Skip to content

Read-only version: For security reasons, the Research Information System (FIS) has been temporarily replaced by this read-only version. All public content and full texts remain available; login, editing, submission and statistics are not available. Questions: fis@uni-bamberg.de

Home › Publications › Privacy-preserving datasets by capturing feature distributions with Conditional VAEs

Publication

Privacy-preserving datasets by capturing feature distributions with Conditional VAEs

Di Salvo, Francesco; Tafler, David; Doerrich, Sebastian; u. a. (2024): Privacy-preserving datasets by capturing feature distributions with Conditional VAEs, in: Proceedings : The 35th British Machine Vision Conference 2024, S. 1–14.

Details

Faculty/Chair
Faculty/Professorship
Author
Tafler, David
By
... ; Doerrich, Sebastian; ...
Title of the compilation
Proceedings : The 35th British Machine Vision Conference 2024
Conference
BMVC 2024, The 35th British Machine Vision Conference, 25th - 28th November 2024, Glasgow
Year of publication
2024
Language
English
URL
Abstract

Large and well-annotated datasets are essential for advancing deep learning applications, however often costly or impossible to obtain by a single entity. In many areas, including the medical domain, approaches relying on data sharing have become critical to address those challenges. While effective in increasing dataset size and diversity, data sharing raises significant privacy concerns. Commonly employed anonymization methods based on the k-anonymity paradigm often fail to preserve data diversity, affecting model robustness. This work introduces a novel approach using Conditional Variational Autoencoders (CVAEs) trained on feature vectors extracted from large pre-trained vision foundation models. Foundation models effectively detect and represent complex patterns across diverse domains, allowing the CVAE to faithfully capture the embedding space of a given data distribution to generate (sample) a diverse, privacy-respecting, and potentially unbounded set of synthetic feature vectors. Our method notably outperforms traditional approaches in both medical and natural image domains, exhibiting greater dataset diversity and higher robustness against perturbations while preserving sample privacy. These results underscore the potential of generative models to significantly impact deep learning applications in data-scarce and privacy-sensitive environments.

GND Keywords
Generative KI
Deep Learning
Datenschutz
Medizin
Keywords
Generative models
Privacy
Foundation models
Robustness
DDC Classification
004 Computer science
RVK Classification
ST 300
Peer Reviewed
yes
International Distribution
yes
Type
Conferenceobject
URI
Activation date
2025-08-06

Permalink: https://fis.uni-bamberg.de/handle/uniba/109551 · Data as JSON · Cite: BibTeX · RIS

Data as of: 2026-11-16

Restricted read-only access: login, editing and submission are temporarily disabled.