基于Grassmann流形几何深度网络的图像集识别方法
Geometry deep network image-set recognition method based on Grassmann manifolds
查看参考文献17篇
|
文摘
|
近年来,深度学习以其强大的非线性计算能力在目标检测和识别任务中取得了巨大的突破。现有的深度学习网络几乎都是以数据的欧氏结构为前提,而在计算机视觉中许多数据都具有严格的流形结构,如图像集可表示为Grassmann流形。基于数据的流形几何结构来设计深度学习网络,将微分几何理论与深度学习理论相结合,提出一种基于Grassmann流形的深度图像集识别网络。同时在模型训练过程中,使用基于矩阵链式法则的反向传播算法来更新模型,并将权值的优化过程转换为Grassmann流形上的黎曼优化问题。实验结果表明:该方法不仅在结果上识别准确率得到了提高,同时在训练和测试速度上也有一个数量级的提升。 |
|
其他语种文摘
|
In recent years,deep learning techniques have achieved great breakthrough for its powerful non -linear computations in the tasks of target recognition and detection.Existing deep networks were almost designed based on the precondition that the visual data reside on the Euclidean space.However,many data in computer vision have rigorous geometry of manifolds,i.e.,image sets can be represented as Grassmann manifolds.The deep network was devised based on the non -Euclidean structure of the manifold-valued data,which combined the differential geometry and deep learning methods theoretically.Furthermore,a deep network for image-set recognition based on the Grassmann manifold was proposed.In the training process,the model was updated by the use of the backpropagation algorithm derived from the matrix chain rule.Learning of the weights can be transformed as the Riemannian optimization problem on the Grassmannian.The experimental results show that this method not only improves the accuracy of recognition,but also accelerates the training and test process in one magnitude. |
|
来源
|
红外与激光工程
,2018,47(7):0703002-1-0703002-7 【核心库】
|
|
DOI
|
10.3788/IRLA201847.0703002
|
|
关键词
|
深度学习
;
Grassmann流形
;
黎曼优化
;
图像集识别
|
|
地址
|
1.
中国科学院沈阳自动化研究所, 辽宁, 沈阳, 110016
2.
中国科学院大学, 北京, 100049
3.
中国科学院光电信息处理重点实验室, 中国科学院光电信息处理重点实验室, 辽宁, 沈阳, 110016
|
|
语种
|
中文 |
|
文献类型
|
研究性论文 |
|
ISSN
|
1007-2276 |
|
学科
|
自动化技术、计算机技术 |
|
基金
|
中国科学院重点创新基金
|
|
文献收藏号
|
CSCD:6373935
|
参考文献 共
17
共1页
|
|
1.
罗海波. 基于深度学习的目标跟踪方法研究现状与展望.
红外与激光工程,2017,46(5):0502002
|
被引
18
次
|
|
|
|
|
2.
罗海波. 基于双模全卷积网络的行人检测算法(特邀).
红外与激光工程,2018,47(2):0203001
|
被引
3
次
|
|
|
|
|
3.
裴晓敏. 时空特征融合深度学习网络人体行为识别方法.
红外与激光工程,2018,47(2):0203007
|
被引
3
次
|
|
|
|
|
4.
柯余洋. 基于递归神经网络和蚁群优化算法的发电环保调度.
信息与控制,2017,46(4):415-421
|
被引
2
次
|
|
|
|
|
5.
吴培良. 一种基于CLM的服务机器人室内功能区分类方法.
机器人,2018,40(2):188-194
|
被引
3
次
|
|
|
|
|
6.
徐朋. 冗余机器人逆运动学解流形的多目标优化.
机器人,2016,38(6):704-710
|
被引
5
次
|
|
|
|
|
7.
徐涛. 基于协同显著性的服务机器人空间物体快速定位方法.
机器人,2017,39(3):307-315
|
被引
3
次
|
|
|
|
|
8.
Huang Z. A Riemannian network for SPD matrix learning.
Association for the Advancement of Artificial Intelligence,2017,2(4):2036-2042
|
被引
1
次
|
|
|
|
|
9.
Huang Z. Building deep networks on Grassmann manifolds.
Association for the Advancement of Artificial Intelligence,2018:3279-3286
|
被引
1
次
|
|
|
|
|
10.
Huang Z. Projection metric learning on Grassmann manifold with application to video based face recognition.
Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2015:140-149
|
被引
4
次
|
|
|
|
|
11.
Harandi M. Graph embedding discriminant analysis on grassmannian manifolds for improved image set matching.
IEEE Conference on Computer Vision and Pattern Recognition,2011:2705-2712
|
被引
1
次
|
|
|
|
|
12.
Hamm J. Grassmann discriminant analysis: a unifying view on subspace-based learning.
Proceedings of the 25th international conference on Machine learning,2008:376-383
|
被引
9
次
|
|
|
|
|
13.
Liu T. Kernel sparse representation on grassmann manifolds for visual clustering.
Optical Engineering,2018,57(5):053104
|
被引
1
次
|
|
|
|
|
14.
Dhall A. Emotion recognition in the wild challenge 2014: Baseline, data and protocol.
Proceedings of the 16th International Conference on Multimodal Interaction,2014:461-466
|
被引
3
次
|
|
|
|
|
15.
Kim T K. Discriminative learning and recognition of image set classes using canonical correlations.
IEEE T-PAMI,2007,29(6):1005-1018
|
被引
36
次
|
|
|
|
|
16.
Liu M. Learning expressionlets on spatio-temporal manifold for dynamic facial expression recognition.
2014 IEEE Conference on Computer Vision and Pattern Recognition,2014:4321-4328
|
被引
1
次
|
|
|
|
|
17.
Ionescu C. Matrix backpropagation for deep networks with structured layers.
Proceedings of the IEEE International Conference on Computer Vision,2015:2965-2973
|
被引
3
次
|
|
|
|
|
了解气象卫星更多信息，请访问