The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
우리는 감독 학습과 증분 비지도 학습이라는 두 가지 학습 프레임워크에서 비디오 시퀀스의 얼굴 인식을 위한 뷰 종속 공분산 행렬을 갖춘 모양 다양체를 제안합니다. 이 방법의 장점은 첫째, 뷰 종속 공분산 행렬 모델을 사용하는 모양 다양체는 포즈 변경에 견고하고 노이즈 불변성입니다. 왜냐하면 삽입된 공분산 행렬은 샘플의 분포를 학습하기 위해 포즈를 기반으로 계산되기 때문입니다. 매니폴드. 또한 제안된 점진적 비지도 학습 프레임워크는 실제 얼굴 인식 애플리케이션에 더 현실적입니다. 훈련을 위해 완전한 포즈(왼쪽 측면에서 오른쪽 측면까지)에서 많은 양의 얼굴 시퀀스를 수집하는 것이 어렵다는 것은 분명합니다. 여기서 증분 비지도 학습 프레임워크를 사용하면 사용 가능한 초기 시퀀스로 시스템을 교육하고 나중에 레이블이 없는 시퀀스가 입력될 때마다 시스템 지식을 점진적으로 업데이트할 수 있습니다. 또한, 점진적 비지도 학습 프레임워크에서 분류 정확도를 향상시키고 병합 프로세스를 위해 중첩된 포즈가 있는 시퀀스를 쉽게 감지하기 위해 포즈 추정 시스템이 있는 뷰 종속 공분산 행렬 모델과 모양 다양체를 통합합니다. 실험 결과, 두 프레임워크 모두에서 제안된 뷰 종속 공분산 행렬 방법을 사용하는 모양 다양체는 비디오 시퀀스에서 얼굴을 정확하게 인식할 수 있음을 보여주었습니다.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
부
Lina, Tomokazu TAKAHASHI, Ichiro IDE, Hiroshi MURASE, "Incremental Unsupervised-Learning of Appearance Manifold with View-Dependent Covariance Matrix for Face Recognition from Video Sequences" in IEICE TRANSACTIONS on Information,
vol. E92-D, no. 4, pp. 642-652, April 2009, doi: 10.1587/transinf.E92.D.642.
Abstract: We propose an appearance manifold with view-dependent covariance matrix for face recognition from video sequences in two learning frameworks: the supervised-learning and the incremental unsupervised-learning. The advantages of this method are, first, the appearance manifold with view-dependent covariance matrix model is robust to pose changes and is also noise invariant, since the embedded covariance matrices are calculated based on their poses in order to learn the samples' distributions along the manifold. Moreover, the proposed incremental unsupervised-learning framework is more realistic for real-world face recognition applications. It is obvious that it is difficult to collect large amounts of face sequences under complete poses (from left sideview to right sideview) for training. Here, an incremental unsupervised-learning framework allows us to train the system with the available initial sequences, and later update the system's knowledge incrementally every time an unlabelled sequence is input. In addition, we also integrate the appearance manifold with view-dependent covariance matrix model with a pose estimation system for improving the classification accuracy and easily detecting sequences with overlapped poses for merging process in the incremental unsupervised-learning framework. The experimental results showed that, in both frameworks, the proposed appearance manifold with view-dependent covariance matrix method could recognize faces from video sequences accurately.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E92.D.642/_p
부
@ARTICLE{e92-d_4_642,
author={Lina, Tomokazu TAKAHASHI, Ichiro IDE, Hiroshi MURASE, },
journal={IEICE TRANSACTIONS on Information},
title={Incremental Unsupervised-Learning of Appearance Manifold with View-Dependent Covariance Matrix for Face Recognition from Video Sequences},
year={2009},
volume={E92-D},
number={4},
pages={642-652},
abstract={We propose an appearance manifold with view-dependent covariance matrix for face recognition from video sequences in two learning frameworks: the supervised-learning and the incremental unsupervised-learning. The advantages of this method are, first, the appearance manifold with view-dependent covariance matrix model is robust to pose changes and is also noise invariant, since the embedded covariance matrices are calculated based on their poses in order to learn the samples' distributions along the manifold. Moreover, the proposed incremental unsupervised-learning framework is more realistic for real-world face recognition applications. It is obvious that it is difficult to collect large amounts of face sequences under complete poses (from left sideview to right sideview) for training. Here, an incremental unsupervised-learning framework allows us to train the system with the available initial sequences, and later update the system's knowledge incrementally every time an unlabelled sequence is input. In addition, we also integrate the appearance manifold with view-dependent covariance matrix model with a pose estimation system for improving the classification accuracy and easily detecting sequences with overlapped poses for merging process in the incremental unsupervised-learning framework. The experimental results showed that, in both frameworks, the proposed appearance manifold with view-dependent covariance matrix method could recognize faces from video sequences accurately.},
keywords={},
doi={10.1587/transinf.E92.D.642},
ISSN={1745-1361},
month={April},}
부
TY - JOUR
TI - Incremental Unsupervised-Learning of Appearance Manifold with View-Dependent Covariance Matrix for Face Recognition from Video Sequences
T2 - IEICE TRANSACTIONS on Information
SP - 642
EP - 652
AU - Lina
AU - Tomokazu TAKAHASHI
AU - Ichiro IDE
AU - Hiroshi MURASE
PY - 2009
DO - 10.1587/transinf.E92.D.642
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E92-D
IS - 4
JA - IEICE TRANSACTIONS on Information
Y1 - April 2009
AB - We propose an appearance manifold with view-dependent covariance matrix for face recognition from video sequences in two learning frameworks: the supervised-learning and the incremental unsupervised-learning. The advantages of this method are, first, the appearance manifold with view-dependent covariance matrix model is robust to pose changes and is also noise invariant, since the embedded covariance matrices are calculated based on their poses in order to learn the samples' distributions along the manifold. Moreover, the proposed incremental unsupervised-learning framework is more realistic for real-world face recognition applications. It is obvious that it is difficult to collect large amounts of face sequences under complete poses (from left sideview to right sideview) for training. Here, an incremental unsupervised-learning framework allows us to train the system with the available initial sequences, and later update the system's knowledge incrementally every time an unlabelled sequence is input. In addition, we also integrate the appearance manifold with view-dependent covariance matrix model with a pose estimation system for improving the classification accuracy and easily detecting sequences with overlapped poses for merging process in the incremental unsupervised-learning framework. The experimental results showed that, in both frameworks, the proposed appearance manifold with view-dependent covariance matrix method could recognize faces from video sequences accurately.
ER -