Sripatum Digital Media Lab

A reconfigurable immersive media lab integrating multi-surface projection with dedicated LiDAR and depth-camera sensor servers, real-time media applications, networked presentation, and spatial audio for teaching, experimentation, and spatial interaction prototyping.

The Sripatum Digital Media Lab is a reconfigurable immersive environment developed to support teaching, interactive media experimentation, spatial interaction prototyping, and immersive content presentation. Unlike an installation designed around a single visitor experience, the lab is structured as reusable technical infrastructure. Multi-surface projection, spatial sensing, real-time processing, networked media distribution, and spatial audio can be configured for different teaching and experimental scenarios. The sensing architecture separates sensor processing from the applications responsible for interactive content. A dedicated Raspberry Pi operates as a LiDAR sensor server, processing data from multiple LiDAR sensors and providing reusable spatial information such as position data and configurable area-detection states across the local network. A second Raspberry Pi operates as a dedicated depth sensor server. The depth-camera pipeline processes sensor input independently and provides multiple representations of the captured environment, including conventional image output, depth information, point-cloud data, and detected spatial objects represented as box colliders. Processed sensor information can then be consumed by applications such as TouchDesigner, Unity, or Unreal Engine without requiring each application to independently implement the complete sensor-processing pipeline. The lab can therefore support different modes of use, ranging from conventional teaching and software demonstrations to immersive projection, sensor-based interaction experiments, interactive game development, and the prototyping of new responsive environments. This modular architecture also provides a foundation for future research instrumentation, where spatial interaction data can be captured independently from the visitor-facing application and subsequently connected to a broader research data infrastructure.

Development and integration of the technical architecture for a reconfigurable immersive lab, including multi-projector presentation, networked media workflows, spatial audio, LiDAR sensing, depth-camera processing, and reusable sensor-to-application communication. The sensing workflow uses dedicated Raspberry Pi devices as independent sensor-processing servers. The LiDAR server consolidates data from multiple LiDAR sensors and provides processed spatial coordinates and configurable area-detection states to applications across the local network. A separate depth sensor server processes depth-camera input and provides multiple data representations for interactive applications, including image output, depth information, point-cloud data, and detected objects represented as spatial box colliders. Separating sensor processing from the application layer allows TouchDesigner, Unity, Unreal Engine, and future applications to consume processed spatial information without each application having to directly manage individual sensors or reproduce the complete sensing pipeline. The architecture is designed to make the lab reusable across teaching, experimentation, interactive development, and future spatial interaction research rather than coupling the infrastructure to a single installation or software workflow.

The lab is being developed as reusable immersive and interactive infrastructure that can support multiple sensing technologies, real-time applications, media pipelines, and teaching scenarios within the same physical environment. Its modular sensor-server architecture allows LiDAR and depth-camera processing to operate independently from the applications used to create interactive experiences, making the infrastructure easier to reuse and extend for future workshops, prototypes, installations, and research experiments. The same architecture is intended to provide a technical foundation for future research instrumentation, allowing spatial interaction data to be captured and analyzed without making the live media application responsible for the complete research-data pipeline.