HERS: Homomorphically Encrypted Representation Search
HERS: Homomorphically Encrypted Representation Search
复制标题
DOI：
10.1109/tbiom.2021.3139866
复制
发表时间：
2020-03
期刊：
影响因子：
--
通讯作者：
Joshua J. Engelsma;Anil K. Jain;Vishnu Naresh Boddeti
中科院分区：
文献类型：
--
作者：
Joshua J. Engelsma;Anil K. Jain;Vishnu Naresh Boddeti
我们提出了一种用于针对加密域中大型画廊搜索探针（或查询）图像表示的方法。我们要求用固定长度表示表示探针和画廊图像，这对于从学习的网络获得的表示是典型的。我们的加密方案对如何获得固定长度表示不可知，因此可以应用于任何应用域中的任何固定长度表示。我们的方法被称为她的方法（同派加密表示搜索），是通过（i）压缩表示其估计的固有维度的表示，而准确性的最小损失（ii）使用拟议的完全同质加密方案和（iii）有效地加密压缩表示形式（ii）直接在加密域中直接与加密表示的画廊进行搜索，而不会解密它们。大型面部，指纹和对象数据集（例如ImageNet）上的数值结果表明，在加密域中，首次准确且快速的图像搜索是可行的（500秒； $ 275 \ tims $ times $ speed胜过状态 - 与1亿个画廊的加密搜索有关）。代码可在https://github.com/human-analysis/hers-ecrypted-image-search上找到。
We present a method to search for a probe (or query) image representation against a large gallery in the encrypted domain. We require that the probe and gallery images be represented in terms of a fixed-length representation, which is typical for representations obtained from learned networks. Our encryption scheme is agnostic to how the fixed-length representation is obtained and can therefore be applied to any fixed-length representation in any application domain. Our method, dubbed HERS (Homomorphically Encrypted Representation Search), operates by (i) compressing the representation towards its estimated intrinsic dimensionality with minimal loss of accuracy (ii) encrypting the compressed representation using the proposed fully homomorphic encryption scheme, and (iii) efficiently searching against a gallery of encrypted representations directly in the encrypted domain, without decrypting them. Numerical results on large galleries of face, fingerprint, and object datasets such as ImageNet show that, for the first time, accurate and fast image search within the encrypted domain is feasible at scale (500 seconds; $275\times $ speed up over state-of-the-art for encrypted search against a gallery of 100 million). Code is available at https://github.com/human-analysis/hers-encrypted-image-search.