Improving Human Motion Classification by Applying Bagging and Symmetry to PCA-Based Features

Abstract

This paper proposes a method for improving human motion classification by applying bagging and symmetry to Principal Component Analysis (PCA)-based features. In contrast to well-known bagging algorithms such as random forest, the proposed method recalculates the motion features for each “weak classifier” (it does not randomly sample a feature set). The proposed classification method was evaluated on a challenging (even to a human observer) motion capture recording dataset of martial arts techniques performed by professional karate sportspeople. The dataset consisted of 360 recordings in 12 motion classes. Because some classes of these motions might be symmetrical (which means that they are performed with a dominant left or right hand/leg), an analysis was conducted to determine whether accounting for symmetry could improve the recognition rate of a classifier. The experimental results show that applying the proposed classifiers’ bagging procedure increased the recognition rate (RR) of the Nearest-Neighbor (NNg) and Support Vector Machine (SVM) classifiers by more than 5% and 3%, respectively. The RR of one trained classifier (SVM) was higher when we did not use symmetry. On the other hand, the application of symmetry information for bagged NNg improved its recognition rate compared with the results without symmetry information. We can conclude that symmetry information might be helpful in situations in which it is not possible to optimize the decision borders of the classifier (for example, when we do not have direct information about class labels). The experiment presented in this paper shows that, in this case, bagging and mirroring might help find a similar object in the training set that shares the same class label. Both the dataset that was used for the evaluation and the implementation of the proposed method can be downloaded, so the experiment is easily reproducible.

The application of topological data analysis to human motion recognition

“…There is no official name for this approach; therefore, the authors may use various names for it, such as 'eigensequences' (Bottino, De Simone & Laurentini, 2007), 'signatures' (Billon, Nédélec & Tisseau, 2008) or they do not specifically mention its name at all. MoCap data that is used as an input for PCA is most often either three dimensional trajectories of body joints (Billon, Nédélec & Tisseau, 2008;Zago et al, 2017;Ko, Han & Newell, 2018;Choi, Ono & Hachimura, 2009) or angle-based features derived from the positions of those joints (Bottino, De Simone & Laurentini, 2007;Mantovani, Ravaschio, Piaggi & Landi, 2010;Choi, Sekiguchi & Hachimura, 2009;Choi, Sekiguchi & Hachimura, 2013;Das, Wilson, Lazarewicz & Finkel, 2006;Świtoński et al, 2011;Lee, Roan & Smith, 2009;Hachaj & Ogiela, 2018;Hachaj, 2019).…”

“…A classifier is applied after features calculation, for example, support vector machine (SVM) (Das, Wilson, Lazarewicz & Finkel, 2006;Zago et al, 2017;Hachaj, 2019) or the k-nearest neighbour method (Hachaj, 2019).…”

“…The data set we used for validation was an open data set containing the motion capture of three experienced Shorin-Ryu karate athletes (world and national medalists) (Github, online). This data set has already been used in research on human action classification (Hachaj, 2019) and proved to be challenging due to the complexity of the motions. Each of the three individuals on the recordings performed twelve types of karate techniques which were repeated 10 times (there are altogether 360 motion actions).…”

Evaluation of Pattern Recognition Methods for Head Gesture-Based Interface of a Virtual Reality Helmet Equipped with a Single IMU Sensor

“…As we already mentioned in Section 1.2 PCA-based features are among most popular approaches used in MoCap data analysis and classification. That approach is described for example in paper [55] so we will summarize only the basic concept.…”

A Study of Accelerometer and Gyroscope Measurements in Physical Life-Log Activities Detection Systems

“…Recent developments in the healthcare industry help patients, especially the elderly, to avoid illness, accidents and disease [ 1 ]. Such strategies have introduced monitoring devices such as wearable, vision and marker-based sensors that secure, examine and improve human life in uncertain situations [ 2 , 3 ] while patients remain mobile. Wearable technology has replaced traditional diagnostics by delivering ubiquitous access to vital patient data via smartphones and wearable sensory clothing [ 4 , 5 ].…”