Quality Control

Acronym: 3Dassess

The use of low‑cost calibrated artefacts will enable the evaluation of the quality of 3D point clouds acquired during the digitization of CH objects.
The possibility of selecting artefact’s characteristics regarding the surface properties, such as reflectivity, similar to those of the target objects ensures that the behaviour of the scanning system under comparable conditions can be assessed. This, in turn, makes it possible to estimate the uncertainties present in the scanned object based on the uncertainties observed in the calibrated artefacts.
The development of a standardized methodology will guarantee the correct on‑site deployment of the artefacts and the appropriate evaluation of scan quality across multiscale processes.
Both the artefacts and the methodology will first be developed and validated at laboratory scale by applying established metrological principles, including dimensional calibration, traceability, and local measurement uncertainty analysis, in order to evaluate the effectiveness of the proposed solutions. Subsequently, they will be implemented in CH demonstration sites, where the methodology will be applied to real use cases by CH professionals, thereby demonstrating the practical utility of the developed tools.

Software tool to assess the quality of digitised data: dimensional measurement uncertainty and accuracy

Acronym: 3DCertify

Photogrammetry is among the main 3D scanning techniques used in cultural heritage sector, if not the most used one due to its scalability and relatively low cost compared to other approaches (e.g.: structured light scanner, LiDAR, etc.). It brings the potential to get 3D assets integrating both geometrical data along with RGB colour information into a 3D mesh or point cloud. Achievable geometrical accuracy ranges at 1/1000 relative performance (e.g.: 1mm error at 1m large scenarios).

However, the geometrical accuracy delivered by a 3D scan by photogrammetry largely depends on a wide set of variables, and specially, on the measuring process applied during the scan (i.e.: the spatial arrangement of images around the object, their relative overlap, quality of the images, etc.). As a result, delivered accuracy is strongly dependant on the skills and experience of the photogrammetrist.

Facing this fact, some photogrammetry software offers post-processing tools to somehow control this skill-dependence, so that quality of the delivered 3D mesh could be under control. However, these software tools still provide non-standard ways of expressing 3D quality (e.g.: confidence heatmaps on the mesh), far away from the standard way the science of measurements, metrology, expresses the quality of a measuring result: the measuring uncertainty.

Heritalise project will develop an innovative post-processing tool for photogrammetry, able to compute and express the measuring uncertainty at each vertex of a 3D mesh. This post-processing and certifying processes will be hosted at Tekniker servers, and could be licenced by Tekniker and exploited through a convenient gate at the ECCCH ecosystem. The delivered post-processing software tool will provide a first sound step on photogrammetry, and it will in turn enable the future development of analogue steps for additional 3D scanning techniques, with special emphasis on LiDAR technology.

HERITALISE CH Conservation Module (CH-CM) – built on MHS (Monitoring Heritage System)

The HERITALISE CH Conservation Module (CH‑CM) is a preventive‑conservation decision‑support module that links continuous monitoring data with 3D digital twins to help stakeholders detect risks early, diagnose root causes and plan timely actions. CH‑CM aggregates heterogeneous data streams from heritage‑friendly IoT sensing (environmental and structural parameters, plus safety, energy and usage/visitor variables) and turns them into actionable indicators through dashboards, alerts and periodic reports.
CH‑CM builds on FSMLR’s MHS (Monitoring Heritage System) experience in intelligent monitoring for heritage environments, combining in‑house and commercial sensor networks (ZigBee/IEEE 802.15.4 and LoRaWAN‑ready) with online visualisation. Within HERITALISE, CH‑CM is extended with AI post‑processing and sequential diagnostic algorithms that generate qualitative and quantitative conservation outputs, and with integration to the GeoHBIM Digital Twin Environment via web APIs. By connecting the ‘visible’ 3D representation with the ‘non‑visible’ dynamics captured by sensors, CH‑CM supports evidence‑based maintenance and preventive interventions, reducing uncertainty, resource waste and the likelihood of costly curative restoration.

CT Scan based 3D printing workflow

Acronym: CT2Print CH

CT2Print CH is a specialised workflow that transforms computed tomography (CT) data of cultural heritage (CH) objects into high quality ready to print 3D models (Fig. 1) that can be shared on the cloud. It is designed and aims for non- invasive documentation, research and inclusive access. The workflow enables the capture, reconstruction, and physical replication of both external and internal features of compel heritage artefacts.
The workflow supports the processing of volumetric CT datasets into segmented, watertight meshes suitable for additive - digital manufacturing, preserving geometric accuracy while allowing selective emphasis of hidden structures such as voids, joins repairs, or internal decay. The workflow is adaptable to different object scales and materials. It supports multiple printing strategies (single part, multipart, transparent or sectioned prints) according to curatorial, research, or educational goals.
CT2Print-CH is intended as an ECCCH-compliant workflow based service, linking raw CT data, derived meshes, paradata (e.g. resolution, uncertainty, segmentation decisions), and printable outputs within a FAIR ecosystem. Enabling the hosting, access, sharing and reuse of CT-derived printable assets across institutions, the service reduces handling risks for originals while unlocking new possibilities for study, conservation planning, accessibility, and public engagement.

HBIM-GIS DTE for CH data models integration and visualisation

Acronym: CH DTE

CH DTE is an HBIM-GIS web-based Digital / Memory Twin Environment for CH landscape, buildings, and objects visualisation, management and operation optimisation. The IDP Twin, a part of IDP's cloud-based platform, integrates 3D models at various levels (landscapes, buildings, objects) using the HBIM-GIS methodology. It seamlessly integrates with existing solutions such as Facility Management, Asset Management, and Monitoring Management. The IDP Twin web application, by default, includes Asset Manager, Maintenance Management, and Operations Monitoring. Its by-default goals include the following:
Authentication: Each tool authenticates via IDP services and receives a token for subsequent HTTPS requests. The token contains user information and assigned roles for server-side transaction management.
Asset Data: Tools access IDP services' model and sensor data. Responses are in JSON format. IDP structures information storage based on each tool's requirements, allowing access to historical simulation/process data and related documentation.
Facility Management: This service controls and tracks maintenance throughout an asset's lifecycle.
Sensor & Monitoring – Connection: Part of an IoT infrastructure, this service captures sensor data for client application asset recording and monitoring.
File Management: Tools can access project documentation as needed.

OGC Blocks

OGC Blocks are modular, reusable components that enhance semantic interoperability across data, data models, and APIs. Each Block encapsulates a well‑scoped concept (such as a schema fragment, constraint set, API component, or semantic mapping) along with its documentation, tests, and machine‑readable artifacts. This modular approach allows data models and interfaces to be assembled in a consistent, testable, and interoperable way.
OGC Blocks help people and computers understand data the same way. They don’t just say how a dataset is laid out; they also make clear what each part means. To do this, every field or attribute can point to a shared, authoritative definition (often published through the OGC Definitions Service) so its meaning is precise and easy to look up.
For example, a field named “date” can be ambiguous (creation, observation, or publication?). With OGC Blocks, date is linked to an exact definition in a definitions service, so anyone, human or machine, can see which “date” it is. That clarity reduces guesswork, lets different datasets reuse the same patterns, and makes it much easier to combine and federate data across systems
OGC Blocks are a core piece of OGC RAINBOW, the knowledge‑graph-based approach that links these components, their relationships, and their tests so they can be discovered and reused across projects.

OGC Definitions Service

The OGC Definitions Service is a shared online reference that helps people and systems use the same concepts in the same way. It is part of the OGC RAINBOW initiative, which focuses on making the meaning of data clearer, easier to find, and easier to reuse across different communities and technologies.
In practice, the Definitions Service acts like a trusted dictionary for concepts from different domains. It provides stable web identifiers (links or URLs) for official terms (such as data properties, concepts, or classifications) that anyone can refer to. When a dataset, data model, or API points to one of these definitions, users can immediately see what a term means and how it relates to others, without relying on guesswork or local documentation.
The service is designed to work for both humans and machines. Definitions can be read as simple web pages, but they are also structured so software tools can understand and use them automatically. Over time, related concepts can be linked together, helping users discover connections across domains. By providing a common point of reference, the OGC Definitions Service supports clearer communication, better data reuse, and more reliable data sharing across platforms and projects.

3D Printing of Digitised Cultural Heritage Objects

Acronym: PrintCH

PrintCH is a workflow pipeline enabling the physical realisation of digitised cultural heritage objects through high-quality 3D printing. Building on validated 3D digitisation pipelines (photogrammetry, structured light), the workflow service supports the creation of accurate, durable, and purpose-driven replicas for research, digital repatriation, conservation, education, and public engagement.
It allows curators and researchers to select appropriate printing strategies based on object complexity, scale, and intended use—ranging from tactile replicas and study models to exhibition-grade reproductions. PrintCH supports the documentation of material choices, print parameters, and fidelity considerations as paradata, ensuring transparency and repeatability.
PrintCH promotes the sharing of ready-to-print heritage assets, enabling decentralised, on-demand manufacturing while preserving links to authoritative digital sources. These can be made available through ECCCH repositories and access services. By reducing transport risks and expanding access, PrintCH contributes to sustainable, inclusive, and collaborative cultural heritage practices.

Virtual Museum Ecosystem support for XR (VMEX), previously referred to as Immersive AR/VR/XR experience

The Virtual Museum Eco System enables stakeholders in the heritage value chain to develop immersive and mobile applications, which include augmented reality, virtual reality and XR components. This empowers heritage practitioners to create, collect, and curate digital media into digital applications which facilitate engagement with heritage, enriching commmunities and enhancing cultural tourism.
The VMEX system is made up of services, clients, tools and example applications. Open source and freely available components are combined together to create a holistic system which connects heritage, communities and sustainable developement integrating VR/AR and XR components into applications.
The approach is based on and enhances the Virtual Museum infrastructure developed in EU-LAC-Museums, CINE and CUPIDO projects. It has seen deployments with community museums in Europe, East Africa, Latin America and the Carribean.
The ecosystem contains services supporting archiving of digital media representations of tangible, intangible, natural and cultural, movable and immovable heritage. It connects these using discrete services to create Virtual Museum at Home, Enhanced Museum Exeriences and Museum Without Walls applications.
The application builders supports Virtual Tours, Digital Galleries, Interactive Maps as well as Citizen Science appllications. It builds upon open source technologies such as Dublin Core, Omeka, Leaflet, IIIF utilising 3D, 360 and digital scene media.

Visible and Non-visible Data Fusion for Cultural Heritage

Acronym: VN-FUSION

VN-FUSION is a tool designed for the integrated acquisition, processing, fusion and visualization of visible and non-visible data of Cultural Heritage objects. The high-resolution external geometry of the object is reconstructed using established geomatics methods (e.g., laser scanning, digital photogrammetry, and total station measurements) and provides the essential spatial framework for the correct localization, processing and interpretation of internal non-visible features. Non-visible features are derived from non-invasive geophysical investigations (e.g. 3D ultrasonic tomography and high-resolution radar scans) which enable the analysis of the object’s internal structure and material conditions without destructive intervention. After processing, visible geometric data and internal non-visible features are harmonized into a common data format and integrated within a unified 3D environment. This environment allows users to explore the object from multiple perspectives and to generate cross-sections at different depths within the object, facilitating a comprehensive understanding of both visible and non-visible features. The proposed tool is particularly effective for stone artifacts (e.g. statues) and for structural or decorative elements of historic buildings, providing valuable insights into hidden conditions and supporting informed decision-making for conservation and restoration activities.

Immersive Analytics Tool

Acronym: Interlumo

Interlumo is a powerful suite of online services designed to transform large volumes of interaction data into actionable insights. It analyses interactions generated by remote applications, public and laboratory installations, virtual and physical spaces, web apps, and other digital platforms, enabling a deep understanding of user behaviour in immersive and interactive contexts such as virtual experience, 360° environments, and 3D content.
Delivered through cloud-based components, Interlumo captures and processes high-density interaction data structured according to attributes defined by the originating application or installation. Data can be filtered, aggregated, and explored to reveal meaningful behavioural patterns and trends.
At the core of the suite is an advanced Visual and Immersive Analytics module, accessible online, which allows analysts to investigate interaction data directly within the same digital environment in which it was generated, whether a digital replica of CH, a 3D reconstruction, or a generative space. Analysts can visually explore and annotate user trajectories and behaviours in 3D, identifying key indicators such as attention hotspots and dwell areas.
A collaborative mode further enhances the analytical workflow, enabling remote teams to discuss, interpret, and share insights in real time, directly from a web browser.

agami

Acronym: agami

Based on 5 years of creative R&D by Mezzo Forte team, agami is an audio platform designed to create and deliver location-based immersive sound experiences. It is conceived as a new kind of system focused on open-ear immersive listening, allowing users to remain connected with the natural acoustic environment while experiencing digitally enhanced soundscapes.
agami’s approach is “ecological”: instead of isolating listeners, it blends virtual audio content with the real world, creating a seamless relationship between sound, space, and place.

Online PlaceMUS Geovisualisation module

Acronym: PlaceMUS

The thematic geovisualisation component of the PlaceMUS initiative, GeoViz (R 01.1), developed by CNRS MAP, offers a web interface dedicated to exploring cultural itineraries. It combines a cartographic component (based on web-compliant open solutions) and additional graphics (chronography and abstract visual formalisms).
The methodology has been already developed by CNRS MAP (current TRL4) in the context of a research aimed at promoting the initiatives of citizens in documenting minor heritage sites.
The GeoViz component technically consists of a multi-view web interface used not only to position “places” and to give access to heterogeneous datasets (visual and acoustic) documenting a place’s heritage, but also to enable comparative and analytical reading of similarities and differences between places.
This web component will enhance the representation of temporal aspects at both conceptual and interface levels. Its purpose is to offer academics, cultural actors, and the wide public a synthetic view over places of music and to access multimedia content, in a technical setting thought-of as lightweight, and reusable.

Harmony Space

Acronym: HSP

Harmony Space is a prototype interactive music system, versions of which can be controlled using dance mats, games controllers, mouse and keyboard, guitars or cameras for spatial tracking. Harmony Space allows physically impaired participants to collaborate as first-class citizens in a wide range of complex musical activities, composing, performing, and analysing harmony. At the same time, it can offer valuable analytical and compositional insights to expert musicians.
Harmony Space exploits cognitive theories of music perception to represent and control harmonic sequences in a principled way as pathways in space, allowing all, regardless of musical experience, to explore and manipulate harmonic patterns using universal human capabilities for spatial navigation.
Harmony Space can be played live like an instrument by tracing out paths. These spatial journeys can be recorded live or stepwise. These can typically be rendered in a variety of harmonic contexts for contrasting musical results. Harmony Space can download chord sequences for existing pieces of music from the 20,000 in the ChoCo Chord database, and can import chord sequences from iReal Pro. From these sources, it can create spatial traces, with full harmonic analysis, typically yielding analytical and compositional insights, and opportunities for making new music.

Web3D Presentation Layer

Acronym: PlaceMUS XR Viewer

The Web3D Presentation Layer is an open-source application designed to visualize and interact with rich 3D scenarios directly through the web. Built by integration the ATON (CNR) and MINERVA (3D Research) frameworks, this tool enables end-user to experience complex “Heritage Digital Twins”, including high-resolution 360° panoramas and detailed (multi-resolution) 3D models. Its primary purpose is to lower technological barriers: leveraging modern web standards, APIs (WebXR) and large open ecosystems (like Node.js, Three.js), it is realized as Web3D application accessible via standard browsers on any device (PC, mobile, VR headsets) without any installation required. The tool also integrates gamification strategies to enhance user engagement and learning. It also supports advanced spatialized binaural sound to simulate historical soundscapes and allows for synchronous multi-user collaborative sessions. In the presentation layer, the same technological framework also supports XR On the Move experiences, enabling geo-localized mobile VR applications that extend 3D soundscapes, narratives, and heritage contents into real urban spaces during on-site visits.

Web3D Scene Editor Framework

Acronym: PlaceMUS XR Editor

The Web3D Scene Editor is an open-source framework designed specifically for Cultural Heritage professionals (curators, creatives, researchers) to enable the creation, the modification and reuse of interactive 3D scenarios directly on the web. Developed by integrating the ATON (CNR) and MINERVA (3D Research) platforms, this tool addresses the need to overcome technical complexity in producing “Heritage Digital Twins” or 3D Digital Experience. Unlink simple viewers, this is an authoring tool: it allows users to import 3D Model and 360° panoramas, create semantic annotations and manage advanced spatialised binaural audio. Its purpose is to empower museums and creative industries to produce complex XR experiences (WebXR) without requiring local software installations, ensuring that the content produced is interoperable and ready for the ECCCH ecosystem. Within the editor, the VR Mockup Tool is integrated as an expert design environment, allowing curators to prototype, simulate, and evaluate exhibition layouts, sound strategies, and user experiences in a 3D cybermap before their physical implementation.

Story Map Building and Visualising Tool

Acronym: SMBVT

The Story Map Building and Visualising Tool (SMBVT) is an open-source and web software for creating narratives visualized as story maps, supported by Semantic Web technologies, developed by CNR-ISTI. Narratives created with SMBVT are modelled based on the Narrative Ontology as an extension of the ISO standard CIDOC CRM. A narrative is represented as a network of geospatial and temporal events related by semantic relations. SMBVT was specifically developed to enable users without expertise in Semantic Web technologies, to create narratives in an easy and semi-automatic way.

The Haptic Bracelets

Acronym: HB

The Haptic Bracelets are prototype haptic motion-sensing bracelets that can be worn on all four limbs simultaneously to allow anyone to feel and learn interlocking musical (and mathematical) patterns in their body, by sense of touch alone. These can be controlled by a computer - or live, by someone else wearing another set of bracelets. Feeling multi-limbed rhythms in this way allows them to be experienced differently – for example any two out of the four voices can be compared instantly by simply focusing mentally on the appropriate limbs. With this approach to generating and experiencing rhythms, profoundly deaf people can collaborate in music making with hearing people as first class citizens. The system has the potential to transform access to music-making for disabled and d/Deaf people by removing the technical, sensory and cognitive barriers that currently exclude many from full participation and enable inclusive music engagement across education, community and professional settings. The bracelets are also valuable for non-deaf people to learn how multi-voiced musical rhythm interlock and can be manipulated and performed.

The VR Mockup Tool for designing sound communication in museums – VR Mockup Tool

The VR Mockup Tool is a specialised module of the Web3D Scene Editor, designed specifically for museum curators and exhibition designers. Unlike the generic editor, this tool functions as a predictive simulator: it allows for the creation of a 3D "Cybermap" of the museum (a cognitive representation of the space) to design virtual exhibitions or prefigure real layouts. Its distinctive feature is the focus on Sound Communication Design: it enables curators to position artworks and sound sources in virtual space and simulate how the audio will propagate. It utilises generative Artificial Intelligence to rapidly populate spaces and suggest optimal configurations, supporting museum decision-making processes

XR On The Move

XR On The Move represents the "mobile-first" and geo-localized extension of the Web3D Presentation Layer. It is a suite of WebXR tools and Apps designed to take the museum experience outside physical walls, transforming cities into open-air museums. While the standard Presentation Layer allows for remote exploration, this module is optimised for the on-site experience. It utilises geo-localisation to activate specific content when the user physically approaches a "Place of Music". It natively integrates "Agami" technology (R 01.6) for Sound Augmented Reality (Sound AR): using open-ear bone-conduction headphones, the user can listen to reconstructed historical soundscapes superimposed over the real sounds of the modern city, maintaining sensory contact with the surrounding environment. It also allows users to "open virtual windows" onto lost architecture or inaccessible interiors via their smartphone screens.

Unified No-Code Platform for Interactive Cultural Heritage Experiences

Acronym: UNICHE Platform

UNICHE is an integrated, no-code digital platform designed to support the full lifecycle of interactive cultural heritage experiences, from content creation and exhibition design to deployment, evaluation, and reuse. It provides cultural heritage institutions, creative professionals, and researchers with a single-entry, collaborative environment for designing, producing, delivering, and assessing multimodal experiences across physical, digital, and hybrid contexts.
The platform brings together a coherent suite of interoperable tools, including narrative authoring, spatial exhibition design, XR prototyping, AI-powered content intelligence, experience deployment applications, visitor analytics, and a shared experience model. These components are unified through a common architectural framework, shared data models, and APIs, ensuring seamless workflows across different experience formats (web, mobile, AR, VR, MR, and phygital installations).
UNICHE is explicitly designed to operate as part of the Cultural Heritage Cloud (ECCCH). It enables direct access to ECCCH-hosted cultural assets and metadata, supports FAIR data principles, and ensures that authored experiences and their evaluation outputs become reusable, interoperable digital heritage resources. By lowering technical barriers and supporting collaborative, cross-institutional workflows, UNICHE democratises the creation of high-quality interactive cultural heritage experiences, particularly benefiting small and resource-constrained institutions.

UNICHE Story Maker

Acronym: UNICHE.StoryMaker

UNICHE.StoryMaker is a no-code, AI-enabled authoring tool designed to support cultural heritage professionals in creating rich, interactive digital narratives without requiring technical expertise. Building upon the Narralive Story Maker (developed by ATHENA RC), USM enables domain experts—such as curators, archaeologists, educators, and researchers—to collaboratively design, structure, and produce multimodal storytelling experiences that can be deployed across physical, digital, and hybrid exhibition contexts.
The tool supports both linear and branching narratives through an intuitive, word-processor-like interface, allowing authors to organise content into story parts, scenes, and conditional paths. Multimedia elements—including text, images, audio, video, quizzes, and interactive prompts—can be seamlessly integrated. AI-assisted features support content discovery, enrichment, and adaptation by leveraging ECCCH-hosted cultural assets and metadata.
UNICHE.StoryMaker is conceived as the core authoring component of the UNICHE platform, serving as the narrative backbone for all narrative-based experiences, from mobile storytelling applications to XR and phygital installations.

UNICHE Place Maker

Acronym: UNICHE.PlaceMaker

UNICHE.PlaceMaker is a no-code spatial authoring tool that enables the design of interactive experiences involving physical, virtual, and hybrid spaces. It supports the creation of virtual tours and VR exhibitions by allowing users to associate digital content with spatial elements such as rooms, hotspots, exhibits, and objects.
Built upon the Narralive Virtual Tour Maker, UNICHE.PlaceMaker allows cultural heritage professionals to upload 360° panoramas, images and videos and enrich them with interactive elements. The tool abstracts technical complexity, making immersive experience design accessible to non-experts while maintaining high-quality spatial interaction.

UNICHE XR-360

Acronym: UNICHE XR-360

The UNICHE.XR360 is a no-code authoring and configuration environment that enables cultural heritage professionals to design immersive 360° XR experiences without technical expertise. It functions as a visual “canvas” where creators can assemble, enrich, and structure XR-360 environments by combining panoramic media, spatial elements, narratives, and interaction logic.
Building upon the Leopoly XR-360 Framework, the Creator Canvas allows users to define immersive scenes based on 360° video or panoramic imagery, place interactive spatial elements (3D objects, text, images, audio, and video), and configure interaction modalities such as gaze-based selection and touchless navigation. The tool abstracts XR complexity behind a technology-agnostic interface, supporting a single-entry, collaborative workflow aligned with the UNICHE vision (no-code, black-box, creator-friendly platform).
The XR-360 Creator Canvas is designed to integrate ECCCH-hosted assets and AI-assisted curation services, supporting content discovery, enrichment, and reuse while keeping human creators fully in control.

UNICHE iGuide CMS

Acronym: UNICHE.iGuideCMS

The UNICHE iGuide CMS is a web-based content management platform designed for the creation, organisation, and maintenance of content used by iGuide mobile and web applications. It enables cultural organisations, and content managers to centrally manage rich, structured, and multilingual content related to places, cultural assets, routes, events, and services.
The platform supports the full content lifecycle of a digital guide, from initial data entry and enrichment to publication, update, and reuse across multiple digital channels. Content is organised through a structured data model covering geographic areas, points of interest, routes, thematic categories, multimedia assets, and user roles, ensuring consistency and scalability across different islands, destinations, and applications.

UNICHE MuseotekBox CMS

Acronym: UNICHE.MuseotekBoxCMS

The MuseotekBox CMS is a web-based authoring and configuration platform for designing, managing, and previewing interactive experiences driven by the physical MuseotekBox. It enables cultural institutions, educators, and exhibit designers to map physical RFID-enabled cards or objects to structured digital content displayed on an external screen.
The platform supports the creation of non-linear, interactive narratives where physical actions trigger digital responses such as videos, quizzes, and textual content. Experiences are modelled through a visual, node-based editor that represents the logical flow of the interaction, allowing authors to define chapters, branching logic, and content behaviour without technical expertise.
As part of the UNICHE ecosystem, the MuseotekBox CMS focuses on the orchestration of phygital experiences, linking tangible interaction with digital storytelling in a transparent, reusable, and configurable way.

Collaborative Interactive XR Spaces

Acronym: CoSpace

CoSpace (UNICHE XR Exhibition Prototyper) is a collaborative co-design tool that supports early-stage prototyping and consultation for physical exhibitions. Developed over the Thalia system, CoSpace extends the Thalia collaborative environment to enable geographically distributed teams to jointly explore exhibition concepts and spatial layout ideas using mixed reality technologies.
The tool facilitates participatory design processes, allowing diverse stakeholders to iteratively co-create, evaluate, and refine exhibition concepts in real-time before full-scale implementation.

UNICHE.MediaEditor

Acronym: UNICHE.MediaEditor

UNICHE.MediaEditor is a no-code multimedia editing and processing toolset designed to support the adaptation, enhancement, and preparation of existing digital media assets for cultural heritage applications. The tool focuses on image, video, and audio processing rather than the creation of new media content, enabling cultural heritage professionals to efficiently refine assets for use in exhibitions, storytelling, and immersive experiences.
UNICHE.MediaEditor integrates and adapts mature open-source and free media processing tools, providing a unified and user-friendly interaction layer. The tool also explores conversational interaction paradigms and selective integration of generative AI capabilities to further simplify media editing workflows.

UNICHE Story Player App

Acronym: UNICHE.StoryPlayer

The UNICHE.StoryPlayer is a runtime application for delivering interactive, narrative-driven cultural heritage experiences authored with the UNICHE Story Maker. It enables museums and cultural heritage organisations to publish and present linear or branching digital stories enriched with multimedia content, quizzes, and conditional logic, without requiring custom software development.
UNICHE.StoryPlayer functions as a configurable execution layer: stories are dynamically loaded and rendered based on their narrative structure and metadata, ensuring consistency across web and mobile platforms. The app supports both on-site and remote use, making it suitable for museum guides, educational applications, and standalone digital storytelling experiences.

UNICHE Place Viewer

Acronym: UNICHE.PlaceViewer

The UNICHE.PlaceViewer is a photorealistic virtual tour application that enables users to explore museums, monuments, and archaeological sites through high-resolution 360° panoramas enriched with interactive hotspots and embedded narratives. It supports both remote access and on-site hybrid use.
The app integrates spatial navigation with narrative storytelling, allowing users to move freely through virtual environments while accessing contextual information linked to exhibits and locations.

UNICHE XR-360 Viewer

Acronym: UNICHE XR-360 Viewer

The UNICHE XR-360 Viewer is a runtime application that delivers fully immersive XR-360 cultural heritage experiences to visitors through VR devices. It represents the visitor-facing execution layer of XR-360 experiences authored in the UNICHE XR-360.
The app supports immersive navigation within 360° VR video environments enriched with spatially embedded content and narratives. Visitors interact using gaze-based and touchless interaction, removing the need for controllers and lowering barriers to participation. The app is designed for museum installations, open-air heritage sites, and portable exhibitions, providing high-quality immersive storytelling while remaining accessible and easy to deploy.
UNICHE XR-360 Viewer is tightly integrated with UNICHE analytics and evaluation services, enabling the collection of interaction data to support evidence-based assessment of visitor engagement.

UNICHE iGuide App

Acronym: UNICHE.iGuideApp

The UNICHE iGuide App is a visitor-facing digital guide designed to support location-based cultural heritage experiences across indoor and outdoor environments. It enables museums, archaeological sites, historic landscapes, and cultural routes to present curated narratives that connect physical places with rich digital content.
The application consumes content authored in the UNICHE iGuide CMS and delivers it through intuitive interfaces optimised for on-site and remote use. It supports different modes of engagement, including points of interest, thematic routes, multimedia narratives and Augmented Reality displays, that can be experienced both within exhibition spaces and across open cultural landscapes. Users can also discover events within integrated calendars.
The iGuide App demonstrates how centrally managed cultural content can be activated at scale through lightweight, reusable, and adaptable visitor applications.

UNICHE Museotek Box

Acronym: UNICHE Museotek Box

The UNICHE Museotek Box is a phygital application combining a physical interactive device (MuseotekBox) with digital cultural content. Interaction is driven through tangible objects equipped with RFID tags, enabling inclusive, playful, and educational experiences.

UNICHE.Repository

Acronym: UNICHE.Repository

The UNICHE.Repository is the core content management and interoperability layer of the UNICHE platform, designed as a federated repository and interoperability gateway rather than a standalone data silo. It provides unified access to locally managed cultural heritage assets and to external content hosted in the Cultural Heritage Cloud (ECCCH) and other European data spaces, without duplicating authoritative records.
UNICHE.Repository functions both as a repository for assets created or curated within UNICHE workflows, while externally hosted assets (e.g. from ECCCH-connected repositories) remain at their source and are accessed through persistent identifiers, metadata links, and APIs..
By abstracting the complexity of heterogeneous repositories, UNICHE.Repository allows creators to work with cultural content seamlessly, regardless of its physical location, while ensuring alignment with FAIR data principles and long-term sustainability.

UNICHE Academy

Acronym: UNICHE Academy

The UNICHE Academy is the training, capacity-building, and knowledge-transfer component of the UNICHE ecosystem. It is designed to support cultural heritage professionals, educators, students, and creative practitioners in developing the skills required to design, deploy, and evaluate interactive cultural heritage experiences using the UNICHE platform and ECCCH-aligned tools.
UNICHE Academy provides structured learning resources, tutorials, use-case–driven guidance, and best-practice frameworks covering no-code authoring, XR and 360° experiences. Rather than functioning as a generic e-learning platform, UNICHE Academy is tightly coupled with the UNICHE tools themselves, enabling learning-through-doing workflows where users move seamlessly between training materials and hands-on experimentation.
The Academy also acts as a dissemination and onboarding mechanism for the ECCCH ecosystem, helping users understand how to access, reuse, and contribute cultural heritage data and experiences in a FAIR and interoperable manner.

Visitor Insights & Behaviour Evaluation

Acronym: UNICHE.VIBE

UNICHE.VIBE (Visitor Insights & Behaviour Evaluation) is an AI-powered analytics tool designed to collect, analyze, and interpret visitor behaviour data generated across UNICHE digital interaction applications. It provides near real-time insights into how visitors interact with exhibitions, narratives, and XR experiences, supporting evidence-based decision-making and continuous experience improvement.
The tool aggregates behavioural signals originating from UNICHE applications. By analysing interaction patterns, preferences, and engagement levels—combined with demographic and preference data derived from behavioural models—UNICHE.VIBE constructs comprehensive user and group profiles. These profiles enable personalised content delivery, adaptive application behaviour, and strategic exhibition optimisation.

UNICHE AI Content Intelligence & Recommender

Acronym: UNICHE.AI

UNICHE.AI integrates advanced AI pipelines for content aggregation, analysis, enrichment, search, and personalised recommendation. It supports both creators and visitors by enabling intelligent discovery of cultural assets and adaptive storytelling.

Intelligent Conversational Agent

Acronym: UNICHE.ICAgent

UNICHE.ICAgent is an AI-based conversational agent designed to enhance visitor engagement, interaction, and guidance across UNICHE digital applications. It provides personalised assistance, responds to visitor queries, and delivers targeted content recommendations related to cultural heritage experiences.
The chatbot leverages advanced Large Language Models (LLMs) combined with Retrieval-Augmented Generation (RAG) and agentic AI techniques. This allows it to dynamically access curated repositories (texts, images, videos) and provide context-aware, accurate, and up-to-date responses aligned with the specific goals of UNICHE applications.
UNICHE.ICAgent integrates seamlessly with other UNICHE components, including visitor profiling, AI-based curation, and visitor behaviour analytics, enabling adaptive, personalised, and continuously improving conversational experiences.

Advanced toolkits for interdisciplinary and enhanced study, conservation, and restoration in burial excavations and findings

Acronym: EXCALIBUR Platform

The EXCALIBUR Platform is a holistic, human-centred framework designed to transform the study, conservation, and restoration of burial cultural heritage (CH). It integrates 12 innovative digital toolkits into a centralised, web-based interface. The platform focuses on four specific levels of heritage objects: (1) burial sites and surroundings, (2) tomb structures and decoration, (3) artefacts and findings, and (4) human remains.
By leveraging Digital Twins (DTs), Artificial Intelligence (AI), and Extended Reality (XR), EXCALIBUR allows professionals to visualise invisible characteristics of objects (e.g., inside funeral urns) through combined representations of 3D surfaces and CT scans. The platform supports the entire lifecycle of heritage management, from remote risk monitoring using Earth Observation data to the digital restoration of damaged artefacts using AI "inpainting" algorithms. It is built to be fully interoperable with the European Collaborative Cloud for Cultural Heritage (ECCCH), ensuring that advanced, cost-affordable tools are accessible to museums, researchers, and small institutions.

European Cultural Heritage Optimized Linked Open Tools

Acronym: ECHOLOT

ECHOLOT supports the adoption of the Cultural Heritage Cloud (ECCCH) by making the creation, reuse and dissemination of high-quality, semantically rich and interoperable cultural heritage data accessible to scholars and institutions of any size. The new end-to-end solution relies on knowledge graphs and open source software (notably MediaWiki suite) for the application of the FAIR data principles and further integration of data within European infrastructures – Europeana, DS4CH, EOSC – as well as collaborative public knowledge bases, such as Wikidata and Wikimedia Commons. Using a combination of AI-, knowledge-, and rule-based approaches, ECHOLOT delivers hybrid curation workflows that efficiently combine automated processing and human input to perform multilingual and multimedia annotation, description generation, and entity recognition. The service features user-friendly interfaces and mechanisms for the detailed capture and embedding of metadata to ensure correct attribution and management of value and ownership in collaborative data generation processes. To ensure sustainability and impact, ECHOLOT includes business model innovation, training, and community engagement, as well as close collaboration with ECCCH stakeholders.

INFINITY

Acronym: INFINITY

The INFINITY ecosystem is an integrated suite of services designed to transform fragmented cultural data into Multidimensional Cultural Heritage Digital Objects (CHDOs). Unlike traditional systems that treat digital files as isolated assets, INFINITY provides a semantic layer that captures the "biography" of an object, including its social, historical, and legal contexts. The project focuses on interoperability: instead of releasing new Collection Management Systems (CMS), INFINITY develops Plug-and-Play software components (APIs and Plugins) for existing platforms like CollectiveAccess, Omeka S, and Arches. These tools allow institutions to bridge the "Provenance Gap" and securely ingest data into the ECCCH Cloud, ensuring it is enriched with ethical validation and ready for advanced AI applications.

KINETIKA STEALTH

Acronym: KINETIKA STEALTH

KINETIKA STEALTH is an IoT-based structural health monitoring system designed for movable and mechanically driven cultural heritage assets, such as bell towers, industrial looms, and movable bridges. The tool enables continuous, non-invasive monitoring of structural behaviour, detecting early signs of wear, deformation, or stress in real time. By integrating sensor networks with advanced data analytics, KINETIKA STEALTH provides heritage professionals with actionable insights to plan preventive maintenance and ensure long-term preservation of complex mechanical structures.

KINETIKA SimPro

Acronym: KINETIKA SimPro

KINETIKA SimPro is a predictive simulation tool for movable and mechanically driven cultural heritage assets, such as industrial looms, bell towers, and movable bridges. The tool combines Finite Element Analysis (FEA), Finite Modelling (FEM), and AI-based analytics to forecast structural behaviour, degradation, and potential failure scenarios. KINETIKA SimPro enables heritage professionals to plan preventive interventions, optimize maintenance strategies, and reduce the risk of unexpected structural failures

KINETIKA MechGuard

Acronym: KINETIKA MechGuard

KINETIKA MechGuard is a predictive maintenance tool designed for movable and mechanically driven cultural heritage assets, such as industrial looms, clock mechanisms, and movable bridges. By integrating sensor data, AI analytics, and operational workflows, MechGuard predicts mechanical system failures and optimizes maintenance schedules. This ensures efficient resource allocation, minimizes downtime, and supports the long-term preservation of heritage assets.

KINETIKA Chronographer

Acronym: KINETIKA Chronographer

KINETIKA Chronographer is a digital tool for estimating and analysing the chronological deterioration of cultural heritage assets with movable or mechanical components. By combining high-resolution photogrammetry, temporal image comparison, and AI-based analytics, the tool reconstructs degradation patterns over time, enabling the assessment of material ageing, surface wear, and structural evolution.

KINETIKA Cracksight

Acronym: KINETIKA Cracksight

KINETIKA Cracksight is an advanced diagnostic tool for detecting, analysing, and monitoring fractures and surface degradation in cultural heritage assets with movable or mechanically driven components. By integrating high-resolution 3D scanning, radiometric data, and AI-driven predictive analytics, Cracksight identifies early-stage cracking, material fatigue, and surface deterioration that are often invisible to traditional inspection methods.

KINETIKA VibroSense

Acronym: KINETIKA VibroSense

KINETIKA VibroSense is an advanced vibro-acoustic diagnostic tool designed to assess the dynamic behaviour and operational health of movable and mechanically driven cultural heritage assets. By analysing vibration and acoustic signals captured through non-invasive sensors, VibroSense identifies anomalies, resonance effects, and early signs of mechanical faults that may compromise structural integrity or functionality. The tool combines signal processing techniques with AI-based pattern recognition to support condition-based maintenance and restoration planning.

KINETIKA ObservAR

Acronym: KINETIKA ObservAR

KINETIKA ObservAR is a remote inspection and assisted intervention tool that enables virtual access, assessment, and guided repair of movable and mechanically driven cultural heritage assets. By combining augmented reality (AR), 3D visualisation, and real-time data streaming from digital twins and monitoring systems, ObservAR allows experts to inspect assets remotely and support on-site teams during maintenance and restoration activities. The tool reduces the need for physical access to fragile or hard-to-reach components, while enhancing decision-making through contextualised digital information.

AmalthAI Machine Learning Platform

Acronym: AmalthAI

A user-friendly machine learning platform designed to make advanced AI technologies accessible to non-technical users. The platform allows users to effortlessly upload, annotate and manage their data while also enabling the creation of new, customized datasets. Users can choose from a wide range of machine learning models, including models capable of semantic segmentation, object detection and classification tasks to initiate training with just a few clicks. Upon completion of training, the platform automatically preserves the best-performing model among multiple trials, allowing seamless inference on new data. Every step of the workflow (data management, model selection, training and inference) is executed through an intuitive and responsive user interface, ensuring a seamless and efficient experience. For experienced users, the platform offers also manual control of the basic training parameters such as hyperparameters and allows for individual augmentations selection. Behind the scenes, the platform leverages Kubernetes-based technologies such as Kubeflow to manage the backend infrastructure, ensuring scalability, reliability and optimal resource allocation. With its modular architecture, AmalthAI can accommodate new computer vision tasks, ensuring adaptability to a wide range of application fields.

DegraTwin: Textile Degradation Digital Twin

Acronym: DegraTwin

Τhis component acts as a digital twin of textile artefacts, integrating sensor data, artefact metadata, and 3D object representations to model and predict material degradation under varying environmental conditions such as humidity, temperature, and light exposure. Combining forward and inverse modelling approaches, the twin can both simulate how degradation evolves over time and infer past exposure conditions from observed damage. When real environmental data is missing, the model can rely on predefined baseline scenarios informed by conservation standards. Ongoing work focuses on defining the data structures for simulation, analysing reference degradation curves and identifying practical decision-making scenarios where conservators could benefit from comparative insights. The resulting module is expected to support comparative assessments in what-if scenarios, for example, estimating damage differences under two imperfect environmental conditions which were a key need identified during early stakeholder feedback.

Dynamics and Motion Analysis (DynaMo)

Acronym: DynaMo

The tool provides finite element (FE) solvers for analyzing the mechanical behavior of textiles across different length scales. Following the typical hierarchy in textile modeling, a model for individual yarns on the fiber-scale, a model for small textile patches on the yarn-scale and a model for entire garments on the macro-scale are developed. Inputs are topology and material description of the textile sample, e.g. weaving distances and material choice. Output are deformed states at discrete timesteps under predefined loading scenarios, as well as internal mechanical fields, such as stresses. The purpose of the tool is a realistic depiction of the textile sample’s mechanical behaviour and motion. In addition, it provides insights to sensitivities with respect to material choice, manufacturing methods, etc.

Hub for Entry, Storage, and Textile Information Access (HESTIA)

Acronym: HESTIA

The HESTIA Datalake serves as the central data backbone for TEXTaiLES, designed to unify all incoming data from sensors, drones, robots, and user uploads. It provides a simple API endpoint for all project tools to submit artifacts and their metadata. The architecture ensures robust and flexible data handling: large files are securely stored in MinIO object storage, while all metadata is organized in a Postgres database for detailed querying. A central Kafka message bus reliably notifies other tools of new data, allowing components like analysis or simulation tools to react instantly. This decoupled approach makes the entire system highly scalable, allowing new tools and data to be easily integrated at any time.
The above architecture is also supported by a web application which provides a user interface to manage the database as well as visualise the data collected. Private access can be granted to those who need to interact with the database, while an extension page provides public access to selected artefacts for viewing by the general audience.

INDRA: Environmental Condition Monitoring

Acronym: INDRA

INDRA is the environmental condition monitoring all-in-one system developed as part of the TEXTaiLES toolbox which collects and sends data to a database for several environmental variables such as temperature, humidity and luminance. The application is focused on monitoring environmental conditions in cultural heritage sites

NEPHELE – NEural Photogrammetry for Heritage with Efficient Learning Engines

Acronym: NEPHELE

NEPHELE is a state-of-the-art tool designed to automate the generation of high-fidelity, background-free 3D models from photographic images of cultural heritage objects. By seamlessly integrating SAM2 (Surface-Aligned Mesh) and SUGAR (Surface-Aligned Gaussian Splatting) techniques, NEPHELE aligns surface geometry with advanced neural rendering, resulting in precise, clean 3D meshes suitable for preservation, study, and dissemination purposes. The system’s approach ensures efficient processing and high-quality outputs, addressing challenges in digitising heritage assets with complex shapes or adverse backgrounds. NEPHELE provides an accessible pipeline from raw images to ready-to-use 3D assets, facilitating direct use in online archives, virtual exhibitions, and research workflows.

TexRevAIve: Digital Textile Restoration Assistant

Acronym: TexRevAlve

TexRevAIve is an intelligent digital restoration assistant to support archaeologists, conservators, and the public to explore and visualize potential restorations of CH textile artefacts. It enables users to digitally synthesize images with user guidance through the deployment of Image-to-Image and Text-to-Image generative AI models. It can perform digital synthesis of damaged, faded, or lost areas based on surrounding or external visual context as well as archeologically grounded context provided through text prompts by human experts (e.g. about the suspected material, colour, style, theme, etc.). Interaction with the tool is meant to be user-friendly and will not require technical expertise related to running or training AI models. Such processes will be handled in the background if and when necessary - the users will provide contextual information and instructions, will tweak parameters, and see the synthesized images. The tool records user interactions (e.g., selected parameters, input prompts, and images, etc.), into session-based data in the background, allowing full tracing of the digital restoration data throughout a user’s session and late retrieval for continuation or inspection.

THOTH 3D Viewer and Annotator

Acronym: THOTH

THOTH is a web app based developed as part of the TEXTaiLES toolbox as a dedicated 3D viewer and annotator. THOTH is based on the ATON Framework, a 3D viewer designed for uses in cultural heritage. Both THOTH and ATON are in turn built on Node.js using Three.js for a cross-platform web-based experience.

TEXTaiLES UAV – Autonomous Aerial System for Oversized Textile Digitisation

Acronym: TEXTaiLES UAV

TEXTaiLES UAV is an autonomous aerial robotic platform designed to safely digitise oversized and large-scale cultural heritage textile artefacts. The system employs an Unmanned Aerial Vehicle (UAV) equipped with RGB imaging and designed with a modular architecture that allows integration of additional sensing modalities when required.
The primary objective of the TEXTaiLES UAV system is to enable non-contact 3D digitisation of delicate textile artefacts that are difficult to document using static or ground-based systems. Dedicated offline path-planning algorithms generate pre-validated, collision-free trajectories, ensuring safe operation in indoor, GPS-denied environments. The platform supports structured data acquisition campaigns where navigation paths are computed, verified, and uploaded prior to flight.
The collected datasets are logged and integrated into the TEXTaiLES digital workflow to support AI-driven analysis, restoration activities, and Digital Twin generation within the ECCCH ecosystem.

Robot ZEUSbot

Acronym: Robot ZEUSbot

A comprehensive robotics platform designed to simplify the development, deployment, and evaluation of autonomous Unmanned Ground Vehicles (UGVs). The system enables users to configure, operate, and monitor a mobile robotic platform equipped with perception and manipulation capabilities, including object detection, pose tracking, data logging, and 3D reconstruction. Through an organized documentation structure, users can seamlessly install the software stack, manage sensor data, and deploy advanced computer vision pipelines for real-world experimentation or simulation in Gazebo environments.
The platform integrates both mobile base control and 6-DOF robotic arm manipulation, allowing coordinated navigation and interaction with objects of interest. Data collection and logging functionalities support structured experimentation, while simulation tools facilitate rapid prototyping and validation before real-world deployment. Each stage of the workflow—from system setup and environment configuration to perception, control, and reconstruction is designed to provide clarity, modularity, and scalability. With its flexible architecture, the platform can be extended to support additional robotic tasks and research scenarios, making it suitable for both academic experimentation and applied robotics development.

StyleShade3D Tool

Acronym: StyleShade3D

StyleShade3D is an existing WebGL-based tool originally developed within the PERCEIVE project. It enhances 3D mesh reconstruction by integrating deep learning segmentation into a conventional geometry and material pipeline. The system incorporates models such as Segment Anything Model to generate a semantic segmentation atlas- a 2D parameterized map aligned with the 3D surface - enabling region-aware shading and stylization. Different materials, rendering models, and artistic effects can therefore be assigned to specific parts of a mesh. The workflow begins in BlenderProc, where a full-orbit camera animation captures optimized multi-view images to ensure complete surface coverage. Inspired by path-planning strategies, the pipeline minimizes the number of views while preserving geometric completeness. Using SAM’s tracking features, consistent multi-view segmentation maps are generated and consolidated into a unified atlas. Since CAD models often lack UV coordinates, a texture atlas is created using Blender’s Smart UV Project algorithm. The segmented projections are merged into a 2D semantic atlas, which drives region-specific shading and deep stylization models.
Within the COLOURS project, StyleShade3D will be further extended and technically enhanced, expanding its capabilities for advanced semantic control, material modeling, and artistic rendering of 3D assets.

Light Damage Estimator

Acronym: LDE

The Light Damage Estimator is a decision-support tool developed within the framework of the PERCEIVE project and currently under improvement in the COLOURS project It addresses a critical challenge in cultural heritage preservation: predicting and mitigating light-induced colour change in sensitive artworks.
Designed for researchers, conservators, and museum professionals, LDE estimates how pigments and materials may fade or alter when exposed to specific lighting conditions over time. By combining the spectral power distribution of a selected light source with scientific datasets describing pigment sensitivity and fading behaviour, the tool simulates potential long-term visual changes.
LDE enables users to define exposure durations, compare lighting scenarios, and establish safe display thresholds. Rather than reacting to irreversible damage, professionals can proactively assess risk and optimize exhibition strategies. The tool supports both preservation and presentation, helping institutions balance conservation requirements with public access.

Multi-Layer eXtensible eXplorer

Acronym: MuLaX

MuLaX is an interactive Web3D/WebXR tool designed and developed under the PERCEIVE European project. It enables the online exploration of multi-layered 3D cultural heritage collections enriched with diagnostic and analytical data. Built on the open-source ATON framework developed by CNR ISPC, MuLaX integrates high-resolution 3D models with multi-band imaging datasets (e.g., VIS, UVL, VIL) and punctual analyses (e.g., XRF, microscope, FORS), allowing users to visually compare and inspect heterogeneous scientific results within a single environment .The tool supports advanced multi-layer discovery techniques, such as lens and split modes, semantic annotations, texture-based masks, and interactive semantic painting. It is accessible directly via web browser on any device, from smartphones to immersive XR headsets, ensuring wide interoperability and remote collaboration.
By integrating analytical datasets into spatially referenced 3D environments, MuLaX facilitates documentation, monitoring, interpretation, and communication of diagnostic processes for both experts and broader audiences.

Lenticular Colour Reconstruction Tool

Acronym: doLCE 2.0

doLCE 2.0 is a digital restoration tool dedicated to the recovery of colour information embedded in early lenticular film formats such as Kodacolor. In the COLOURS project, the tool was initially referred to as Kodacolor Colorizer.
Kodacolor, introduced in the 1920s, used a black-and-white film embossed with microscopic lenticules. When combined with a red-green-blue filter during filming and projection, these lenses encoded colour information as fine vertical RGB components within the monochrome emulsion.
doLCE 2.0 computationally extracts and reconstructs this hidden chromatic data. By analysing the lenticular structure and separating encoded RGB signals, the tool digitally reassembles full-colour frames from deteriorated or fragile historical films.
The system supports heritage professionals in restoring rare home movies and archival footage, ensuring accurate colour recovery while preserving original material characteristics. It represents a bridge between historical optical technologies and modern digital preservation methods.

ATON

Acronym: ATON

ATON is an open-source Web3D/WebXR framework developed by CNR ISPC to present and interact with complex and multi-dimensional data on the web. It allows to build interactive Web3D/WebXR applications without any installation required for final users. ATON provides a modular, cross-device platform for presenting and interacting with 3D models, scenes, and associated metadata in browsers using mobile devices, desktop workstation, museum kiosks, up to immersive VR/AR devices.
Built on modern web standards and large open ecosystems (Node.js, Three.js), ATON supports high-performance, responsive 3D visualization and collaboration. It offers a simple but extensible API for scene management, semantic annotations, measurements, advanced materials, synchronous collaboration, and spatial user interfaces for XR sessions. The framework embraces robust international formats (glTF, Cesium 3D Tiles) to ensure interoperability, streaming efficiency, and scalability for large geospatial or artifact datasets.
The framework also offers a plug&play architecture, to accelerate the creation and deployment of vertical applications (built on top of ATON features) - or extend horizontally its functionalities through plugins (“flares”) that may equip one or more web-applications.
ATON is thus a foundational framework for tools like MuLaX, traceable within COLOURS’ objectives to develop interoperable, web-based services for cultural heritage research and dissemination.

De-greening Tool

Acronym: De-greening Tool

The Degreening Tool is an AI-driven restoration service developed within the PERCEIVE framework and further aligned with the objectives of the COLOURS project. It addresses one of the most critical conservation issues affecting Autochromes, among the earliest successful colour photography processes, namely “greening” defects caused by material degradation.
Autochromes are extremely fragile and light sensitive; once degraded, their chromatic alterations are often irreversible. Since no ground-truth restored references exist, the tool employs synthetic defect generation to train deep learning models. By analysing real greening patterns, small spot-like stains and larger leakage-style areas with multi-ringed green halos and subtle orange borders, the system creates realistic simulated defects on digitized plates.
Paired datasets (defected vs. clean images) are used to train neural networks with optimization techniques, emphasizing correction of stained regions while preserving original detail. The result is a restoration model capable of removing green artifacts and reconstructing colour balance.
An additional fine-tuned inpainting module refills pale residual areas, ensuring visually coherent, historically respectful digital restoration.

PERCEIVE Colour Knowledge Repository

Acronym: CKR

The PERCEIVE Colour Knowledge Repository is an open data management platform designed to collect, structure, preserve, and openly share colour-related resources in cultural heritage. Developed within the PERCEIVE project and aligned with the objectives of the COLOURS, CKR supports interdisciplinary research on colour in artworks ranging from ancient polychrome sculpture to paintings, textiles, and historical photography and film. The repository follows FAIR data principles supporting the full data lifecycle from acquisition to analysis and long-term preservation. It hosts 3D models, scientific datasets, high-resolution images, documentation, simulation outputs, and risk assessment data related to colour change and deterioration.
Built on the InvenioRDM framework developed by CERN, CKR ensures robust data management, advanced metadata description, and integration with external research tools. Through federated cloud storage and REST APIs, the platform enables secure, distributed access and interoperability with other digital heritage services, including PERCEIVE and future Cultural Heritage Cloud infrastructures.
Repository access: https://perceive-data.iesl.forth.gr//

Negative to Positive

Acronym: neg2pos

The widespread success of color photography is largely attributed to the negative-positive process. In fact, the vast majority of photographs were historically captured on color negatives. Positive prints from these negatives were traditionally produced using enlargers. Today, billions of color negatives are held in archives and private collections worldwide, often without corresponding prints. A common preservation approach is to digitize the negatives and use image processing techniques to invert them into positive images. While tools like Negative Lab Pro and FilmLab facilitate this process, research into optimizing both the image capture and processing workflows remains limited. neg2pos addresses this gap by introducing a calibrated transformation framework designed to minimize aesthetic deviation from chromogenic C-prints while producing accurate and reproducible positive renderings. Collaboration with peers is essential to accurately determine the parameters of this transformation. By consulting diverse negative/print pairs preserved in photographic archives, the process can be systematically evaluated and calibrated to ensure consistency, historical fidelity, and aesthetic accuracy.

CoDesign Tool

Acronym: CoDesign Tool

The PERCEIVE CoDesign Tool is a structured, card-based framework developed by Consiglio Nazionale delle Ricerche (CNR ISPC) and Fraunhofer IGD within the Horizon Europe PERCEIVE project. It supports exhibition designers, curators, researchers, and cultural professionals in collaboratively creating engaging visitor experiences, physical, digital, or hybrid. The tool exists in two complementary versions: a paper-based board game, designed for in-person workshops; a web-based collaborative application, currently under development, enabling remote and hybrid sessions.
The framework builds on the open-access VisitorBox (Horizon 2020 GIFT project) and introduces an updated seven-step workflow, including the innovative Audience Goals stage focused on Authenticity and Sense of Care.
The tool will be adapted to support COLOURS’ objectives in colour-driven exhibition and heritage design.

ChatBot Assistant

Acronym: ChatBot Assistant

The ChatBot Assistant is StratiGraph’s embedded, multilingual AI helper for hands‑free, offline‑capable guidance in field and lab. It supports speech‑to‑text capture of stratigraphic notes, controlled vocabulary look‑ups, stepwise 3D capture guidance, and contextual documentation rules as part of the project’s validate‑as‑you‑collect workflow. “Frugal AI” models run locally on ordinary laptops/tablets and switch seamlessly to Cloud‑powered services when connectivity returns via the Variable Connectivity Protocol (VCP). The assistant also triggers the project’s NLP services to surface relevant publications and grey literature during documentation, binding citations and paradata to the Stratigraphic Knowledge Graph (SKG) with persistent identifiers. This hybrid design reduces curation burden and time‑to‑publication while keeping data FAIR and semantically aligned with the ECCCH.

Tropy (no acronym), with StratiGraph plugins

Within StratiGraph, Tropy is the image and document asset manager for ingesting, describing and semantically tagging photos, scans and notebook pages. Project plugins add ECCCH‑aligned templates, controlled vocabularies and knowledge‑graph synchronisation, promoting each image to a first‑class SKG entity linked to SUs, features, contexts and citations. NLP add‑ons support automatic extraction of dates, places, bibliographic mentions and paradata from captions and sheets, with expert review. Tropy works offline and synchronises via VCP; when online, image sets are packaged with IIIF for CATALOG publication. By turning imagery into semantically linked evidence, Tropy enables graph‑aware search, filtering and cross‑site comparison within the ECCCH.

pyArchInit

pyArchInit acts as the GIS‑centric field/post‑excavation client for spatial and stratigraphic forms, context sheets and SU relationships. StratiGraph aligns templates with the SKG so relationship declarations (e.g., fills, cuts, overlies) and attributes become typed graph edges. The edition emphasises offline reliability, map‑first entry/validation and exports that push structured, PID‑ready records to the SKG and CATALOG. It interworks with Extended Matrix (3D) and iDAI.Field, allowing teams to keep preferred tools while StratiGraph’s semantic layer harmonises and publishes outputs into ECCCH.

Extended Matrix

Acronym: EM

Extended Matrix delivers 3D‑centric stratigraphic documentation, formalising stratigraphy as a knowledge graph and enabling direct annotation/interpretation over 3D models (e.g., in Blender). It separates observations from interpretations, manages stratigraphic relations (including TSU—Transformation Stratigraphic Unit), and binds each 3D annotation to SKG entities and bibliographic evidence. EM implements validate‑as‑you‑collect checks against vocabularies and relation rules, exports to ECCCH, and interconnects with Tropy (imagery), pyArchInit (GIS forms) and CATALOG (publishable 3D collections with DOIs).

Collaborative Virtual Research Environment

Acronym: CVRE

The CVRE is a web‑based collaborative analysis and training space built on the ATON framework. It offers shared 3D/2D viewers, synchronous annotation, audio/chat channels, and structured “teaching lanes” used by StratiGraph Academy. Session annotations and decisions are persisted as graph content linked to SKG entities and, when relevant, published through CATALOG. With VCP, CVRE supports delayed sync from field devices, making telepresence reviews viable during campaigns and enabling multi‑party reconciliation across tools (pyArchInit, Extended Matrix, Tropy, iDAI.Field).

CATALOG

CATALOG assembles ready‑to‑publish collections (context registers, 3D views, image sets, method notes) and mints PIDs/DOIs for durable citation. Reading from the SKG, it makes relationships explicit (e.g., which image evidences which SU; which 3D view supports which interpretation), packages FAIR metadata/paradata/licences, and exposes sets to ECCCH discovery. Designed for research and preventive archaeology alike, CATALOG streamlines reports and publications, reducing the time from validated documentation to citable online outputs and generating media kits for dissemination and teaching subsets for training.

iDAI.Field

iDAI.Field is a modern, cross‑platform, open‑source stratigraphic recording application operated by the Deutsches Archäologisches Institut. Within StratiGraph, it is one of the interoperable field clients: teams who already use it can keep their workflows while StratiGraph’s semantic layer maps forms and relations to the SKG and forwards publishable registers to CATALOG. This protects prior investments and accelerates ECCCH‑aligned adoption alongside pyArchInit and Extended Matrix.

ATON Framework

Acronym: ATON

ATON is the web framework underpinning StratiGraph’s CVRE. It provides real‑time 3D/2D rendering, annotation, session management, and modular plug‑ins for communication and role‑based tools (teacher/reviewer/student). StratiGraph extends ATON with ECCCH‑aware persistence so that session artefacts become graph assertions tied to PIDs and licences. Thanks to VCP, ATON‑based clients cache actions during poor connectivity and replay them when back online. The framework integrates viewers for photogrammetry, point clouds and meshes, enabling telepresence research, training at scale, and multi‑party reconciliation ahead of CATALOG publication.

Online Collaborative Bibliographies Management Platform

Acronym: OCBMP

The Online Collaborative Bibliographies Management Platform is a web-based authoring and management environment designed to support the creation, enrichment, and long-term maintenance of annotated bibliographies in the cultural heritage domain. The platform addresses the fragmentation and manual effort currently involved in managing bibliographic resources by providing an integrated workspace where references, annotations, and linked cultural heritage objects can be collaboratively curated
The platform enables users to ingest bibliographic records from multiple sources, including existing repositories, digitised materials, and private collections, and to enrich them with semantic annotations aligned with established cultural heritage ontologies and controlled vocabularies. Bibliographic entries can be linked to digital cultural heritage assets such as 3D models, images, audiovisual material, and stratigraphic records, enabling bibliographies to function as structured, contextual knowledge artefacts rather than static lists of references.

Annotated Bibliography Authoring Assistant

Acronym: ABAA

The Annotated Bibliography Authoring Assistant is an AI-powered service designed to support cultural heritage professionals and researchers in the efficient creation, enrichment, and evaluation of annotated bibliographies. Building on advances in natural language processing and semantic technologies, the assistant automates time-consuming tasks such as reference discovery, metadata extraction, annotation drafting, and thematic classification, while keeping the user fully in control of scholarly decisions.
The assistant operates as an intelligent companion within bibliographic workflows, facilitating already analysed textual, visual, and audiovisual sources to suggest relevant references, generate preliminary annotations, and highlight semantic relationships between sources and cultural heritage objects. It supports multilingual and multiscript content, enabling the processing of historical documents, digitized manuscripts, and legacy bibliographic records that are not readily machine-readable.
Rather than replacing scholarly judgement, the tool is designed to augment it. All AI-generated suggestions are transparent, explainable, and editable, ensuring traceability and academic accountability.

Retro-Digitised Content Parsers

Acronym: RCP

The Retro-Digitised Content Parsers are a set of specialised tools designed to transform heterogeneous, legacy, and non–machine-readable cultural heritage sources into structured, semantically enriched digital resources. The parsers address a critical bottleneck in cultural heritage research and documentation: the vast amount of knowledge locked in scanned documents, historical manuscripts and other retro-digitised materials.
The tools support advanced digitisation workflows, including optical character recognition (OCR) for printed and handwritten texts, multiscript and multilingual text extraction and quality assessment of recognition outputs. In addition to producing machine-readable content, the parsers generate paradata that document the digitisation process, uncertainty levels, and transformations applied to the source material.
By aligning extracted content with controlled vocabularies and semantic models, the parsers enable downstream services such as annotated bibliography creation, archival documentation, and narrative generation. Designed for integration with the European Collaborative Cloud for Cultural Heritage (ECCCH), the tools ensure that previously inaccessible or fragmented sources can be systematically ingested, reused, and linked within interoperable cultural heritage workflows.

Multimodal Content Analysis and Classification Service

Acronym: MCACS

The Multimodal Content Analysis and Classification Service is an AI-enabled service designed to analyse, classify, and semantically enrich heterogeneous cultural heritage content, including text, images, and structured metadata. The service supports the creation of high-quality annotations and classifications that enable discovery, reuse, and contextualisation of cultural heritage resources across disciplines.
By combining natural language processing, the service identifies entities, events, concepts, and relationships within and across different media types. It supports domain-specific analysis tasks such as bibliographic classification, thematic tagging, event and actor recognition, and controlled vocabulary assignment. The service is designed to work with both born-digital content and outputs generated by retro-digitisation workflows.
The service acts as a core analytical layer for multiple ARXIVE tools, feeding enriched metadata into bibliographic authoring, archival documentation, visualization, and narrative generation services. Integrated with the European Collaborative Cloud for Cultural Heritage (ECCCH), it ensures that analytical outputs are interoperable, traceable, and compliant with FAIR principles.

Interactive Visualization and Annotation Platform

Acronym: IVAP

The Interactive Visualization and Annotation Platform is a web-based environment that enables cultural heritage professionals to explore, visualise, and annotate heterogeneous cultural heritage data in an integrated and semantically enriched workspace. The platform supports the interactive examination of digital cultural heritage assets, including 3D models, images, stratigraphic data, and associated metadata, providing a unified interface for documentation and analysis.
The platform is designed to support granular, context-aware annotations that can be linked to specific components of digital assets, such as areas of a 3D model, stratigraphic layers, or regions within images. Annotations are semantically mapped using established ontologies and controlled vocabularies, ensuring interoperability and reuse across tools and institutions.
By combining advanced visualisation techniques with collaborative workflows, the platform enables multiple users to contribute annotations, track changes, and explore relationships between assets, bibliographic references, and narratives. Integrated with the European Collaborative Cloud for Cultural Heritage (ECCCH), the platform acts as a central interface for interacting with semantically enriched digital twins and archival documentation.

Reporting and Narrative Generation Tools

Acronym: RNGT

The Reporting and Narrative Generation Tools are a suite of applications designed to support the creation of structured reports, narratives, and dissemination outputs based on semantically enriched cultural heritage data. The tools enable cultural heritage professionals to transform complex archival records, annotations, and bibliographic resources into coherent textual and visual outputs tailored to different audiences and regulatory contexts.
The suite includes a flexible template engine, a collaborative authoring environment, and automated content generation components that draw directly on metadata, annotations, and linked digital assets stored in the Cloud. Reports can be generated for diverse purposes, such as excavation documentation, conservation records, research publications, and public-facing narratives, while ensuring consistency with institutional and national reporting requirements.
By leveraging semantic models and linked data, the tools ensure that generated narratives remain traceable to their underlying sources, preserving provenance and citation integrity. Integrated with the European Collaborative Cloud for Cultural Heritage (ECCCH), the suite supports interoperable, reusable, and FAIR-compliant reporting workflows.

Smart Archival Assistance System

Acronym: SAAS

The Smart Archival Assistance System is an intelligent support service designed to enhance the quality, consistency, and efficiency of cultural heritage archival workflows. The system provides context-aware assistance throughout the documentation lifecycle, helping professionals validate, enrich, and maintain archival records and annotations linked to cultural heritage objects.
By combining machine learning techniques with semantic technologies, the system analyses archival content to suggest improvements, detect inconsistencies, and highlight missing or incomplete information. It supports tasks such as metadata quality checks, validation against controlled vocabularies, provenance verification, and contextual recommendations based on related assets and bibliographic references.
The system operates as a non-intrusive assistant integrated into existing tools, offering recommendations and alerts rather than enforcing automated decisions. All suggestions are explainable and traceable, preserving professional accountability. Integrated within the European Collaborative Cloud for Cultural Heritage (ECCCH), the Smart Archival Assistance System contributes to sustainable, high-quality archival practices across institutions.

Knowledge Infrastructure and FAIR Data Management Service

Acronym: KIFDMS

The Knowledge Infrastructure and FAIR Data Management Service provides the foundational semantic and data management backbone that enables ARXIVE tools to operate as interoperable, sustainable services within the European Collaborative Cloud for Cultural Heritage (ECCCH). The service delivers a coherent framework for representing, storing, validating, and interlinking cultural heritage data, metadata, and paradata across diverse domains and institutions.
The service implements extensions of established cultural heritage standards, such as CIDOC CRM, and aligns them with FAIR principles to support consistent representation, provenance tracking, and versioning of cultural heritage objects and their digital twins. It supports the ingestion, reconciliation, and synchronisation of data from heterogeneous repositories, ensuring that knowledge produced by different tools and workflows remains connected and reusable over time.
Rather than acting as a user-facing application, the service underpins all ARXIVE tools by providing shared semantic artefacts, APIs, validation mechanisms, and quality control processes. Integrated with ECCCH infrastructure, it ensures that cultural heritage knowledge produced within ARXIVE is discoverable, interoperable, and sustainably managed.

Simulation models of windpipes and traditional musical instruments: airflow calculation, parameterization and data generation.

Acronym: TK1

Development of Finite Element Method (FEM) and Computational Fluid Dynamics (CFD) simulation models of windpipes and traditional music instruments as well as parameterisation of the simulation results for efficient and generalizable data generation until the point to make the simulations audible (auralization). This toolkit focuses on simulating windpipes and traditional organs through three key aspects. First, it utilizes FEM and CFD for frequency, amplitude and directivity modelling. Second, it develops a parameterization approach that allows for faster data generation across various pipe geometries, which can later be adapted into AI-based models. Third, it integrates single pipe directivities to a body-organ directivity for fast auralization algorithms. A complete organ sound simulation workflow for wide sound sources in historical rooms will be ready, implementing the new advances into the existing state-of-the art procedures. Fourth, it incorporates material aging into the simulation to estimate the change in sound quality.

Physical - Virtual Restoration of Cultural Heritage Musical Instruments

Acronym: TK2

Recreating the original sound and appearance of cultural heritage musical instruments by physical model reconstruction based on digitization of artifacts, simulations, and craftsmanship. The toolkit will aim towards the recreation of the original appearance of cultural heritage musical instruments, such as the Hydraulis and the Pipe Organ, through digitization, craftsmanship, and physical model reconstruction. The process will involve historical research, artifact study, and documentation analysis to understand the instruments' original design and materials. Advanced imaging techniques, including 3D scanning and photogrammetry, will facilitate precise digital modelling, while CAD simulations will ensure accurate physical and acoustic properties. Prototyping through 3D printing, acoustic testing, and iterative refinement will guarantee authenticity. The toolkit will try to integrate comprehensive documentation, user-friendly software, and training modules to bridge digital reconstruction with traditional craftsmanship. Dissemination efforts will include workshops, open-access resources, and community engagement to ensure broad applicability. A sustainability plan will be incorporated to refine and preserve reconstructed instruments. This approach will try to offer a holistic solution for cultural heritage preservation, ensuring the revival and continued study of historical musical instruments.

Acoustic Profile Digitization of Architectural Spaces

Acronym: TK3

Capturing and modelling the acoustic profiles of architectural spaces using advanced techniques such as directional impulse responses and ambisonic recordings as well as autonomous measurements using a robotic platform. Provision of the necessary data for accurate simulations, supporting applications in cultural heritage, immersive auralization, and acoustic parameter analysis and research. This toolkit enables the digitization of the acoustic characteristics of architectural spaces, capturing how sound sources dynamically interact with their surroundings. The process involves the collection and analysis of directional impulse responses at a large number of spatially dense points in the environment. Alternatively, it can model the space using geometric properties and wall absorption and scattering profiles, which can then be used to synthesize directional impulse responses. This data provides a comprehensive understanding of the sound propagation, reflections, and reverberation patterns unique to the environment, enabling accurate simulations for cultural heritage preservation, virtual reconstructions, and immersive experiences.

Long-term Acoustic Projections

Acronym: TK4

AI and computational models project sound quality over extended periods, simulating environmental and usage effects on acoustic properties of cultural heritage objects. Sensor-based monitoring tools for real-time data capture (temperature, humidity, etc), enabling ongoing evaluation of an object’s condition. This toolkit enables the simulation of long-term changes in the acoustic characteristics of architectural spaces due to environmental and structural evolution. By integrating historical data, material science, and acoustic measurements and modelling, it reconstructs past soundscapes and predicts future acoustic transformations. The toolkit minimizes the need for physical interventions in heritage sites, reducing risks to their structural integrity. It supports sustainable restoration practices by projecting how changes to architectural or environmental conditions may impact acoustics. Additionally, for collaborative projects like restoring churches with organs, it allows simulations that ensure historically accurate and efficient restoration efforts.

Predictive Modelling for Material Deterioration and Sound Degradation

Acronym: TK5

AI models that predict material degradation based on historical data and environmental conditions help track the deterioration of objects like organs or ancient musical instruments. AI algorithms coupled with simulations that predict detuning based on environmental data (temperature, humidity, UV exposure, etc.) apply to dynamic objects like musical instruments. This toolkit enables predictive maintenance for musical instruments based on simulated acoustic data and environmental factors like temperature and humidity. It detects degradation patterns, predicts aging effects on sound quality, and informs proactive measures to mitigate environmental impacts. This approach also aids in material selection for new instruments, estimation of service intervals, and helps to predict maintenance costs. It also allows for reverse predictions of an instrument's original sound.

Interactive XR Experiences with Spatialized Acoustics

Acronym: TK6

Virtual and Augmented Reality platforms allow users to interact with digital reconstructions of cultural heritage objects and their environments. This toolkit integrates 3D visual reconstructions of architectural spaces with corresponding spatialized acoustic environments, delivering highly immersive and interactive Extended Reality (XR) experiences. Users can navigate these reconstructions and experience dynamic, location-specific acoustics, creating a seamless fusion of auditory and visual realism. By leveraging advanced technologies like ambisonic recordings and real-time synchronization, the toolkit reproduces the unique acoustic properties of each space, tailored to user positions and movements. For example, visitors can explore a virtual church and perceive its acoustic characteristics from the altar, aisles, or pews, or experience how the organ and acoustics sounded in a different historical period, provided sufficient documentation exists. Additionally, it offers unique perspectives, such as simulating how an organist perceives the music from their position, enhancing educational and immersive storytelling applications. This dynamic integration supports cultural heritage preservation, academic study, and immersive storytelling, bridging historical environments with cutting-edge XR technologies.

Cross-modal Data Integration

Acronym: TK7

CASPAR provides a domain-agnostic tool for the automated retrieval and fusion of in-house structured data from disparate sources into domain-specific semantic models (ontologies), in order to enable the discovery of new knowledge and facilitate the extraction of actionable insights in a way that simulates the human reasoning. The product has been tested in numerous projects such as MuseIT, SILVANUS, PREVISION, etc, reaching TRL5 on these domains. For MusicSphere, we will start from TRL4 using the skeleton of the CASPAR framework as it has been deployed in MuseIT and will be expanded up to TRL7 by the end of the project satisfying the project needs.

Real-time polyphonic music transcription with tuning adaptation

Acronym: TK8

AI-based algorithms that transcribe real or simulated musical performances on arbitrary harmonic instruments in real-time. Analysis results include score-like music representations as well as parameters related to tuning frequencies and instrument quality. This toolkit provides software for real-time music transcription (AMT) of (historic) polyphonic instruments, which are often based on different tuning systems. The underlying processing steps include an analysis of the tuning and timbre qualities of the instrument and a combination of multi-pitch estimation and note formation resulting in a note-by-note transcription of the musical performance. The parameters extracted using this toolkit allow for comparing real and simulated performances on the instrument and evaluating the simulation quality in a feedback loop. At the same time, the toolkit opens up new perspectives for interdisciplinary research into historical music practices. In the context of interactive XR experiences, real-time music transcription can be integrated as performance feedback for musicians learning to perform on historical instruments.

Real-time AI-based synthesis of historic music organs for virtual instrument performances

Acronym: TK9

AI-based sound synthesis models that convert note-level performance parameters extracted with TK8 into realistic music organ sounds. Real-time processing allows for the integration of the synthesis models into a virtual instrument that is playable as part of an interactive XR experience. This toolkit provides software for real-time synthesis of music organ sounds. The underlying AI-based synthesis models are trained using simulation results from the physical modelling algorithms as provided in TK1 and TK2. While the physical modelling approach incorporates knowledge about precise instrument construction and sound production mechanisms, it is computationally too expensive for a real-time application scenario. Data-driven audio synthesis algorithms offer an alternative approach for audio synthesis, which can be implemented in real-time but at the same time requires large amounts of recorded or simulated audio data to learn the complex mapping between instrument control input (such as pressed keys) and the resulting instrument sound. Complementary to TK8, real-time audio synthesis can be integrated into interactive XR experiences, allowing musicians to virtually play historic music organs in realistically simulated acoustic environments. Virtual keyboard interfaces generate control signals to guide the audio synthesis.

AUTOMATA AI Recognition & Knowledge Discovery Services

Acronym: AUTOMATA-AI

The AI tooling delivers automated classification and knowledge discovery across multimodal datasets (e.g., images, 3D, and analytical measurements). The AI-based Enrichment and Classification Services are a set of modular, cloud-native services developed within the AUTOMATA project to automate the analysis, enrichment, and interpretation of archaeological artefacts. These services apply machine learning techniques to support the classification and comparison of artefacts across collections. The tool incorporates XAI techniques so that results can be inspected and debated by domain experts, supporting controlled improvement as curated datasets expand during the project. In ECCCH, the intent is to deliver reusable analytical capabilities that can be invoked by other services (e.g., visualisation modules) while maintaining transparency through confidence reporting.

AUTOMATA Interactive Data Visualisation Module

Acronym: AUTOMATA-VIZ

AUTOMATA-VIZ is an interactive data visualisation and visual analytics module designed to help users explore, summarise, and interpret enriched digitisation outputs at scale. It supports dataset-level exploration and comparison across collections by bringing together multiple evidence layers such as object geometry, appearance information, archaeometric measurements, and contextual metadata. The module provides explanatory views that help users identify patterns, compare subsets of artefacts, and inspect relationships among variables including correlation-oriented views where data support them. Within ECCCH, AUTOMATA-VIZ operates directly on cloud-hosted, authoritative datasets, enabling web-based exploration where possible through lightweight online access, complemented by optional advanced local exploration when higher-resolution data or more complex analysis are required.
It is intended to serve both research workflows and institutional dissemination needs. AUTOMATA-VIZ is not a repository or reference information system itself; it acts as a visual analytics layer built on top of existing infrastructures and repositories. It is the insight layer that turns complex, multi-modal datasets into interpretable views, while maintaining links to the underlying object records held in ECCCH.