AppleNeuralHash2ONNX
nhcalc
Our great sponsors
AppleNeuralHash2ONNX | nhcalc | |
---|---|---|
107 | 4 | |
1,517 | 127 | |
- | - | |
0.0 | 3.2 | |
over 2 years ago | over 2 years ago | |
Python | Swift | |
Apache License 2.0 | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
AppleNeuralHash2ONNX
- Legit app in Google Play turns malicious and sends mic recordings every 15 minutes
- Daily General Discussion - October 27, 2022
-
How did apple train its cp filtering algorithm?
Perhaps you're referring to the hash collisions that were generated for the client-side NeuralHash which was reverse engineered about a year ago: https://github.com/AsuharietYgvar/AppleNeuralHash2ONNX/issues/1
- [Request] A way to remove Apple’s new NeuralHash ( iCloud CSAM scanner )
-
Google AI flags parents' accounts for potential abuse over kid's photos
Except Apple’s hashes have collisions where images that are similar in content but not identical will have the same hash. Meaning their process will have the same result: benign images being flagged as CSAM. See an example here https://github.com/AsuharietYgvar/AppleNeuralHash2ONNX/issues/1
- Apple Remains Silent About Plans to Detect Known CSAM Stored in iCloud Photos
-
Apple should scan iPhones for child abuse images, says scanning technology inventor
perceptual hashes are completely broken due to false positives, like making a picture of a dog having the same hash as CSAM images. They are vulnerable to collision and have no preimage resistance, the most important feature. Because of that you can DOS conversations and people by sending dog images with the same hash as a CSAM: https://github.com/AsuharietYgvar/AppleNeuralHash2ONNX/issues/1
-
UK cybersecurity chiefs back plan to scan phones for child abuse images | GCHQ | The Guardian
Specifically the fact that you can have one hash on two images, for example. There are other ways to abuse the CSAM system, which is confirmed by several researchers (one, two). And Apple saying it's not a concern... only to pull back on these plans.
- Frage zur drohenden Chatkontrolle
- Apple’s CSAM troubles may be back, as EU plans a law requiring detection
nhcalc
-
[P] AppleNeuralHash2ONNX: Reverse-Engineered Apple NeuralHash, in ONNX and Python
The hidden APIs were found by someone else here. I'm not going to talk about the reverse-engineering process in too much detail. Basically what I did was to use Xcode debugger+Hopper disassembler+LLDB commands trying to understand how the function works under the hood in assembly code (which was very tedious). There were some parts that I didn't understand and by guessing I managed to get the same hash results from my script as what came from the function.
-
"Having your device know you is cool. Having some cloud person know you is creepy." - Craig Federighi
not iOS, but macOS Monterey https://github.com/KhaosT/nhcalc
-
Apple's New CSAM Protections May Make iCloud Photos Bruteforceable
This isn't really true in a world where it's trivial to reverse engineer and decompile binaries.
For example, we already now have a tool for generating NeuralHash hashes for arbitrary images, thanks to KhaosT:
https://github.com/khaost/nhcalc
- Compute Apple NeuralHash for a given image
What are some alternatives?
GmsCore - Free implementation of Play Services
TNN - TNN: developed by Tencent Youtu Lab and Guangying Lab, a lightweight and high-performance deep learning framework for mobile inference. TNN is distinguished by several outstanding features, including its cross-platform capability, high performance, model compression and code pruning. Based on ncnn and Rapidnet, TNN further strengthens the support and performance optimization for mobile devices, and also draws on the advantages of good extensibility and high performance from existed open source efforts. TNN has been deployed in multiple Apps from Tencent, such as Mobile QQ, Weishi, Pitu, etc. Contributions are welcome to work in collaborative with us and make TNN a better framework. TNN:由腾讯优图实验室和光影实验室协同打造,移动端高性能、轻量级推理框架,同时拥有跨平台、高性能、模型压缩、代码裁剪等众多突出优势。TNN框架在原有Rapidnet、ncnn框架的基础上进一步加强了移动端设备的支持以及性能优化,同时也借鉴了业界主流开源框架高性能和良好拓展性的优点。目前TNN已经在手Q、微视、P图等应用中落地,欢迎大家参与协同共建,促进TNN推理框架进一步完善。
hardened_malloc - Hardened allocator designed for modern systems. It has integration into Android's Bionic libc and can be used externally with musl and glibc as a dynamic library for use on other Linux-based platforms. It will gain more portability / integration over time.
onnxruntime - ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
OpenWifiPass - An open source implementation of Apple's Wi-Fi Password Sharing protocol in Python.
ExpansionCards - Reference designs and documentation to create Expansion Cards for the Framework Laptop
neuralhash-collisions - A catalog of naturally occurring images whose Apple NeuralHash is identical.
neural-hash-collider - Preimage attack against NeuralHash 💣
img-cryptor - Image AES256 crypt-decrypt
appleprivacyletter - An open letter against Apple's new privacy-invasive client-side content scanning.
onnxjs - ONNX.js: run ONNX models using JavaScript
neural_hash_collision