Find target hash collisions for Apple's NeuralHash perceptual hash function.💣

Overview

neural-hash-collider

Find target hash collisions for Apple's NeuralHash perceptual hash function.

For example, starting from a picture of this cat, we can find an adversarial image that has the same hash as the picture of the dog in this post:

python collide.py --image cat.jpg --target 59a34eabe31910abfb06f308

Cat image with NeuralHash 59a34eabe31910abfb06f308 Dog image with NeuralHash 59a34eabe31910abfb06f308

We can confirm the hash collision using nnhash.py from AsuharietYgvar/AppleNeuralHash2ONNX:

$ python nnhash.py dog.png
59a34eabe31910abfb06f308
$ python nnhash.py adv.png
59a34eabe31910abfb06f308

How it works

NeuralHash is a perceptual hash function that uses a neural network. Images are resized to 360x360 and passed through a neural network to produce a 128-dimensional feature vector. Then, the vector is projected onto R^96 using a 128x96 "seed" matrix. Finally, to produce a 96-bit hash, the 96-dimensional vector is thresholded: negative entries turn into a 0 bit, and non-negative entries turn into a 1 bit.

This entire process, except for the thresholding, is differentiable, so we can use gradient descent to find hash collisions. This is a well-known property of neural networks, that they are vulnerable to adversarial examples.

We can define a loss that captures how close an image is to a given target hash: this loss is basically just the NeuralHash algorithm as described above, but with the final "hard" thresholding step tweaked so that it is "soft" (in particular, differentiable). Exactly how this is done (choices of activation functions, parameters, etc.) can affect convergence, so it can require some experimentation. After choosing the loss function, we can follow the standard method to find adversarial examples for neural networks: gradient descent.

Details

The implementation currently does an alternating projections style attack to find an adversarial example that has the intended hash and also looks similar to the original. See collide.py for the full details. The implementation uses two different loss functions: one measures the distance to the target hash, and the other measures the quality of the perturbation (l2 norm + total variation). We first optimize for a collision, focusing only on matching the target hash. Once we find a projection, we alternate between minimizing the perturbation and ensuring that the hash value does not change. The attack has a number of parameters; run python collide.py --help or refer to the code for a full list. Tweaking these parameters can make a big difference in convergence time and the quality of the output.

The implementation also supports a flag --blur [sigma] that blurs the perturbation on every step of the search. This can slow down or break convergence, but on some examples, it can be helpful for getting results that look more natural and less like glitch art.

Examples

Reproducing the Lena/Barbara result from this post:

The first image above is the original Lena image. The second was produced with --target a426dae78cc63799d01adc32 to collide with Barbara. The third was produced with the additional argument --blur 1.0. The fourth is the original Barbara image. Checking their hashes:

$ python nnhash.py lena.png
32dac883f7b91bbf45a48296
$ python nnhash.py lena-adv.png
a426dae78cc63799d01adc32
$ python nnhash.py lena-adv-blur-1.0.png
a426dae78cc63799d01adc32
$ python nnhash.py barbara.png
a426dae78cc63799d01adc32

Reproducing the Picard/Sidious result from this post:

The first image above is the original Picard image. The second was produced with --target e34b3da852103c3c0828fbd1 --tv-weight 3e-4 to collide with Sidious. The third was produced with the additional argument --blur 0.5. The fourth is the original Sidious image. Checking their hashes:

$ python nnhash.py picard.png
73fae120ad3191075efd5580
$ python nnhash.py picard-adv.png
e34b2da852103c3c0828fbd1
$ python nnhash.py picard-adv-blur-0.5.png
e34b2da852103c3c0828fbd1
$ python nnhash.py sidious.png
e34b2da852103c3c0828fbd1

Prerequisites

  • Get Apple's NeuralHash model following the instructions in AsuharietYgvar/AppleNeuralHash2ONNX and either put all the files in this directory or supply the --model / --seed arguments
  • Install Python dependencies: pip install -r requirements.txt

Usage

Run python collide.py --image [path to image] --target [target hash] to generate a hash collision. Run python collide.py --help to see all the options, including some knobs you can tweak, like the learning rate and some other parameters.

Limitations

The code in this repository is intended to be a demonstration, and perhaps a starting point for other exploration. Tweaking the implementation (choice of loss function, choice of parameters, etc.) might produce much better results than this code currently achieves.

Owner
Anish Athalye
grad student @mit-pdos
Anish Athalye
Rembg is a tool to remove images background.

Rembg is a tool to remove images background.

Daniel Gatis 7.8k Jan 05, 2023
🛹 Turn an SVG into an STL for stencil creation purposes

svg2stl This repository provides a script which takes as input an SVG such as this one: It outputs an STL file like this one: You can also see an inte

Max Halford 3 Dec 29, 2021
Fast Image Retrieval (FIRe) is an open source image retrieval project

Fast Image Retrieval (FIRe) is an open source image retrieval project release by Center of Image and Signal Processing Lab (CISiP Lab), Universiti Malaya. This project implements most of the major bi

CISiP Lab 39 Nov 25, 2022
Generative Art Synthesizer - a python program that generates python programs that generates generative art

GAS - Generative Art Synthesizer Generative Art Synthesizer - a python program that generates python programs that generates generative art. Examples

Alexey Borsky 43 Dec 03, 2022
Tools for making image cutouts from sets of TESS full frame images

Cutout tools for astronomical images Astrocut provides tools for making cutouts from sets of astronomical images with shared footprints. It is under a

Space Telescope Science Institute 20 Dec 16, 2022
Parking management project which generates barcode parking ticket with user-friendly Tkinter program GUI

Parking-management-system Parking management project which generates barcode parking ticket with user-friendly Tkinter program GUI How to run Download

1 Jul 03, 2022
This tool allows the user to convert a 16 by 16 image into a list with numbers representing an object/character.

Room Formatter This tool allows the user to convert a 16 by 16 image into a list with numbers representing an object/character. There is cur

Thomas Landstra 1 Nov 13, 2021
A Python package implementing various CFA (Colour Filter Array) demosaicing algorithms and related utilities.

Colour - Demosaicing A Python package implementing various CFA (Colour Filter Array) demosaicing algorithms and related utilities. It is open source a

colour-science 218 Dec 04, 2022
This script is for photographers to do timeslice with one click.

One Click TimeSlice Tool What is this for This is for photographers who want to create TimeSlice pictures without installing PS plugins. Before using

Xi Zhao 13 Sep 23, 2022
Convert photos to paintings with python

Convert-photos-to-paintings Before the changes After the changes Before the changes After the changes This code is written in the Python programming l

Amir Hussein Sharifnezhad 3 May 31, 2022
Image manipulation package used for EpicBot.

Image manipulation package used for EpicBot.

Nirlep_5252_ 7 May 26, 2022
PIX is an image processing library in JAX, for JAX.

PIX PIX is an image processing library in JAX, for JAX. Overview JAX is a library resulting from the union of Autograd and XLA for high-performance ma

DeepMind 294 Jan 08, 2023
Png2Jpg tool will help you convert from png image format to jpg images format.

PNG 2 JPG All codes assume running from root directory. Please update the sys path at the beginning of the codes before running. Over View Png2Jpg too

Nguyễn Trường Lâu 2 Dec 27, 2021
Napari plugin for iteratively improving 3D instance segmentation of cells (u-net x watershed)

iterseg napari plugin for iteratively improving unet-watershed segmentation This napari plugin was generated with Cookiecutter using @napari's cookiec

Abigail McGovern 3 May 16, 2022
Create a static HTML/CSS image gallery from a bunch of images.

gallerize Create a static HTML/CSS image gallery from a bunch of images.

Jochen Kupperschmidt 19 Aug 21, 2022
ModernGL is a python wrapper over OpenGL 3.3+ core

ModernGL is a python wrapper over OpenGL 3.3+ core that simplifies the creation of simple graphics applications like scientific simulations, games or user interface

ModernGL 1.4k Jan 01, 2023
Nanosensor Image Processor (NanoImgPro), a python-based image analysis tool for dopamine nanosensors

NanoImgPro Nanosensor Image Processor (NanoImgPro), a python-based image analysis tool for dopamine nanosensors NanoImgPro.py contains the main class

1 Mar 02, 2022
Snowfall - helpful image handling utils - abstracts various file and opencv and pil features into result oriented functions

snowfall helpful image handling utils - abstracts various file and opencv and pil features into result oriented functions usage examples: from image_h

Less Wright 2 Jan 09, 2022
Extracts random colours from an image

EXTRACT COLOURS This repo contains all the project files. Project Description A Program that extracts 10 random colours from an image and saves the rg

David .K. Danso 3 Dec 13, 2021
Paper backup of files using QR codes

Generate paper backups for Linux. Currently command-linux Linux only. Takes any file, and outputs a "paper backup": a printable black-and-white pdf fu

Zachary Vance 27 Dec 28, 2022