Development of learning subspace-based methods for pattern recognition

基于学习子空间的模式识别方法的开发

基本信息

项目摘要

In fiscal year 2022, we worked to address the problem that traditional deep neural network frameworks process image sets independently, without considering the underlying feature distribution and the variance of the images in the set. To overcome this limitation, we devised a new subspace learning method called Grassmannian learning mutual subspace method (G-LMSM), which is an NN layer that can be integrated into deep neural networks.G-LMSM maps the image set into a low-dimensional input subspace representation, which is then matched with dictionary subspaces using a similarity metric of their canonical angles, an interpretable and computationally efficient metric. The key idea of G-LMSM is to learn dictionary subspaces as points on the Grassmann manifold, which is a smooth, non-linear manifold that captures the geometric structure of subspaces. This learning is optimized with Riemannian stochastic gradient descent, which is stable, efficient, and theoretically well-grounded.The proposed method was evaluated on three different tasks: hand shape recognition, face identification, and facial emotion recognition. Our experimental results showed that G-LMSM outperformed state-of-the-art methods on all three tasks, demonstrating its potential to improve the performance of deep frameworks for object recognition from image sets.
在2022财年,我们致力于解决传统的深度神经网络框架独立处理图像集的问题,而不考虑图像集中的潜在特征分布和方差。为了克服这一局限性,我们设计了一种新的子空间学习方法--Grassman学习互动子空间方法(G-LMSM),它是一个可以集成到深层神经网络中的NN层,将图像集映射到低维输入子空间表示,然后使用它们的典型角的相似性度量与字典子空间进行匹配,该度量是一种可解释且计算高效的度量。G-LMSM的核心思想是将字典子空间学习为Grassmann流形上的点,Grassmann流形是一种光滑的非线性流形,它捕获子空间的几何结构。该方法采用Riemannian随机梯度下降法进行优化,具有稳定、高效和理论基础良好的特点,并在手形识别、人脸识别和表情识别三个不同的任务上进行了评估。我们的实验结果表明,G-LMSM在所有三个任务上都优于最先进的方法,表明它有潜力提高图像集中目标识别的深层框架的性能。

项目成果

期刊论文数量(4)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
Analysis of Temporal Tensor Datasets on Product Grassmann Manifold
Grassmannian learning mutual subspace method for image set recognition
  • DOI:
    10.1016/j.neucom.2022.10.040
  • 发表时间:
    2021-11
  • 期刊:
  • 影响因子:
    6
  • 作者:
    L. S. Souza;Naoya Sogi;B. Gatto;Takumi Kobayashi;K. Fukui
  • 通讯作者:
    L. S. Souza;Naoya Sogi;B. Gatto;Takumi Kobayashi;K. Fukui
Temporal-stochastic tensor features for action recognition
  • DOI:
    10.1016/j.mlwa.2022.100407
  • 发表时间:
    2022-09
  • 期刊:
  • 影响因子:
    0
  • 作者:
    Bojan Batalo;L. S. Souza;B. Gatto;Naoya Sogi;K. Fukui
  • 通讯作者:
    Bojan Batalo;L. S. Souza;B. Gatto;Naoya Sogi;K. Fukui
Environmental sound classification based on CNN latent subspaces
基于CNN潜在子空间的环境声音分类
  • DOI:
  • 发表时间:
    2022
  • 期刊:
  • 影响因子:
    0
  • 作者:
    Maha Mahyub;Lincon S. Souza;Bojan Batalo;Kazuhiro Fukui
  • 通讯作者:
    Kazuhiro Fukui
{{ item.title }}
{{ item.translation_title }}
  • DOI:
    {{ item.doi }}
  • 发表时间:
    {{ item.publish_year }}
  • 期刊:
  • 影响因子:
    {{ item.factor }}
  • 作者:
    {{ item.authors }}
  • 通讯作者:
    {{ item.author }}

数据更新时间:{{ journalArticles.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ monograph.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ sciAawards.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ conferencePapers.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ patent.updateTime }}

SALESDESOUZA LINCON其他文献

SALESDESOUZA LINCON的其他文献

{{ item.title }}
{{ item.translation_title }}
  • DOI:
    {{ item.doi }}
  • 发表时间:
    {{ item.publish_year }}
  • 期刊:
  • 影响因子:
    {{ item.factor }}
  • 作者:
    {{ item.authors }}
  • 通讯作者:
    {{ item.author }}

相似海外基金

DMS-EPSRC: Asymptotic Analysis of Online Training Algorithms in Machine Learning: Recurrent, Graphical, and Deep Neural Networks
DMS-EPSRC:机器学习中在线训练算法的渐近分析:循环、图形和深度神经网络
  • 批准号:
    EP/Y029089/1
  • 财政年份:
    2024
  • 资助金额:
    $ 3万
  • 项目类别:
    Research Grant
Deep neural networks
深度神经网络
  • 批准号:
    2902331
  • 财政年份:
    2024
  • 资助金额:
    $ 3万
  • 项目类别:
    Studentship
CAREER: Reliable and Accelerated Deep Neural Networks via Co-Design of Hardware and Algorithms
职业:通过硬件和算法的协同设计实现可靠且加速的深度神经网络
  • 批准号:
    2340516
  • 财政年份:
    2024
  • 资助金额:
    $ 3万
  • 项目类别:
    Continuing Grant
Collaborative Research: SHF: Medium: Verifying Deep Neural Networks with Spintronic Probabilistic Computers
合作研究:SHF:中:使用自旋电子概率计算机验证深度神经网络
  • 批准号:
    2311295
  • 财政年份:
    2023
  • 资助金额:
    $ 3万
  • 项目类别:
    Continuing Grant
CAREER: Deep Neural Networks That Can See Shape From Images: Models, Algorithms, and Applications
职业:可以从图像中看到形状的深度神经网络:模型、算法和应用
  • 批准号:
    2239977
  • 财政年份:
    2023
  • 资助金额:
    $ 3万
  • 项目类别:
    Continuing Grant
Property-Driven Quality Assurance of Adversarial Robustness of Deep Neural Networks
深度神经网络对抗鲁棒性的属性驱动质量保证
  • 批准号:
    23K11049
  • 财政年份:
    2023
  • 资助金额:
    $ 3万
  • 项目类别:
    Grant-in-Aid for Scientific Research (C)
Imaging Epilepsy Sources with Biophysically Constrained Deep Neural Networks
使用生物物理约束的深度神经网络对癫痫源进行成像
  • 批准号:
    10655833
  • 财政年份:
    2023
  • 资助金额:
    $ 3万
  • 项目类别:
High-performance deep neural networks for medical image analysis
用于医学图像分析的高性能深度神经网络
  • 批准号:
    10723553
  • 财政年份:
    2023
  • 资助金额:
    $ 3万
  • 项目类别:
Crop design simulator utilizing deep neural networks with crop growth model as knowledge layer
利用深度神经网络以作物生长模型作为知识层的作物设计模拟器
  • 批准号:
    23H02200
  • 财政年份:
    2023
  • 资助金额:
    $ 3万
  • 项目类别:
    Grant-in-Aid for Scientific Research (B)
DMS-EPSRC: Asymptotic Analysis of Online Training Algorithms in Machine Learning: Recurrent, Graphical, and Deep Neural Networks
DMS-EPSRC:机器学习中在线训练算法的渐近分析:循环、图形和深度神经网络
  • 批准号:
    2311500
  • 财政年份:
    2023
  • 资助金额:
    $ 3万
  • 项目类别:
    Standard Grant
{{ showInfoDetail.title }}

作者:{{ showInfoDetail.author }}

知道了