Transparent AR Markers

A novel AR system using stacked transparent markers as a tangible, compositional input method.

Year Active: 2023

Stack AR introduces a novel alternative input method for augmented reality: stacking transparent fiducial markers as a tangible, compositional interface. Unlike traditional AR systems that rely on single, opaque markers or touchscreens, Stack AR leverages the physicality and combinatorial potential of transparent markers—users can physically layer, rearrange, and combine them to create new digital commands and musical sequences in real time.

This approach transforms the act of input into a playful, hands-on experience, where each unique stack of markers is instantly recognized and mapped to distinct digital outcomes. The system’s technical innovation lies in its custom marker design and robust computer vision pipeline, which together enable reliable detection and decoding of multiple, semi-transparent, and overlapping markers. This unlocks a new dimension of tangible interaction, allowing users to “compose” with their hands by stacking, mixing, and remixing physical elements—blurring the boundary between the physical and digital in AR.

If you are interested in the technical paper, please read (Chen et al., 2023).

Technical Highlight


  • Custom ArUco Marker Library
    A bespoke ArUco dictionary was created to support marker stacking on transparent film, minimizing interference and ensuring that stacked combinations generate unique, detectable codes. Marker regions are designed in concentric “rings,” allowing selective stacking and clear layer separation.

  • Computer Vision Integration in Unity
    The Unity application employs the OpenCV for Unity plugin to enable real-time marker recognition and pose estimation from a top-down camera. Custom logic parses the detected markers, identifies their combination, and maps them to distinct musical or visual elements in the AR environment.

  • Detection and Interactive Synthesis
    By precomputing all viable marker combinations, Stack AR can identify both individual and stacked markers in a single frame, triggering the correct set of audio, 3D models, and particle effects. The system is resilient to partial occlusion, variable lighting, and the material challenges of transparent.

Presentation


Stack AR was selected for presentation at the MIT Reality Hack 2025 (Jan 23–27) as part of an art grant. The project can be easily set up with a desktop, camera, and printed transparent markers.

Demo Video


Presentation Video


References

2023

  1. stack-ar-teaser.png
    Stackable Music: A Marker-Based Augmented Reality Music Synthesis Game
    Max Chen, Shano Liang, and Gillian Smith
    In Companion Proceedings of the Annual Symposium on Computer-Human Interaction in Play, Stratford, ON, Canada, Mar 2023