In this paper, we handle the problem of human action recognition by combining covariance matrices as local spatio-temporal (ST) descriptors and local ST features extracted densely from action video. Unlike traditional methods that separately utilizing gradient-based feature and optical flow-based feature, we use covariance matrix to fuse the two types of feature. Since covariance matrices are Symmetric Positive Definite (SPD) matrices, which form a special type of Riemannian manifold. To measure the distance of SPDs while avoid computing the geodesic distance between them, covariance features are transformed to log-Euclidean covariance matrices (LECM) by matrix logarithm operation. After encoding LECM by Locality-constrained Linear Coding method, in order to provide position information to ST-LECM features, spatial pyramid is used to partition the video frames, and the average-pooling-on-absolute-value function is implemented over each sub-frames. Finally, non-linear support vector machine is used as classifier. Experiments on public human action datasets show that the proposed method obtains great improvements in recognition accuracy, in comparison to several state-of-the-art methods.
목차
Abstract 1. Introduction 2. Spatio-Temporal Log-Euclidean Covariance Matrix (ST-LECM). 2.1. Log-Euclidean Framework on SPD Matrices 2.2. Spatio-Temporal Log-Euclidean Covariance Matrix (ST-LECM) 2.3. Encoding ST-LECM Features by LLC Method 3. The System Framework 4. Experiments 4.1. KTH Action Datasets 4.2. ADL Datasets 5. Conclusion References
Shilei Cheng [ School of Electronic Engineering University of Electronic Science and Technology of China, Xiyuan Ave, No.2006, West Hi-Tech Zone ]
Jiangfeng Yang [ School of Communication and Information Engineering, University of Electronic Science and Technology of China, Xiyuan Ave, No.2006, West Hi-Tech Zone, 61173 ]
Zheng Ma [ School of Communication and Information Engineering, University of Electronic Science and Technology of China, Xiyuan Ave, No.2006, West Hi-Tech Zone, 61173 ]
Mei Xie [ School of Electronic Engineering University of Electronic Science and Technology of China, Xiyuan Ave, No.2006, West Hi-Tech Zone ]
보안공학연구지원센터(IJSIP) [Science & Engineering Research Support Center, Republic of Korea(IJSIP)]
설립연도
2006
분야
공학>컴퓨터학
소개
1. 보안공학에 대한 각종 조사 및 연구
2. 보안공학에 대한 응용기술 연구 및 발표
3. 보안공학에 관한 각종 학술 발표회 및 전시회 개최
4. 보안공학 기술의 상호 협조 및 정보교환
5. 보안공학에 관한 표준화 사업 및 규격의 제정
6. 보안공학에 관한 산학연 협동의 증진
7. 국제적 학술 교류 및 기술 협력
8. 보안공학에 관한 논문지 발간
9. 기타 본 회 목적 달성에 필요한 사업
간행물
간행물명
International Journal of Signal Processing, Image Processing and Pattern Recognition
간기
격월간
pISSN
2005-4254
수록기간
2008~2016
십진분류
KDC 505DDC 605
이 권호 내 다른 논문 / International Journal of Signal Processing, Image Processing and Pattern Recognition Vol.9 No.2