AI Paper Atlas

Research map / Architectures and efficiency / CNN design

ENSEI: Efficient Secure Inference via Frequency-Domain Homomorphic Convolution for Privacy-Preserving Visual Recognition

Song Bian, Tianchen Wang, Masayuki Hiromoto, Yiyu Shi, Takashi Sato

CVPR 2020 · Architectures and efficiency › CNN design · approach: Dense matching and estimation CNNs · 42 citations (OpenAlex)

Check your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use frequency-domain secure convolution instead of Gazelle's time-domain convolutionENSEI-NTT · Convolution (32 × 32 × 32 input, 3 × 3 × 32 filter)online time: 195 ms -> 16.9 msImproved
Use frequency-domain secure convolution instead of Gazelle's time-domain convolutionENSEI-NTT · Convolution (32 × 32 × 32 input, 3 × 3 × 32 filter)setup time: 704 ms -> 21.4 msImproved
Use frequency-domain secure convolution instead of Gazelle's time-domain convolutionENSEI-NTT · Convolution (28 × 28 × 1 input, 5 × 5 × 5 filter)online time: 9.20 ms -> 1.8 msImproved
Use ENSEI with BC3 architecture instead of prior-art secure inferenceENSEI · CIFAR-10, BC3 in [30]inference time: 35.8 s -> 14.7 sImproved
Use ENSEI with BC4 architecture instead of prior-art secure inferenceENSEI · CIFAR-10, BC4 in [30]inference time: 123.9 s -> 30.91 sImproved
Replace the second and fifth ReLU layers with square activation (ablation)ENSEI on the Fig. 13 in [25] architecture · CIFAR-10inference time: 7.72 s -> 1.38 sImproved
Replace a small portion of ReLU layers with square activation (ablation)High-Square · CIFAR-10, same architecture as [23,25]accuracy: 82% -> 82%No clear effect
Replace all ReLU layers with square activation (ablation)The square-activation variant · CIFAR-10accuracy: 82% -> 10%Hurt
Use binary-weight precision rather than the reported original accuracy setting (ablation)ENSEI, same architecture as [23,25] · CIFAR-10accuracy: 82% -> 81%Hurt
Use high precision (11-bit features and 7-bit weights) rather than full-bit precision (ablation)ENSEI, same architecture as [23,25] · CIFAR-10accuracy: 83% -> 82%No clear effect
Use medium precision (8-bit input and 4-bit filter) rather than full-bit precision (ablation)ENSEI, same architecture as [23,25] · CIFAR-10accuracy: 83% -> 79%Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy