Hitoshi Kiya
YOU?
Author Swipe
View article: Scale and Rotation Estimation of Similarity-Transformed Images via Cross-Correlation Maximization Based on Auxiliary Function Method
Scale and Rotation Estimation of Similarity-Transformed Images via Cross-Correlation Maximization Based on Auxiliary Function Method Open
This paper introduces a highly efficient algorithm capable of jointly estimating scale and rotation between two images with sub-pixel precision. Image alignment serves as a critical process for spatially registering images captured from di…
View article: A Privacy-Preserving Semantic-Segmentation Method Using Domain-Adaptation Technique
A Privacy-Preserving Semantic-Segmentation Method Using Domain-Adaptation Technique Open
We propose a privacy-preserving semantic-segmentation method for applying perceptual encryption to images used for model training in addition to test images. This method also provides almost the same accuracy as models without any encrypti…
View article: Effective Fine-Tuning of Vision Transformers with Low-Rank Adaptation for Privacy-Preserving Image Classification
Effective Fine-Tuning of Vision Transformers with Low-Rank Adaptation for Privacy-Preserving Image Classification Open
We propose a low-rank adaptation method for training privacy-preserving vision transformer (ViT) models that efficiently freezes pre-trained ViT model weights. In the proposed method, trainable rank decomposition matrices are injected into…
View article: Learnable Image Encryption Without Key Management for Privacy-Preserving Vision Transformer
Learnable Image Encryption Without Key Management for Privacy-Preserving Vision Transformer Open
We propose a privacy-preserving image classification method based on perceptual encryption that does not require centralized key management. In the proposed method, each client independently generates an encryption key to protect visual in…
View article: Privacy-Preserving ConvMixer Without Any Accuracy Degradation Using Compressible Encrypted Images
Privacy-Preserving ConvMixer Without Any Accuracy Degradation Using Compressible Encrypted Images Open
We propose an enhanced privacy-preserving method for image classification using ConvMixer, which is an extremely simple model that is similar in spirit to the Vision Transformer (ViT). Most privacy-preserving methods using encrypted images…
View article: Speech Privacy-preserving Methods Using Secret Key for Convolutional Neural Network Models and Their Robustness Evaluation
Speech Privacy-preserving Methods Using Secret Key for Convolutional Neural Network Models and Their Robustness Evaluation Open
In this paper, we propose privacy-preserving methods with a secret key for convolutional neural network (CNN)-based models in speech processing tasks. In environments where untrusted third parties, like cloud servers, provide CNN-based sys…
View article: Scene-Segmentation-Based Exposure Compensation for Tone Mapping of High Dynamic Range Scenes
Scene-Segmentation-Based Exposure Compensation for Tone Mapping of High Dynamic Range Scenes Open
We propose a novel scene-segmentation-based exposure compensation method for multi-exposure image fusion (MEF) based tone mapping. The aim of MEF-based tone mapping is to display high dynamic range (HDR) images on devices with limited dyna…
View article: On the Security of Bitstream-level JPEG Encryption with Restart Markers
On the Security of Bitstream-level JPEG Encryption with Restart Markers Open
This paper aims to evaluate the security of a bitstream-level JPEG encryption method using restart (RST) markers, where encrypted image can keep the JPEG file format with the same file size as non-encrypted image. Data encrypted using this…
View article: ENCRYPTION INSPIRED ADVERSARIAL DEFENSE FOR VISUAL CLASSIFICATION
ENCRYPTION INSPIRED ADVERSARIAL DEFENSE FOR VISUAL CLASSIFICATION Open
Conventional adversarial defenses reduce classification accuracy whether or not a model is under attacks. Moreover, most of image processing based defenses are defeated due to the problem of obfuscated gradients. In this paper, we propose …
View article: Enhancing Security Using Random Binary Weights in Privacy-Preserving Federated Learning
Enhancing Security Using Random Binary Weights in Privacy-Preserving Federated Learning Open
In this paper, we propose a novel method for enhancing security in privacy-preserving federated learning using the Vision Transformer. In federated learning, learning is performed by collecting updated information without collecting raw da…
View article: Privacy-Preserving Vision Transformer Using Images Encrypted with Restricted Random Permutation Matrices
Privacy-Preserving Vision Transformer Using Images Encrypted with Restricted Random Permutation Matrices Open
We propose a novel method for privacy-preserving fine-tuning vision transformers (ViTs) with encrypted images. Conventional methods using encrypted images degrade model performance compared with that of using plain images due to the influe…
View article: Disposable-key-based image encryption for collaborative learning of Vision Transformer
Disposable-key-based image encryption for collaborative learning of Vision Transformer Open
We propose a novel method for securely training the vision transformer (ViT) with sensitive data shared from multiple clients similar to privacy-preserving federated learning. In the proposed method, training images are independently encry…
View article: Speech privacy-preserving methods using secret key for convolutional neural network models and their robustness evaluation
Speech privacy-preserving methods using secret key for convolutional neural network models and their robustness evaluation Open
In this paper, we propose privacy-preserving methods with a secret key for convolutional neural network (CNN)-based models in speech processing tasks. In environments where untrusted third parties, like cloud servers, provide CNN-based sys…
View article: Encryption Method for JPEG Bitstreams for Partially Disclosing Visual Information
Encryption Method for JPEG Bitstreams for Partially Disclosing Visual Information Open
In this paper, we propose a novel encryption method for JPEG bitstreams in which encrypted data can preserve the JPEG file format with the same size as that without encryption. Accordingly, data encrypted with the method can be decoded wit…
View article: Efficient Fine-Tuning with Domain Adaptation for Privacy-Preserving Vision Transformer
Efficient Fine-Tuning with Domain Adaptation for Privacy-Preserving Vision Transformer Open
We propose a novel method for privacy-preserving deep neural networks (DNNs) with the Vision Transformer (ViT). The method allows us not only to train models and test with visually protected images but to also avoid the performance degrada…
View article: Fine-Tuning Text-To-Image Diffusion Models for Class-Wise Spurious Feature Generation
Fine-Tuning Text-To-Image Diffusion Models for Class-Wise Spurious Feature Generation Open
We propose a method for generating spurious features by leveraging large-scale text-to-image diffusion models. Although the previous work detects spurious features in a large-scale dataset like ImageNet and introduces Spurious ImageNet, we…
View article: A Random Ensemble of Encrypted Vision Transformers for Adversarially Robust Defense
A Random Ensemble of Encrypted Vision Transformers for Adversarially Robust Defense Open
Deep neural networks (DNNs) are well known to be vulnerable to adversarial examples (AEs). In previous studies, the use of models encrypted with a secret key was demonstrated to be robust against white-box attacks, but not against black-bo…
View article: Efficient Fine-Tuning with Domain Adaptation for Privacy-Preserving Vision Transformer
Efficient Fine-Tuning with Domain Adaptation for Privacy-Preserving Vision Transformer Open
We propose a novel method for privacy-preserving deep neural networks (DNNs) with the Vision Transformer (ViT). The method allows us not only to train models and test with visually protected images but to also avoid the performance degrada…
View article: A Random Ensemble of Encrypted models for Enhancing Robustness against Adversarial Examples
A Random Ensemble of Encrypted models for Enhancing Robustness against Adversarial Examples Open
Deep neural networks (DNNs) are well known to be vulnerable to adversarial examples (AEs). In addition, AEs have adversarial transferability, which means AEs generated for a source model can fool another black-box model (target model) with…
View article: On the Security of Learnable Image Encryption for Privacy-Preserving Deep Learning
On the Security of Learnable Image Encryption for Privacy-Preserving Deep Learning Open
In this paper, we evaluate the security of learnable image encryption methods proposed for privacy-preserving deep learning. In addition, we also propose a new generative model-based attack based on latent diffusion models. Various learnab…
View article: A Random Ensemble of Encrypted Vision Transformers for Adversarially Robust Defense
A Random Ensemble of Encrypted Vision Transformers for Adversarially Robust Defense Open
Deep neural networks (DNNs) are well known to be vulnerable to adversarial examples (AEs). In previous studies, the use of models encrypted with a secret key was demonstrated to be robust against white-box attacks, but not against black-bo…
View article: Efficient Key-Based Adversarial Defense for ImageNet by Using Pre-Trained Models
Efficient Key-Based Adversarial Defense for ImageNet by Using Pre-Trained Models Open
In this paper, we propose key-based defense model proliferation by leveraging pre-trained models and utilizing recent efficient fine-tuning techniques on ImageNet-1 k classification. First, we stress that deploying key-based models on edge…
View article: Efficient Key-Based Adversarial Defense for ImageNet by Using Pre-trained Model
Efficient Key-Based Adversarial Defense for ImageNet by Using Pre-trained Model Open
In this paper, we propose key-based defense model proliferation by leveraging pre-trained models and utilizing recent efficient fine-tuning techniques on ImageNet-1k classification. First, we stress that deploying key-based models on edge …
View article: A privacy-preserving method using secret key for convolutional neural network-based speech classification
A privacy-preserving method using secret key for convolutional neural network-based speech classification Open
In this paper, we propose a privacy-preserving method with a secret key for convolutional neural network (CNN)-based speech classification tasks. Recently, many methods related to privacy preservation have been developed in image classific…
View article: Domain Adaptation for Efficiently Fine-tuning Vision Transformer with Encrypted Images
Domain Adaptation for Efficiently Fine-tuning Vision Transformer with Encrypted Images Open
In recent years, deep neural networks (DNNs) trained with transformed data have been applied to various applications such as privacy-preserving learning, access control, and adversarial defenses. However, the use of transformed data decrea…
View article: Hindering Adversarial Attacks with Multiple Encrypted Patch Embeddings
Hindering Adversarial Attacks with Multiple Encrypted Patch Embeddings Open
In this paper, we propose a new key-based defense focusing on both efficiency and robustness. Although the previous key-based defense seems effective in defending against adversarial examples, carefully designed adaptive attacks can bypass…
View article: Block-Wise Encryption for Reliable Vision Transformer models
Block-Wise Encryption for Reliable Vision Transformer models Open
This article presents block-wise image encryption for the vision transformer and its applications. Perceptual image encryption for deep learning enables us not only to protect the visual information of plain images but to also embed unique…
View article: Security Evaluation of Compressible and Learnable Image Encryption Against Jigsaw Puzzle Solver Attacks
Security Evaluation of Compressible and Learnable Image Encryption Against Jigsaw Puzzle Solver Attacks Open
Several learnable image encryption schemes have been developed for privacy-preserving image classification. This paper focuses on the security block-based image encryption methods that are learnable and JPEG-friendly. Permuting divided blo…
View article: Enhanced Security with Encrypted Vision Transformer in Federated Learning
Enhanced Security with Encrypted Vision Transformer in Federated Learning Open
Federated learning is a learning method for training models over multiple participants without directly sharing their raw data, and it has been expected to be a privacy protection method for training data. In contrast, attack methods have …
View article: Enhanced Security against Adversarial Examples Using a Random Ensemble of Encrypted Vision Transformer Models
Enhanced Security against Adversarial Examples Using a Random Ensemble of Encrypted Vision Transformer Models Open
Deep neural networks (DNNs) are well known to be vulnerable to adversarial examples (AEs). In addition, AEs have adversarial transferability, which means AEs generated for a source model can fool another black-box model (target model) with…