Bai, Fenhua, Yang, Xianlin, Shen, Tao, Gong, Bei, Waqas, Muhammad ORCID: https://orcid.org/0000-0003-0814-7544 and Alfuhaid, Hisham (2026) Two-level IVC: scalable zero-knowledge proof for trustworthy video stream verification. IEEE Transactions on Circuits and Systems for Video Technology (TCSVT). ISSN 1051-8215 (Print), 1558-2205 (Online) (doi:10.1109/TCSVT.2026.3720431)

Preview

PDF (Author's Accepted Manuscript) 54012 WAQAS_Two-Level_IVC_Scalable_Zero-Knowledge_Proof_(AAM)_2026.pdf - Accepted Version Available under License Creative Commons Attribution. Download (1MB) | Preview

Abstract

In content-sensitive applications such as video journalism, forensic video authentication, and digital media provenance verification, there is a growing demand for AI analysis of video streams. However, this creates a profound contradiction between data privacy protection and trust in computational processes. Existing solutions struggle to provide scalable cryptographic proofs with temporal integrity for AI computational processes involving complex non-linear operations while simultaneously protecting video content privacy. To address this challenge, this paper proposes an end-to-end Two-Level Incremental Verifiable Computation (IVC) architecture that verifies both intra-frame spatial computation and inter-frame temporal continuity. The architecture’s pluggable design instantiates multiple Verifiable Computational Modules (VCM) ranging from pixel-wise transformations and spatiotemporal convolution to Vision Transformer inference, without altering the recursive outer structure. At the hardware deployment level, an adjustable block size mechanism accommodates specific hardware resource constraints by flexibly trading proof generation time against peak memory consumption, achieving constant O(1) peak memory overhead with respect to video length on commodity hardware. Additionally, our framework incorporates the Coalition for Content Provenance and Authenticity (C2PA) standard, ensuring complete sensor-to-proof traceability without requiring hardware trust at the processing level.

Item Type:	Article
Uncontrolled Keywords:	privacy preserving video stream processing, zero-knowledge proof, Incremental Verifiable Computation, temporal integrity, C2PA content provenance
Subjects:	Q Science > Q Science (General) Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Faculty / School / Research Centre / Research Group:	Faculty of Engineering & Science Faculty of Engineering & Science > School of Computing & Mathematical Sciences (CMS)
Last Modified:	10 Aug 2026 10:03
URI:	https://gala.gre.ac.uk/id/eprint/54012

Actions (login required)

View Item

Downloads