From Anonymous Faces to Provable Privacy: A Journey Through Deep De-Identification and Beyond

Author

Tao Li
Dept. of Computer Science
Purdue University
taoli@purdue.edu

Download

➜ sha256 dissertation.pdf
SHA256 (dissertation.pdf) = 65f2b4d3922132aaee3d86ba5abd7e7d5697a54e31221b196fbf39bb30d64a6d

Abstract

As the use of visual data grows in surveillance, social media, and machine learning, safeguarding individual privacy while preserving data utility has become an urgent challenge. This dissertation, From Anonymous Faces to Provable Privacy, presents a comprehensive progression in visual data de-identification—from early heuristic methods to rigorous privacy guarantees—spanning diverse representation spaces and biometric modalities. The first technical contribution introduces AnonymousNet, a structured face obfuscation framework operating in a discrete facial attribute space. It demonstrates that facial privacy can be quantified and manipulated while maintaining image realism. Next, DeepBlur offers a lightweight and practical latent-space obfuscation method with strong empirical performance, though without formal guarantees. The dissertation then advances a formal privacy definition tailored to single-image publication, proposing a latent-space mechanism that satisfies this definition under ε-differential privacy. This framework bridges empirical utility with worst-case privacy guarantees. To improve the visual fidelity of privatized images, CAGFace, a facial component-aware super-resolution module, is introduced. Finally, the proposed formal privacy framework is extended beyond facial imagery to additional modalities, including gait, voice, and full-body appearance. In particular, the dissertation introduces the first provably private gait de-identification method based on skeleton data. Together, these contributions establish a foundation for modality-agnostic, privacy-preserving data publishing, grounded in robust theoretical foundations and practical relevance.

Citation

If you find this dissertation useful, please consider citing it:

@phdthesis{li2025from,
  title     = {From Anonymous Faces to Provable Privacy: A Journey Through Deep De-Identification and Beyond},
  author    = {Tao Li},
  year      = {2025},
  month     = {August},
  school    = {Purdue University, Department of Computer Science},
  type      = {Ph.D. Dissertation},
  url       = {https://drive.google.com/file/d/1IBjvNb2OINQ0efyhFV0tHMUva2Z76dGd/view}
}