Three-dimensional action recognition using volume integrals

被引:4
作者
Diaz-Mas, Luis [1 ]
Munoz-Salinas, Rafael [1 ]
Madrid-Cuevas, F. J. [1 ]
Medina-Carnicer, R. [1 ]
机构
[1] Univ Cordoba, Dept Comp & Numer Anal, E-14071 Cordoba, Spain
关键词
Action recognition; View invariance; Multi-camera; Motion descriptor; POSTURE CLASSIFICATION;
D O I
10.1007/s10044-011-0239-5
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This work proposes the volume integral (VI) as a new descriptor for three-dimensional action recognition. The descriptor transforms the actor's volumetric information into a two-dimensional representation by projecting the voxel data to a set of planes that maximize the discrimination of actions. Our descriptor significantly reduces the amount of data of the three-dimensional representations yet preserves the most important information. As a consequence, the action recognition process is greatly speeded up while achieving very high success rates. The method proposed is therefore especially appropriate for applications in which limitations of computing power and space are significant aspects to consider, such as real-time applications or mobile devices. Additionally, the descriptor is sensitive to reflected actions, i.e., same actions performed with different limbs can be differentiated. This paper tests the VI using several Dimensionality Reduction techniques (namely PCA, 2D-PCA, LDA) and different Machine Learning approaches (namely Clustering, SVM and HMM) so as to determine the best combination of these for the action recognition task. Experiments conducted on the public IXMAS dataset show that the VI compares favorably with state-of-the-art descriptors both in terms of classification rates and computing times.
引用
收藏
页码:289 / 298
页数:10
相关论文
共 50 条
[21]   Action Recognition Using Form and Motion Modalities [J].
Meng, Quanling ;
Zhu, Heyan ;
Zhang, Weigang ;
Piao, Xuefeng ;
Zhang, Aijie .
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS, 2020, 16 (01)
[22]   Using SAX representation for human action recognition [J].
Junejo, Imran N. ;
Al Aghbari, Zaher .
JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION, 2012, 23 (06) :853-861
[23]   Action Recognition and Benchmark Using Event Cameras [J].
Gao, Yue ;
Lu, Jiaxuan ;
Li, Siqi ;
Ma, Nan ;
Du, Shaoyi ;
Li, Yipeng ;
Dai, Qionghai .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2023, 45 (12) :14081-14097
[24]   Human Action Recognition using Topic Model [J].
Xiao, Qian ;
Cheng, Jun .
2014 4TH IEEE INTERNATIONAL CONFERENCE ON INFORMATION SCIENCE AND TECHNOLOGY (ICIST), 2014, :694-697
[25]   Action Recognition using Time of Flight Cameras [J].
Oprisescu, Serban ;
Burlacu, Constantin ;
Buzuloiu, Vasile .
PROCEEDINGS OF THE 2010 8TH INTERNATIONAL CONFERENCE ON COMMUNICATIONS (COMM), 2010, :153-156
[26]   Action recognition using dynamic hierarchical trees [J].
Wang, Tingwei ;
Duan, Peng ;
Ma, Bingxian ;
Wu, Peng ;
Lu, Weizhi .
JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION, 2019, 61 :315-325
[27]   Action recognition using linear dynamic systems [J].
Wang, Haoran ;
Yuan, Chunfeng ;
Luo, Guan ;
Hu, Weiming ;
Sun, Changyin .
PATTERN RECOGNITION, 2013, 46 (06) :1710-1718
[28]   Action recognition using polyhedron neighborhood features [J].
Yang, Jiangfeng ;
Ma, Zheng .
International Journal of Multimedia and Ubiquitous Engineering, 2015, 10 (01) :391-402
[29]   Human Action Recognition using Skeleton features [J].
Patil, Akash Anil ;
Swaminathan, A. ;
Rajan, Ashoka R. ;
Narayanan, Neela, V ;
Gayathri, R. .
2022 IEEE INTERNATIONAL SYMPOSIUM ON MIXED AND AUGMENTED REALITY ADJUNCT (ISMAR-ADJUNCT 2022), 2022, :289-296
[30]   ACTION RECOGNITION USING SALIENT NEIGHBORING HISTOGRAMS [J].
Ren, Huamin ;
Moeslund, Thomas B. .
2013 20TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP 2013), 2013, :2807-2811