REVIEW 5 cited by
DP-Image: Differential Privacy for Image Data in Feature Space
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
DP-Image: Differential Privacy for Image Data in Feature Space
read the original abstract
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.
Forward citations
Cited by 5 Pith papers
-
Contrastive Privacy: A Semantic Approach to Measuring Privacy of AI-based Sanitization
Contrastive privacy is a new corpus-contrast test for semantic privacy in AI-sanitized media that uses latent concept measures and requires no manual labeling.
-
Privacy-Preserving Semantic Communication over Wiretap Channels with Learnable Differential Privacy
A framework that perturbs private semantic representations with learnable DP noise via GAN inversion and adversarial training to secure image SemCom over wiretap channels with tunable privacy levels.
-
End-to-End Differential Privacy in Training Deep Neural Network Classifiers
Perturbing softmax outputs with the Dirichlet mechanism during training yields input-private, label-public classifiers that beat prior differentially private training accuracy on five image benchmarks.
-
Bit-ViP: Leveraging Bit-planes to Preserve Visual Privacy in Images through Obfuscation
Bit-ViP applies bit-plane processing with Lorenz chaotic noise and differential privacy to produce obfuscated images that resist reconstruction attacks yet support activity recognition on UCF101 and HMDB51.
-
Assessing Privacy Preservation and Utility in Online Vision-Language Models
The work proposes and evaluates techniques to reduce PII exposure from image context in online vision-language models while preserving utility for downstream applications.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.