ENSEI: Efficient Secure Inference via Frequency-Domain Homomorphic Convolution for Privacy-Preserving Visual Recognition.
Song BianTianchen WangMasayuki HiromotoYiyu ShiTakashi SatoPublished in: CoRR (2020)
Keyphrases
- privacy preserving
- frequency domain
- visual recognition
- vertically partitioned data
- scalar product
- spatial domain
- horizontally partitioned data
- sensitive data
- multi party
- secure multiparty computation
- fourier transform
- privacy preserving data mining
- fast fourier transform
- privacy preservation
- discrete fourier transform
- fourier domain
- data privacy
- homomorphic encryption
- privacy concerns
- user privacy
- object recognition
- privacy preserving association rule mining
- private information
- privacy sensitive
- denoising
- subband
- horizontally partitioned
- image classification
- feature extraction
- sensitive information
- privacy protection
- privacy guarantees
- image processing
- cloud computing
- image compression
- feature set
- multiresolution
- high quality
- image segmentation