no code implementations • 27 Mar 2024 • Taro Togo, Ren Togo, Keisuke Maeda, Takahiro Ogawa, Miki Haseyama
This study presents a novel approach to Generative Class Incremental Learning (GCIL) by introducing the forgetting mechanism, aimed at dynamically managing class information for better adaptation to streaming data.
no code implementations • 15 Feb 2024 • He Zhu, Ren Togo, Takahiro Ogawa, Miki Haseyama
We present a novel prompt-based personalized federated learning (pFL) method to address data heterogeneity and privacy concerns in traditional medical visual question answering (VQA) methods.
Medical Visual Question Answering Personalized Federated Learning +2
no code implementations • 29 Jan 2024 • Guang Li, Ren Togo, Takahiro Ogawa, Miki Haseyama
Based on this observation, we propose an importance-aware adaptive dataset distillation (IADD) method that can improve distillation performance by automatically assigning importance weights to different network parameters during distillation, thereby synthesizing more robust distilled datasets.
no code implementations • 8 Mar 2023 • He Zhu, Ren Togo, Takahiro Ogawa, Miki Haseyama
We present a novel multimodal interpretable VQA model that can answer the question more accurately and generate diverse explanations.
no code implementations • 19 Dec 2022 • Guang Li, Ren Togo, Takahiro Ogawa, Miki Haseyama
On the other hand, batch knowledge ensembling-based fine-tuning can utilize category knowledge of images in a batch according to their visual feature similarities to improve detection performance.
no code implementations • 19 Dec 2022 • Guang Li, Ren Togo, Takahiro Ogawa, Miki Haseyama
We compared six self-supervised learning (SSL) methods (Cross, BYOL, SimSiam, SimCLR, PIRL-jigsaw, and PIRL-rotation) with the proposed method.
1 code implementation • 6 Dec 2022 • Zongyao Li, Ren Togo, Takahiro Ogawa, Miki Haseyama
Model adaptation is proposed as a new domain adaptation problem which requires access to a pre-trained model instead of data for the source domain.
no code implementations • 1 Nov 2022 • Guang Li, Ren Togo, Takahiro Ogawa, Miki Haseyama
To address this issue, this work aims to improve MIM for medical images and evaluate its effectiveness in an open X-ray image dataset.
no code implementations • 29 Sep 2022 • Guang Li, Ren Togo, Takahiro Ogawa, Miki Haseyama
Dataset complexity assessment aims to predict classification performance on a dataset with complexity calculation before training a classifier, which can also be used for classifier selection and dataset reduction.
1 code implementation • 29 Sep 2022 • Guang Li, Ren Togo, Takahiro Ogawa, Miki Haseyama
Furthermore, our method can extract essential weights of DCNN models to reduce the memory required to save trained models for efficient medical data sharing.
1 code implementation • 29 Sep 2022 • Guang Li, Ren Togo, Takahiro Ogawa, Miki Haseyama
In this study, we propose a novel dataset distillation method based on parameter pruning.
1 code implementation • 29 Sep 2022 • Guang Li, Ren Togo, Takahiro Ogawa, Miki Haseyama
Sharing medical datasets between hospitals is challenging because of the privacy-protection problem and the massive cost of transmitting and storing many high-resolution medical images.
1 code implementation • 15 Sep 2022 • Nao Nakagawa, Ren Togo, Takahiro Ogawa, Miki Haseyama
The GW metric measures the distance structure-oriented discrepancy between distributions even with different dimensionalities, which provides a direct measure between the latent and data spaces.
no code implementations • 7 Jun 2022 • Guang Li, Ren Togo, Takahiro Ogawa, Miki Haseyama
Our method provides a feasible solution for self-supervised learning with real-world high-resolution images that uses small batch sizes.
no code implementations • 7 Jun 2022 • Guang Li, Ren Togo, Takahiro Ogawa, Miki Haseyama
The global outbreak of the Coronavirus 2019 (COVID-19) has overloaded worldwide healthcare systems.
no code implementations • 7 Apr 2021 • Guang Li, Ren Togo, Takahiro Ogawa, Miki Haseyama
The effectiveness of the proposed self-supervised learning method in gastritis detection is verified using a few annotated gastric X-ray images.
1 code implementation • 7 Apr 2021 • Guang Li, Ren Togo, Takahiro Ogawa, Miki Haseyama
The idea of our distillation method is to extract the valid information of the medical dataset and generate a tiny distilled dataset that has a different data distribution.