PDQ & TMK + PDQF -- A Test Drive of Facebook's Perceptual Hashing Algorithms

TL;DR

This paper evaluates Facebook's open-sourced PDQ and TMK + PDQF algorithms for detecting modified images and videos, highlighting their effectiveness in real-world law enforcement scenarios and providing a reference implementation.

Contribution

It offers a performance review of Facebook's perceptual hashing algorithms on real-world data and demonstrates their integration potential in law enforcement workflows.

Findings

High accuracy in detecting common image transformations

Effective video similarity measurement in real-world cases

Practical reference implementation provided

Abstract

Efficient and reliable automated detection of modified image and multimedia files has long been a challenge for law enforcement, compounded by the harm caused by repeated exposure to psychologically harmful materials. In August 2019 Facebook open-sourced their PDQ and TMK + PDQF algorithms for image and video similarity measurement, respectively. In this report, we review the algorithms' performance on detecting commonly encountered transformations on real-world case data, sourced from contemporary investigations. We also provide a reference implementation to demonstrate the potential application and integration of such algorithms within existing law enforcement systems.