PixCrypt speeds up secure pixel-level data analysis by 35 times
PixCrypt: Fast Fine-Grained FHE with Range-Aware Caching
Cryptography and Security
Summary
Analyzing secret data one pixel at a time can be very slow when using fully secure encryption methods. The authors created PixCrypt, a technique that speeds this up by storing and reusing parts of the encrypted data instead of always making brand new encrypted pieces. This makes it much faster to analyze detailed images securely, like detecting tumors, without exposing sensitive information. Their method also keeps the encrypted data secure and works well with current hardware.
What this means in practice
- •For medical imaging teams: Process encrypted medical images pixel-by-pixel faster to support secure tumor detection and anomaly analysis without compromising patient privacy.
- •For cloud service operators: Accelerate privacy-preserving image analytics on encrypted data to offer improved secure computation services to clients handling sensitive visual data.
Authors
Chao Wang, Shubing Yang, Xiaoyan Sun, Yan Bai, Jun Dai, Dongfang Zhao
Abstract
Many analytics tasks require secure computation over encrypted data. In particular, fine-grained data such as pixel-level images require higher precision, as every pixel can directly affect outcomes in tasks like tumor segmentation and anomaly detection. While Multi-Party Computation (MPC) is interactive, Differential Privacy (DP) protects only aggregate values, and Partially Homomorphic Encryption (PHE) lacks multiplicative support, none of them can efficiently handle fine-grained data analytics. Fully Homomorphic Encryption (FHE) uniquely enables arbitrary operations on encrypted pixels but remains computationally expensive, posing significant challenges for both software and hardware accelerators. We present PixCrypt, a caching-based acceleration mechanism for fine-grained fully homomorphic encryption. PixCrypt replaces expensive fresh ciphertext generation with cache retrieval and coefficient-level operations across CKKS, BFV, and BGV, while randomized reconstruction ensures that ciphertexts do not repeat. Its linear noise growth reduces the need for bootstrapping and lowers NTT load, improving hardware accelerator efficiency. This design yields up to 35x faster fine-grained encryption and maintains IND-CPA (Indistinguishability under Chosen Plaintext Attack) security. Experiments on five real-world pixel-level image processing tasks show that PixCrypt significantly improves the practicality of FHE for privacy-preserving analytics.