Exploiting and Defending Against the Approximate Linearity of Apple's NeuralHash

@article{Bhatia2022ExploitingAD,
  title={Exploiting and Defending Against the Approximate Linearity of Apple's NeuralHash},
  author={Jagdeep Bhatia and Kevin Meng},
  journal={ArXiv},
  year={2022},
  volume={abs/2207.14258}
}
Perceptual hashes map images with identical semantic content to the same n -bit hash value, while mapping semantically-different images to different hashes. These algorithms carry important ap-plications in cybersecurity such as copyright in-fringement detection, content fingerprinting, and surveillance. Apple’s N EURAL H ASH is one such system that aims to detect the presence of illegal content on users’ devices without compromising consumer privacy. We make the surprising discovery that N… 

References

SHOWING 1-10 OF 18 REFERENCES

Learning to Break Deep Perceptual Hashing: The Use Case NeuralHash

It is shown that current deep perceptual hashing may not be robust, generally not ready for robust client-side scanning and should not be used from a privacy perspective.

Near Duplicate Image Detection: min-Hash and tf-idf Weighting

This paper proposes two novel image similarity measures for fast indexing via locality sensitive hashing and an efficient way of exploiting more sophisticated similarity measures that have proven to be essential in image / particular object retrieval.

Adversarial Examples Are Not Bugs, They Are Features

It is demonstrated that adversarial examples can be directly attributed to the presence of non-robust features: features derived from patterns in the data distribution that are highly predictive, yet brittle and incomprehensible to humans.

Explaining and Harnessing Adversarial Examples

It is argued that the primary cause of neural networks' vulnerability to adversarial perturbation is their linear nature, supported by new quantitative results while giving the first explanation of the most intriguing fact about them: their generalization across architectures and training sets.

The Apple PSI System

The constraints that drove the design of the Apple private set intersection (PSI) protocol are described, a protocol that satisfies the constraints, and its security is analyzed.

Image Matching by Using Fuzzy Transforms

The concept of Fuzzy Transform (for short, F-transform) is applied for improving the results of the image matching based on the Greatest Eigen FuzzY Set with respect tomax-min composition and the Smallest Eigen fuzzy Set with Respect to min-max composition already studied in the literature.

MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications

This work introduces two simple global hyper-parameters that efficiently trade off between latency and accuracy and demonstrates the effectiveness of MobileNets across a wide range of applications and use cases including object detection, finegrain classification, face attributes and large scale geo-localization.

A genetic algorithm tutorial

This tutorial covers the canonical genetic algorithm as well as more experimental forms of genetic algorithms, including parallel island models and parallel cellular genetic algorithms. The tutorial

Imagenette: A subset of imagenet

  • 2019

The pdq perceptual hash

  • 2020