3D textured model encryption via 3D Lu chaotic mapping

Xin Jin, Shuyun Zhu, Chaoen Xiao, Hongbo Sun, Xiaodong Li, Geng Zhao, Shiming Ge

Abstract

This paper presents a novel hierarchical encryption scheme for 3D textured models designed to balance high security with computational efficiency. By decomposing 3D models into their constituent parts—vertices, polygons, and textures—and applying varying levels of chaotic mapping, the method achieves robust protection against unauthorized access. Specifically, the high-complexity 3D Lu chaotic map is used for critical structural data, while faster 1D and 2D maps secure secondary data. Experimental results demonstrate a massive key space of $10^{135}$ and high sensitivity to key changes, significantly outperforming uniform high-dimensional encryption methods in processing speed.

Introduction

As Virtual Reality (VR), Augmented Reality (AR), and 3D printing become mainstream, the intellectual property and confidentiality of 3D models have become paramount. 3D models are complex "multimodal" entities consisting of point clouds (vertices), meshes (polygons), and 2D image wraps (textures).

Previous encryption attempts often faced a trade-off: they either encrypted components individually without considering the whole model, or they applied heavy-duty, high-dimensional encryption to every bit of data, resulting in slow processing for complex models. This research addresses the efficiency-security trade-off by observing that the "skeleton" (vertices) requires heavier encryption than the "skin" (texture) to render a model unrecognizable.

Methodology

The core innovation is a hierarchical (multi-level) encryption architecture based on chaotic maps. Chaotic systems are ideal for cryptography due to their high sensitivity to initial conditions and pseudo-random outputs. The proposed method decomposes the 3D model into three layers, applying different "strengths" of chaos to each:

  1. Vertices (The Skeleton): Since vertices are critical for identifying a model's shape, they are encrypted using a 3D Lu chaotic map. This high-complexity system produces a random vector to shift the 3D coordinates of every point.
  2. Polygons (The Mesh): To scramble the connections between points, a 2D Arnold’s Cat Map is used to shuffle the indices of the triangles forming the mesh.
  3. Textures (The Skin): Textures are treated as 2D images and encrypted using a 1D Logistic Map combined with DNA coding. DNA coding translates pixel data into biological sequences (A, C, G, T) for algebraic operations, adding complexity to the fast 1D process.
Decomposition of a 3D textured model into its constituent components: vertices, polygons, and textures.

Finally, these three encrypted components are reassembled into a "cipher" model that appears as random noise.

Key Results and Security Analysis

The authors evaluated the method on a variety of 3D textured models with different geometric complexities, such as clocks, animals, and radios.

  • Security & Key Space: The system boasts a massive key space of approximately $10^{135}$ (roughly $2^{449}$), making brute-force attacks practically impossible compared to the $2^{256}$ required for top-tier AES.
  • Statistical Robustness: Using Viewpoint Feature Histograms (VFH), the researchers found that while plain models have distinct signatures, the encrypted models' histograms are uniform and random, ensuring no structural information leaks.
  • Sensitivity: The system is extremely sensitive; changing a key by as little as $10^{-10}$ results in a failed decryption.
  • Efficiency: By utilizing lower-dimensional maps for textures and polygons, the method is significantly faster than previous state-of-the-art methods that applied 3D maps to all components.

Conclusion and Future Work

This hierarchical approach demonstrates that focusing "heavy" mathematical operations on vertices achieves high security without the computational overhead of treating the entire file as a high-dimensional system. Future work will involve psychological experiments to determine human recognition thresholds for optimized encryption and migrating the implementation to C/C++ to further boost speed for real-time VR/AR applications.

Summary updated

BibTeX
@article{jin2017d,
  author = {Xin Jin and Shuyun Zhu and Chaoen Xiao and Hongbo Sun and Xiaodong Li and Geng Zhao and Shiming Ge},
  title = {{3D textured model encryption via 3D Lu chaotic mapping}},
  journal = {Science China Information Sciences},
  year = {2017},
  url = {https://openreview.net/forum?id=thXBHrdpuq0},
}
Figure