Blind-compute proof on Zama Concrete ML: client holds the key, server runs inference on ciphertext. Measured, honest about what it does not prove.
-
Updated
Sep 7, 2026 - Python
#
concrete-ml
Here are 3 public repositories matching this topic...
Identify fingerprints on encrypted images using Fully Homomorphic Encryption (Zama Concrete-ML). The server never sees the raw biometric — all ridge detection, enhancement, and feature extraction run directly on ciphertext.
python privacy image-processing fingerprint python3 biometrics convolution encrypted-computation homomorphic-encryption privacy-preserving fingerprint-recognition fhe ridge-detection fully-homomorphic-encryption zama concrete-ml
-
Updated
Aug 11, 2026 - Python
Privacy-preserving machine learning with Fully Homomorphic Encryption (FHE): theory plus encrypted-inference demos using Concrete-ML (Iris, sentiment, image filtering). Cryptography coursework (M.Sc. IT, HUTECH).
cryptography academic-project homomorphic-encryption secure-computation privacy-preserving-ml fully-homomorphic-encryption encrypted-inference concrete-ml
-
Updated
Jun 9, 2026
Add this topic to your repo
To associate your repository with the concrete-ml topic, visit your repo's landing page and select "manage topics."