mlsanityData to 7 Oct 2026

DP-Image: Differential Privacy for Image Data in Feature Space

arXiv 2103.07073 ↗First posted 12 Mar 2021cs.CR cs.CVcited by 23

Not on the map. Why?

Abstract (arXiv)

The excessive use of images in social networks, government databases, and industrial applications has posed great privacy risks and raised serious concerns from the public. Even though differential privacy (DP) is a widely accepted criterion that can provide a provable privacy guarantee, the application of DP on unstructured data such as images is not trivial due to the lack of a clear qualification on the meaningful difference between any two images. In this paper, for the first time, we introduce a novel notion of image-aware differential privacy, referred to as DP-image, that can protect user's personal information in images, from both human and AI adversaries. The DP-Image definition is formulated as an extended version of traditional differential privacy, considering the distance measurements between feature space vectors of images. Then we propose a mechanism to achieve DP-Image by adding noise to an image feature vector. Finally, we conduct experiments with a case study on face image privacy. Our results show that the proposed DP-Image method provides excellent DP protection on images, with a controllable distortion to faces.

Key facts. Cited by 23 papers in the map (data through 7 Oct 2026). Builds on 26 of its 55 references that are in the map. First posted on arXiv 12 Mar 2021.

Builds on

26 of 55 references.

References · 26

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  4. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  5. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  6. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  7. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,845
  8. 2014Going Deeper with Convolutionscited by 4,231
  9. 2013Auto-Encoding Variational Bayescited by 3,980
  10. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,580
  11. 2014Conditional Generative Adversarial Netscited by 2,152
  12. 2016Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learningcited by 1,509
14 more
  1. 2020Encoding in Style: a StyleGAN Encoder for Image-to-Image Translationcited by 727
  2. 2017Renyi Differential Privacycited by 643
  3. 2018Certified Robustness to Adversarial Examples with Differential Privacycited by 462
  4. 2019FaceShifter: Towards High Fidelity And Occlusion Aware Face Swappingcited by 221
  5. 2017Adversarial Variational Bayes: Unifying Variational Autoencoders and Generative Adversarial Networkscited by 107
  6. 2018Privacy Amplification by Iterationcited by 97
  7. 2020When Machine Learning Meets Privacy: A Survey and Outlookcited by 67
  8. 2017Natural and Effective Obfuscation by Head Inpaintingcited by 60
  9. 2019AnonymousNet: Natural Face De-Identification with Measurable Privacycited by 50
  10. 2016Defeating Image Obfuscation with Deep Learningcited by 43
  11. 2018A Hybrid Model for Identity Obfuscation by Face Replacementcited by 43
  12. 2017Adversarial Image Perturbation for Privacy Protection -- A Game Theory Perspectivecited by 38
  13. 2016Faceless Person Recognition; Privacy Implications in Social Mediacited by 27
  14. 2020InfoScrub: Towards Attribute Privacy by Targeted Obfuscationcited by 10

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 3; 2023: 5; 2024: 4; 2025: 6; 2026: 5.

23 papers.

Sort:

By area:

  1. 2023A Survey of What to Share in Federated Learning: Perspectives on Model Utility, Privacy Leakage, and Communication Efficiencycited by 21
  2. 2023Bounding the Invertibility of Privacy-preserving Instance Encoding using Fisher Informationcited by 11
  3. 2024Face De-identification: State-of-the-art Methods and Comparative Studiescited by 9
  4. 2023Disguise without Disruption: Utility-Preserving Face De-Identificationcited by 8
  5. 2022Local Differential Privacy Image Generation Using Flow-based Deep Generative Modelscited by 5
  6. 2024PrivateGaze: Preserving User Privacy in Black-box Mobile Gaze Tracking Servicescited by 4
  7. 2022Syfer: Neural Obfuscation for Private Data Releasecited by 3
  8. 2025Semantics versus Identity: A Divide-and-Conquer Approach towards Adjustable Medical Image De-Identificationcited by 2
  9. 2023Privacy Protectability: An Information-theoretical Approachcited by 1
  10. 2024Enhancing User-Centric Privacy Protection: An Interactive Framework through Diffusion Models and Machine Unlearningcited by 1
  11. 2024SemDP: Semantic-level Differential Privacy Protection for Face Datasetscited by 1
  12. 2025PASS: Private Attributes Protection with Stochastic Data Substitutioncited by 1
11 more
  1. 2025ROFI: A Deep Learning-Based Ophthalmic Sign-Preserving and Reversible Patient Face Anonymizercited by 1
  2. 2022Contextualize differential privacy in image database: a lightweight image differential privacy approach based on principle component analysis inversecited by 0
  3. 2023Differentially Private Cross-camera Person Re-identificationcited by 0
  4. 2025SpinML: Customized Synthetic Data Generation for Private Training of Specialized ML Modelscited by 0
  5. 2025Computation-Efficient and Recognition-Friendly 3D Point Cloud Privacy Protectioncited by 0
  6. 2025Beyond Anonymization: Object Scrubbing for Privacy-Preserving 2D and 3D Vision Taskscited by 0
  7. 2026Amplified Patch-Level Differential Privacy for Free via Random Croppingcited by 0
  8. 2026Assessing Privacy Preservation and Utility in Online Vision-Language Modelscited by 0
  9. 2026Bit-ViP: Leveraging Bit-planes to Preserve Visual Privacy in Images through Obfuscationcited by 0
  10. 2026End-to-End Differential Privacy in Training Deep Neural Network Classifierscited by 0
  11. 2026Picture the Epsilon: Pursuing Identity-Level Privacy Guarantees for Imagescited by 0