From Photos to Presence: Exploring Accessible 3D Capture, Photogrammetry, and Web-Based AR for Cultural Heritage, Education, and Training

Abstract (Draft)

This draft white paper explores how accessible 3D capture workflows—including photogrammetry, LiDAR, Gaussian splats, and AI-assisted reconstruction—can bridge the gap between highly technical 3D pipelines and real-world needs in cultural heritage, education, training, and e-commerce.

The goal of this document is not to present a finished empirical study, but to share a working synthesis of practice-based insights and a concrete direction for future research, tooling, and open collaboration.

Introduction

Over the last decade, 3D capture and immersive media have become dramatically more accessible. Commodity cameras, depth sensors, LiDAR-equipped mobile devices, and AI-driven reconstruction now make it possible for non-specialists to digitize objects, spaces, and scenes that would once have required dedicated scanning hardware.

Yet many of the organizations that would benefit most—museums, archives, educators, training commands, and even product teams—still struggle to move from experimentation to sustainable practice. Common blockers include:

• Lack of in-house 3D specialists or pipeline ownership.
• Fragmented tooling across capture, processing, optimization, and publishing.
• Difficulty connecting 3D assets to narrative, instructional, or commercial outcomes.

This paper frames these challenges through the lens of an "object-first" approach: starting with the real-world artifacts, devices, or products that matter to people, and designing workflows that make it easy to capture, preserve, and present them over time.

Background & practice-based context

The perspective of this draft is practice-based. It synthesizes work across:

• Co-founding and co-inventing a patented AR SaaS platform used in defense and enterprise training.
• Photogrammetry-driven AR experiences for museums and cultural organizations.
• Experiments with Gaussian splats and web-based viewers for immersive storytelling.
• Ongoing qualitative research with museum professionals, educators, developers, and 3D creators as part of Startup Virginia’s Idea Factory.

Rather than treating photogrammetry or AI reconstruction as isolated technologies, the paper treats them as components inside a broader lifecycle: capture → process → optimize → publish → interpret → revisit over time.

Problem space

Interviews and prior work suggest that the barrier is not simply access to tools, but the lack of a unified, opinionated workflow that connects tools to outcomes. Common patterns include:

• 3D initiatives framed as one-off "projects" instead of ongoing, maintainable practices.
• Confusion over which capture method (photogrammetry, LiDAR, splats, AI) is appropriate for which object, space, or use case.
• Limited guidance on how to translate raw scans into accessible, performant web experiences that can be embedded in existing platforms.

These gaps mirror what emerged in the Idea Factory research: the need for standardized workflows, lightweight tools, and storytelling layers that make 3D assets more than just technical artifacts.

Approach & early direction

The pilot implementation of these ideas is Codex3D—the minimum viable experiment you see on this site. It focuses on a service-oriented and later platform-oriented workflow:

• Accept simple photo uploads from non-specialists (educators, curators, product owners).
• Route objects through appropriate 3D pipelines: AI, photogrammetry, Gaussian splats, or hybrid workflows.
• Return ready-to-use outputs (GLB, USDZ, web viewers) and eventually embeddable experiences with narrative layers.

In the current phase this is deliberately manual behind the scenes: a way to test demand and refine workflows before codifying them into software. Insights from this experiment will shape the more formal research questions and structure of the full paper.

Planned additions & open questions

As a work in progress, this draft intentionally leaves space for:

• A more formal literature review on photogrammetry in museums, education, and training.
• Clearer taxonomy of when to use meshes, splats, or volumetric approaches.
• Deeper analysis of qualitative interview data from Idea Factory and beyond.
• Design patterns for narrative-first 3D experiences on the web.

Feedback, critique, and collaboration are welcome—this page is meant to be a transparent snapshot of the thinking before it becomes a polished, citable paper.