LmCast :: Stay tuned in

ZK-JPEG: Zero-Knowledge Image Editing and Compression

Recorded: Sept. 19, 2026, 8 p.m.

Original Summarized

ZK-JPEG: Zero-knowledge Image Editing and Compression

What a lovely hat
Is it made out of tin foil?

Cryptology ePrint Archive

Papers

Updates from the last:
7 days
31 days
6 months
365 days

Listing by year
All papers
Compact view
Subscribe

How to cite

Harvesting metadata

Submissions

Submit a paper
Revise or withdraw a paper
Acceptance and publishing conditions

About

Goals and history
News
Statistics
Contact

Search

Advanced search

Paper 2026/2039
ZK-JPEG: Zero-knowledge Image Editing and Compression
Samuel Dittmer, Stealth Software Technologies, Inc.
Steve Lu, Stealth Software Technologies, Inc.
Kimberlee Model, Stealth Software Technologies, Inc.
Joseph Near, University of Vermont
Abstract
Tools for generating deep fake photographs are proliferating with greater ease of use and prominence in pop culture. Image authentication tools can defeat these deceitful developments by verifying that a digital image was actually produced by a physical camera. The challenge is that these tools must be robust to desirable image transformations.

Camera attestation uses digital signatures to prove an image's provenance from a camera. Lossy compression makes minute changes in order to reduce an image's size, and blurring or redacting regions of an image can protect its subjects. These changes invalidate an image's signature. Prior works use zero-knowledge (ZK) to prove a published image's edit history, but they do not survive lossy encoding such as the JPEG format.

We present \zkjpeg, a cryptographic tool for JPEG compression that proves an image was correctly compressed from a secret, committed input. In addition, our tool can verify a large family of image transformations by integrating them into JPEG compression with minimal cost.

Our system is fast, flexible, and can be instantiated from off-the-shelf ZK tools. We use PicoZK to convert Python image editing code into a ZK circuit for the line-point zero knowledge (LPZK) proof system.

Metadata

Available format(s)

PDF

Category
Applications
Publication info
Published elsewhere. Minor revision. SCN 2026
Keywords
Zero-knowledge proofsimage provenanceverifiable image editing
Contact author(s)

sam @ stealthsoftwareinc comsteve @ stealthsoftwareinc comkimee @ stealthsoftwareinc comjnear @ uvm edu

History
2026-09-17: approved
2026-09-15: received
See all versions
Short URL
https://ia.cr/2026/2039
License

CC BY

BibTeX
Copy to clipboard

@misc{cryptoeprint:2026/2039,
author = {Samuel Dittmer and Steve Lu and Kimberlee Model and Joseph Near},
title = {{ZK}-{JPEG}: Zero-knowledge Image Editing and Compression},
howpublished = {Cryptology {ePrint} Archive, Paper 2026/2039},
year = {2026},
url = {https://eprint.iacr.org/2026/2039}
}

Note: In order to protect the privacy of readers, eprint.iacr.org
does not use cookies or embedded third party content.

The proposed work introduces zkjjpeg, a cryptographic tool designed to facilitate JPEG compression while proving the integrity of the image data. This system addresses the challenge of ensuring image provenance by verifying that a digital image originated from a physical camera, a necessity in an environment where deep fake photographs are increasingly prevalent through sophisticated editing tools. Camera attestation typically relies on digital signatures to establish provenance, but these signatures are invalidated by common image transformations, such as lossy compression inherent in the JPEG format, as well as blurring or redaction of image regions. Previous research utilizing zero-knowledge proofs has managed to prove the edit history of published images, but these methods do not maintain validity when subjected to lossy encoding formats like JPEG.

To overcome this limitation, the authors developed zkjjpeg to cryptographically prove that an image has been correctly compressed from a secretly committed input. Furthermore, the system is engineered to efficiently verify a wide range of image transformations by integrating these transformations directly into the JPEG compression process with minimal computational overhead. The system is characterized by being fast and flexible, allowing for instantiation using existing zero-knowledge tools. Specifically, the implementation leverages PicoZK to translate Python-based image editing code into a zero-knowledge circuit utilizing the line-point zero knowledge (LPZK) proof system. This integration allows for the verification of complex editing and compression operations within the constraints of the JPEG structure, establishing a verifiable link between the original data, the transformations applied, and the final compressed image.