Preprint · 2023 · Open access
PrivImage: Differentially Private Synthetic Image Generation using Diffusion Models with Semantic-Aware Pretraining
arXiv (Cornell University) · Published 19 October 2023 · DOI 10.48550/arxiv.2311.12850
In brief
“PrivImage: Differentially Private Synthetic Image Generation using Diffusion Models with Semantic-Aware Pretraining” is a 2023 preprint by Kecen Li and 5 co-authors, published in arXiv (Cornell University). OpenAlex counts 25 citations as of 19 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).
- Citations
- 25
- Most cited in
- 2025 (13)
- References
- 58
- Authors
- 6
Abstract
Differential Privacy (DP) image data synthesis, which leverages the DP technique to generate synthetic data to replace the sensitive data, allowing organizations to share and utilize synthetic images without privacy concerns. Previous methods incorporate the advanced techniques of generative models and pre-training on a public dataset to produce exceptional DP image data, but suffer from problems of unstable training and massive computational resource demands. This paper proposes a novel DP image synthesis method, termed PRIVIMAGE, which meticulously selects pre-training data, promoting the efficient creation of DP datasets with high fidelity and utility. PRIVIMAGE first establishes a semantic query function using a public dataset. Then, this function assists in querying the semantic distribution of the sensitive dataset, facilitating the selection of data from the public dataset with analogous semantics for pre-training. Finally, we pre-train an image generative model using the selected data and then fine-tune this model on the sensitive dataset using Differentially Private Stochastic Gradient Descent (DP-SGD). PRIVIMAGE allows us to train a lightly parameterized generative model, reducing the noise in the gradient during DP-SGD training and enhancing training stability. Extensive experiments demonstrate that PRIVIMAGE uses only 1% of the public dataset for pre-training and 7.6% of the parameters in the generative model compared to the state-of-the-art method, whereas achieves superior synthetic performance and conserves more computational resources. On average, PRIVIMAGE achieves 30.1% lower FID and 12.6% higher Classification Accuracy than the state-of-the-art method. The replication package and datasets can be accessed online.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Li, K., Gong, C., Li, Z., Zhao, Y., Hou, X., & Wang, T. (2023). PrivImage: Differentially Private Synthetic Image Generation using Diffusion Models with Semantic-Aware Pretraining. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2311.12850
MLA 9
Li, Kecen, et al. “PrivImage: Differentially Private Synthetic Image Generation using Diffusion Models with Semantic-Aware Pretraining.” arXiv (Cornell University), 2023. https://doi.org/10.48550/arxiv.2311.12850.
Chicago
Li, Kecen, Chen Gong, Zhixiang Li, Yuzhong Zhao, Xinwen Hou, and Tianhao Wang. 2023. “PrivImage: Differentially Private Synthetic Image Generation using Diffusion Models with Semantic-Aware Pretraining.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2311.12850.
Harvard
Li, K. et al. (2023) ‘PrivImage: Differentially Private Synthetic Image Generation using Diffusion Models with Semantic-Aware Pretraining’, arXiv (Cornell University). doi:10.48550/arxiv.2311.12850.
IEEE
K. Li, C. Gong, Z. Li, Y. Zhao, X. Hou, and T. Wang, “PrivImage: Differentially Private Synthetic Image Generation using Diffusion Models with Semantic-Aware Pretraining,” arXiv (Cornell University), 2023, doi: 10.48550/arxiv.2311.12850.
BibTeX
@misc{li2023privimage,
title = {{PrivImage: Differentially Private Synthetic Image Generation using Diffusion Models with Semantic-Aware Pretraining}},
author = {Li, Kecen and Gong, Chen and Li, Zhixiang and Zhao, Yuzhong and Hou, Xinwen and Wang, Tianhao},
howpublished = {arXiv (Cornell University)},
year = {2023},
doi = {10.48550/arxiv.2311.12850},
url = {https://doi.org/10.48550/arxiv.2311.12850}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Citations by year
| Year | Citations | Share of the best year |
|---|---|---|
| 2024 | 4 | |
| 2025 | 13 | |
| 2026 | 8 |
Topic and keywords
Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- differential privacy
- synthetic image generation
- diffusion models
- DP-SGD
- privacy-preserving data synthesis
- training stability
- computational efficiency
Work with this paper
Most-cited papers in Computer Science
- ImageNet classification with deep convolutional neural networksAlex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton · Communications of the ACM · 2017 · 114,726 citations
- Long Short-Term MemorySepp Hochreiter, Jürgen Schmidhuber · Neural Computation · 1997 · 109,097 citations
- A Mathematical Theory of CommunicationClaude E. Shannon · Bell System Technical Journal · 1948 · 91,736 citations
- Deep learningYann LeCun, Yoshua Bengio, Geoffrey E. Hinton · Nature · 2015 · 90,226 citations
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksShaoqing Ren, Kaiming He, Ross Girshick and 1 more · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2016 · 74,143 citations
Questions about this paper
How do I cite this paper in APA style?
Li, K., Gong, C., Li, Z., Zhao, Y., Hou, X., & Wang, T. (2023). PrivImage: Differentially Private Synthetic Image Generation using Diffusion Models with Semantic-Aware Pretraining. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2311.12850
How many times has this paper been cited?
25 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.
Is this paper free to read?
Yes. It is open access at arXiv (Cornell University): sign in and open the PDF from this page.
Who wrote this paper?
Kecen Li, Chen Gong, Zhixiang Li, Yuzhong Zhao, Xinwen Hou, Tianhao Wang.
What is the DOI of this paper?
10.48550/arxiv.2311.12850, permanently at https://doi.org/10.48550/arxiv.2311.12850. Sign in and open it from this page.
Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.