PIXEL VR: Optimizing Photogrammetric Datasets for Standalone VR

Photogrammetric models are increasingly employed for heritage documentation, education, and interactive visualization. However, their complexity and size limit in their applicability on standalone Virtual Reality (VR) devices or low-end machines, which typically operate under significant hardware constraints. This research addresses these limitations through the development of an automated optimization workflow implemented as a Blender Python script. The proposed pipeline integrates a series of processes, remeshing, decimation, UV unwrapping, and texture baking, to significantly reduce polygon count while preserving visual fidelity. Case studies have been retrieved using open-access datasets and original surveys from the Carleton Immersive Media Studio (CIMS) and demonstrate polygon reductions exceeding 99% with minimal visual degradation, enabling real-time visualization on limited hardware. The study emphasizes accessibility and replicability by exclusively utilizing open-source and or free to use software, allowing a scalable, cost-effective solution for immersive cultural applications.