Convert Apple NeuralHash model for CSAM Detection to ONNX.

Overview

AppleNeuralHash2ONNX

Convert Apple NeuralHash model for CSAM Detection to ONNX.

Intro

Apple NeuralHash is a perceptual hashing method for images based on neural networks. It can tolerate image resize and compression. The steps of hashing is as the following:

  1. Convert image to RGB.
  2. Resize image to 360x360.
  3. Normalize RGB values to [-1, 1] range.
  4. Perform inference on the NeuralHash model.
  5. Calculate dot product of a 96x128 matrix with the resulting vector of 128 floats.
  6. Apply binary step to the resulting 96 float vector.
  7. Convert the vector of 1.0 and 0.0 to bits, resulting in 96-bit binary data.

In this project, we convert Apple's NeuralHash model to ONNX format. A demo script for testing the model is also included.

Prerequisite

OS

Both macOS and Linux will work. In the following sections Debian is used for Linux example.

LZFSE decoder

  • macOS: Install by running brew install lzfse.
  • Linux: Build and install from lzfse source.

Python

Python 3.6 and above should work. Install the following dependencies:

pip install onnx coremltools

Conversion Guide

Step 1: Get NeuralHash model

You will need 4 files from a recent macOS or iOS build:

  • neuralhash_128x96_seed1.dat
  • NeuralHashv3b-current.espresso.net
  • NeuralHashv3b-current.espresso.shape
  • NeuralHashv3b-current.espresso.weights

Option 1: From macOS or jailbroken iOS device (Recommended)

If you have a recent version of macOS (11.4+) or jailbroken iOS (14.7+) installed, simply grab these files from /System/Library/Frameworks/Vision.framework/Resources/ (on macOS) or /System/Library/Frameworks/Vision.framework/ (on iOS).

Option 2: From iOS IPSW (click to reveal)
  1. Download any .ipsw of a recent iOS build (14.7+) from ipsw.me.
  2. Unpack the file:
cd /path/to/ipsw/file
mkdir unpacked_ipsw
cd unpacked_ipsw
unzip ../*.ipsw
  1. Locate system image:
ls -lh

What you need is the largest .dmg file, for example 018-63036-003.dmg.

  1. Mount system image. On macOS simply open the file in Finder. On Linux run the following commands:
# Build and install apfs-fuse
sudo apt install fuse libfuse3-dev bzip2 libbz2-dev cmake g++ git libattr1-dev zlib1g-dev
git clone https://github.com/sgan81/apfs-fuse.git
cd apfs-fuse
git submodule init
git submodule update
mkdir build
cd build
cmake ..
make
sudo make install
sudo ln -s /bin/fusermount /bin/fusermount3
# Mount image
mkdir rootfs
apfs-fuse 018-63036-003.dmg rootfs

Required files are under /System/Library/Frameworks/Vision.framework/ in mounted path.

Put them under the same directory:

mkdir NeuralHash
cd NeuralHash
cp /System/Library/Frameworks/Vision.framework/Resources/NeuralHashv3b-current.espresso.* .
cp /System/Library/Frameworks/Vision.framework/Resources/neuralhash_128x96_seed1.dat .

Step 2: Decode model structure and shapes

Normally compiled Core ML models store structure in model.espresso.net and shapes in model.espresso.shape, both in JSON. It's the same for NeuralHash model but compressed with LZFSE.

dd if=NeuralHashv3b-current.espresso.net bs=4 skip=7 | lzfse -decode -o model.espresso.net
dd if=NeuralHashv3b-current.espresso.shape bs=4 skip=7 | lzfse -decode -o model.espresso.shape
cp NeuralHashv3b-current.espresso.weights model.espresso.weights

Step 3: Convert model to ONNX

cd ..
git clone https://github.com/AsuharietYgvar/TNN.git
cd TNN
python3 tools/onnx2tnn/onnx-coreml/coreml2onnx.py ../NeuralHash

The resulting model is NeuralHash/model.onnx.

Usage

Inspect model

Netron is a perfect tool for this purpose.

Calculate neural hash with onnxruntime

  1. Install required libraries:
pip install onnxruntime pillow
  1. Run nnhash.py on an image:
python3 nnhash.py /path/to/model.onnx /path/to/neuralhash_128x96_seed1.dat image.jpg

Example output:

ab14febaa837b6c1484c35e6

Note: Neural hash generated here might be a few bits off from one generated on an iOS device. This is expected since different iOS devices generate slightly different hashes anyway. The reason is that neural networks are based on floating-point calculations. The accuracy is highly dependent on the hardware. For smaller networks it won't make any difference. But NeuralHash has 200+ layers, resulting in significant cumulative errors.

Device Hash
iPad Pro 10.5-inch 2b186faa6b36ffcc4c4635e1
M1 Mac 2b5c6faa6bb7bdcc4c4731a1
iOS Simulator 2b5c6faa6bb6bdcc4c4731a1
ONNX Runtime 2b5c6faa6bb6bdcc4c4735a1

Credits

  • nhcalc for uncovering NeuralHash private API.
  • TNN for compiled Core ML to ONNX script.
Owner
Asuhariet Ygvar
Asuhariet Ygvar
[NeurIPS 2021] A weak-shot object detection approach by transferring semantic similarity and mask prior.

[NeurIPS 2021] A weak-shot object detection approach by transferring semantic similarity and mask prior.

BCMI 49 Jul 27, 2022
Coarse implement of the paper "A Simultaneous Denoising and Dereverberation Framework with Target Decoupling", On DNS-2020 dataset, the DNSMOS of first stage is 3.42 and second stage is 3.47.

SDDNet Coarse implement of the paper "A Simultaneous Denoising and Dereverberation Framework with Target Decoupling", On DNS-2020 dataset, the DNSMOS

Cyril Lv 43 Nov 21, 2022
Lightweight library to build and train neural networks in Theano

Lasagne Lasagne is a lightweight library to build and train neural networks in Theano. Its main features are: Supports feed-forward networks such as C

Lasagne 3.8k Dec 29, 2022
This repository is an official implementation of the paper MOTR: End-to-End Multiple-Object Tracking with TRansformer.

MOTR: End-to-End Multiple-Object Tracking with TRansformer This repository is an official implementation of the paper MOTR: End-to-End Multiple-Object

348 Jan 07, 2023
Code for the RA-L (ICRA) 2021 paper "SeqNet: Learning Descriptors for Sequence-Based Hierarchical Place Recognition"

SeqNet: Learning Descriptors for Sequence-Based Hierarchical Place Recognition [ArXiv+Supplementary] [IEEE Xplore RA-L 2021] [ICRA 2021 YouTube Video]

Sourav Garg 63 Dec 12, 2022
Code for visualizing the loss landscape of neural nets

Visualizing the Loss Landscape of Neural Nets This repository contains the PyTorch code for the paper Hao Li, Zheng Xu, Gavin Taylor, Christoph Studer

Tom Goldstein 2.2k Jan 09, 2023
A model to classify a piece of news as REAL or FAKE

Fake_news_classification A model to classify a piece of news as REAL or FAKE. This python project of detecting fake news deals with fake and real news

Gokul Stark 1 Jan 29, 2022
QICK: Quantum Instrumentation Control Kit

QICK: Quantum Instrumentation Control Kit The QICK is a kit of firmware and software to use the Xilinx RFSoC to control quantum systems. It consists o

81 Dec 15, 2022
Gapmm2: gapped alignment using minimap2 (align transcripts to genome)

gapmm2: gapped alignment using minimap2 This tool is a wrapper for minimap2 to r

Jon Palmer 2 Jan 27, 2022
A simple baseline for 3d human pose estimation in tensorflow. Presented at ICCV 17.

3d-pose-baseline This is the code for the paper Julieta Martinez, Rayat Hossain, Javier Romero, James J. Little. A simple yet effective baseline for 3

Julieta Martinez 1.3k Jan 03, 2023
A MatConvNet-based implementation of the Fully-Convolutional Networks for image segmentation

MatConvNet implementation of the FCN models for semantic segmentation This package contains an implementation of the FCN models (training and evaluati

VLFeat.org 175 Feb 18, 2022
A FAIR dataset of TCV experimental results for validating edge/divertor turbulence models.

TCV-X21 validation for divertor turbulence simulations Quick links Intro Welcome to TCV-X21. We're glad you've found us! This repository is designed t

0 Dec 18, 2021
Research code of ICCV 2021 paper "Mesh Graphormer"

MeshGraphormer ✨ ✨ This is our research code of Mesh Graphormer. Mesh Graphormer is a new transformer-based method for human pose and mesh reconsructi

Microsoft 251 Jan 08, 2023
SweiNet is an uncertainty-quantifying shear wave speed (SWS) estimator for ultrasound shear wave elasticity (SWE) imaging.

SweiNet SweiNet is an uncertainty-quantifying shear wave speed (SWS) estimator for ultrasound shear wave elasticity (SWE) imaging. SweiNet takes as in

Felix Jin 3 Mar 31, 2022
Episodic-memory - Ego4D Episodic Memory Benchmark

Ego4D Episodic Memory Benchmark EGO4D is the world's largest egocentric (first p

3 Feb 18, 2022
Py-FEAT: Python Facial Expression Analysis Toolbox

Py-FEAT is a suite for facial expressions (FEX) research written in Python. This package includes tools to detect faces, extract emotional facial expressions (e.g., happiness, sadness, anger), facial

Computational Social Affective Neuroscience Laboratory 147 Jan 06, 2023
An implementation of EWC with PyTorch

EWC.pytorch An implementation of Elastic Weight Consolidation (EWC), proposed in James Kirkpatrick et al. Overcoming catastrophic forgetting in neural

Ryuichiro Hataya 166 Dec 22, 2022
Attention-driven Robot Manipulation (ARM) which includes Q-attention

Attention-driven Robotic Manipulation (ARM) This codebase is home to: Q-attention: Enabling Efficient Learning for Vision-based Robotic Manipulation I

Stephen James 84 Dec 29, 2022
Scalable, Portable and Distributed Gradient Boosting (GBDT, GBRT or GBM) Library, for Python, R, Java, Scala, C++ and more. Runs on single machine, Hadoop, Spark, Dask, Flink and DataFlow

eXtreme Gradient Boosting Community | Documentation | Resources | Contributors | Release Notes XGBoost is an optimized distributed gradient boosting l

Distributed (Deep) Machine Learning Community 23.6k Dec 31, 2022
Model Zoo of BDD100K Dataset

Model Zoo of BDD100K Dataset

ETH VIS Group 200 Dec 27, 2022