Skip to content
md.mudassir
← all work
07 / case study

Auth3D

Digitally signing and verifying 3D models — presented at ICTIS 2026.

Type
Research
Year
2025–26
Role
Author & developer
For
MSc · Christ University
Auth3D website

A platform for proving the ownership and authenticity of 3D models using geometry-aware fingerprints and embedded manifests instead of plain file hashes. The paper, "A Platform to Digitally Sign and Verify 3D Graphic Models", was presented at the 11th ICTIS conference in Bangkok.

The problem

A file hash breaks the moment a mesh is re-exported, re-scaled or lightly edited — so it can't prove that the model you found is the one an artist actually made.

What I built

  • Geometry-aware fingerprinting for OBJ models that survives common mesh transformations.
  • Auth3D manifest creation and embedding directly into the model file.
  • Ownership registration with conflict detection, and offline-style verification against the manifest.
  • Research benchmarks comparing fingerprint stability against file hashing and steganography.
  • Artist console with dashboard, marketplace, signing and verification flows, plus a protocol user guide.

Outcome

  • Paper presented at the 11th International Conference on ICT for Intelligent Systems (ICTIS 2026), Bangkok, 9–11 April 2026.
  • Interactive prototype live at auth3d.netlify.app.