Mohammad Amin Arab

Mohammad Amin Arab

Research Assistant @ Network Systems Lab -- Simon Fraser University

Posts by Collection

portfolio

publications

Band and Quality Selection for Efficient Transmission of Hyperspectral Images

Published in ACM Multimedia, 2019

This paper presents a method to prioritize the transmission of various components of hyperspectral data based on the application needs, the level of details required, and available bandwidth.

Recommended citation: M. Arab, et al. 2019. Band and Quality Selection for Efficient Transmission of Hyperspectral Images. In Proceedings of the 27th ACM International Conference on Multimedia. https://dl.acm.org/doi/10.1145/3343031.3351047

Revealing True Identity: Detecting Makeup Attacks in Face-based Biometric Systems

Published in ACM Multimedia, 2020

This paper proposes a novel solution to address makeup attacks, which are the hardest to detect in such systems because makeup can substantially alter the facial features of a person, including making them appear older/younger by adding/hiding wrinkles, modifying the shape of eyebrows, beard, and moustache, and changing the color of lips and cheeks.

Recommended citation: M. Arab, et al. 2020. Revealing True Identity: Detecting Makeup Attacks in Face-based Biometric Systems. In Proceedings of the 28th ACM International Conference on Multimedia. https://dl.acm.org/doi/10.1145/3394171.3413606

FlexMark: Adaptive Watermarking Method for Images

Published in ACM Multimedia Systems, 2024

We proposed a flexible watermarking method that can be used in different scenarios from copyright protection to data hiding.

Recommended citation: M. Arab, et al. 2024. FlexMark: Adaptive Watermarking Method for Images. In Proceedings of the 15th ACM Multimedia Systems Conference.

RDIAS: Robust and decentralized image authentication system

Published in ACM Transactions on Multimedia Computing, Communications and Applications, 2025

This paper presents RDIAS, a robust and decentralized image authentication system designed to withstand sophisticated AI-powered manipulations while tolerating common platform transformations such as resizing and transcoding. RDIAS embeds secure, semantic fingerprints directly into images using perceptual hashing, deep-learning-based watermarking, and error-correcting codes. Extensive evaluations with realistic AI attacks demonstrate up to 99% accuracy in detecting manipulated images while preserving visual quality and enabling real-time verification.

Recommended citation: Ghorbanpour, Ali, Mohammad Amin Arab, and Mohamed Hefeeda. "RDIAS: Robust and decentralized image authentication system." ACM Transactions on Multimedia Computing, Communications and Applications 21.10 (2025): 1-28.

ImageShield: A System for Image Tampering Detection, Localization, and Recovery

Published in , 1900

This paper introduces ImageShield, an end-to-end system for image authentication, tampering localization, and semantic content recovery that is robust against advanced AI-enabled attacks. ImageShield employs a novel dual-embedding strategy that separately embeds cryptographically protected authentication data and high-capacity recovery information using specialized neural networks. Extensive experiments and a human user study demonstrate that ImageShield significantly outperforms prior state-of-the-art methods in robustness, localization accuracy, and restored image quality under realistic attacks and transformations.

talks

teaching

Teaching experience 1

Undergraduate course, University 1, Department, 2014

This is a description of a teaching experience. You can use markdown like any other post.

Teaching experience 2

Workshop, University 1, Department, 2015

This is a description of a teaching experience. You can use markdown like any other post.