Masayuki Hiromoto
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Masayuki Hiromoto has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Hiroki Sugano
Graduate School of Informatics, Kyoto University
1 shared publication - Kentaro Nakahara
Graduate School of Informatics, Kyoto University, Yoshida-hon-machi, Sakyo, Kyoto, 606-8501, Japan
1 shared publication - Ryusuke Miyamoto
Graduate School of Information Science, Nara Institute of Science and Technology
1 shared publication - Song Bian
Beihang University, Beijing, China
1 shared publication - Takashi Sato
Department of Communications and Computer Engineering Kyoto University
1 shared publication - Tianchen Wang
Department of Computer Science and Engineering University of Notre Dame
1 shared publication
Latest Publications
- Open Access
ENSEI: Efficient Secure Inference via Frequency-Domain Homomorphic Convolution for Privacy-Preserving Visual Recognition
In this work, we propose ENSEI, a secure inference (SI) framework based on the frequency-domain secure convolution (FDSC) protocol for the efficient execution of privacy-preserving visual recognition. Our observation is that, under the combination of homomorphic encryption and secret sharing, homomorphic convolution can be obliviously carried out in the frequency domain, significantly simplifying the related computations. We provide protocol designs and parameter derivations for number-theoretic transform (NTT) based FDSC. In the experiment, we thoroughly study the accuracy-efficiency trade-offs between time- and frequency-domain homomorphic convolution. With ENSEI, compared to the best known works, we achieve 5--11x online time reduction, up to 33x setup time reduction, and up to 10x reduction in the overall inference time. A further 33% of bandwidth reductions can be obtained on binary neural networks with only 1% of accuracy degradation on the CIFAR-10 dataset.
Open AccessENSEI: Efficient Secure Inference via Frequency-Domain Homomorphic Convolution for Privacy-Preserving Visual Recognition
In this work, we propose ENSEI, a secure inference (SI) framework based on the frequency-domain secure convolution (FDSC) protocol for the efficient execution of privacy-preserving visual recognition. Our observation is that, under the combination of homomorphic encryption and secret sharing, homomorphic convolution can be obliviously carried out in the frequency domain, significantly simplifying the related computations. We provide protocol designs and parameter derivations for number-theoretic transform (NTT) based FDSC. In the experiment, we thoroughly study the accuracy-efficiency trade-offs between time- and frequency-domain homomorphic convolution. With ENSEI, compared to the best known works, we achieve 5--11x online time reduction, up to 33x setup time reduction, and up to 10x reduction in the overall inference time. A further 33% of bandwidth reductions can be obtained on binary neural networks with only 1% of accuracy degradation on the CIFAR-10 dataset.
- 1 June 2007
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2005 (10636919) , 1-8
A Specialized Processor Suitable for AdaBoost-Based Detection with Haar-like Features
Robust and rapid object detection is one of the great challenges in the field of computer vision. This paper proposes a hardware architecture suitable for object detection by Viola and Jones based on an AdaBoost learning algorithm with Haar-like features as weak classifiers. Our architecture realizes rapid and robust detection with two major features: hybrid parallel execution and an image scaling method. The first exploits the cascade structure of classifiers, in which classifiers located near the beginning of the cascade are used more frequently than subsequent classifiers. We assign more resources to the former classifiers to execute in parallel than subsequent classifiers. This dramatically improves the total processing speed without a great increase in circuit area. The second feature is a method of scaling input images instead of scaling classifiers. This increases the efficiency of hardware implementation while retaining a high detection rate. In addition we implement the proposed architecture on a Virtex-5 FPGA to show that it achieves real-time object detection at 30 frames per second on VGA video.
- 1 June 2007
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2005 (10636919) , 1-8
A Specialized Processor Suitable for AdaBoost-Based Detection with Haar-like Features
Robust and rapid object detection is one of the great challenges in the field of computer vision. This paper proposes a hardware architecture suitable for object detection by Viola and Jones based on an AdaBoost learning algorithm with Haar-like features as weak classifiers. Our architecture realizes rapid and robust detection with two major features: hybrid parallel execution and an image scaling method. The first exploits the cascade structure of classifiers, in which classifiers located near the beginning of the cascade are used more frequently than subsequent classifiers. We assign more resources to the former classifiers to execute in parallel than subsequent classifiers. This dramatically improves the total processing speed without a great increase in circuit area. The second feature is a method of scaling input images instead of scaling classifiers. This increases the efficiency of hardware implementation while retaining a high detection rate. In addition we implement the proposed architecture on a Virtex-5 FPGA to show that it achieves real-time object detection at 30 frames per second on VGA video.