An extension of kernel learning methods using a modified Log-Euclidean distance for fast and accurate skeleton-based Human Action Recognition

Abstract

In this article, we introduce a fast, accurate and invariant method for RGB-D based human action recognition using a Hierarchical Kinematic Covariance (HKC) descriptor.Recently, non singular covariance matrices of pattern features which are elements of the space of Symmetric Definite Positive (SPD) matrices, have been proven to be very efficient descriptors in the field of pattern recognition.However, in the case of action recognition, singular covariance matrices cannot be avoided because the dimension of features could be higher than the number of samples. Such covariance matrices (non singular and singular) belong to the space of Symmetric Positive semi-Definite (SPsD) matrices.Thus, in order to classify actions, we propose to adapt kernel methods such as Support Vector Machines (SVM) and Multiple Kernel Learning (MKL) to the space of SPsD matrices by using a perturbed Log-Euclidean distance (Arsigny et al., 2006). The mathematical validity of this perturbed distance (called Modified Log-Euclidean distance) for SPsD is therefore studied.The offline experiments are conducted on three challenging benchmarks, namely MSRAction3D, UTKinect and Multiview3D datasets. A fair comparison demonstrates that our approach competes with state-of-the-art methods in terms of accuracy and computational latency. Finally, our method is extended to an online scenario and experiments on MSRC12 prove the efficiency of this extension. (Enjie Ghorbel) can be found in (Poppe, 2010;Weinland et al., 2011). Unfortunately, these methods suffer from some limitations: their performance is negatively affected by occlusions, view-point variation, illumination changes and body segmentation.With the availability of low-cost RGB-D cameras, a renewed interest for action recognition has been observed. Additionally to the classical RGB images, this kind of camera also provides depth images. Furthermore, the recent algorithm proposed by Shotton et al. (2013) allows the real-time human skeleton extraction from depth maps. Thus, new methods (Rahmani et al., 2016;Amor et al., 2016;Brun et al., 2016;Liu et al., 2016) have been pro-

Application of Artificial Intelligence in College Dance Teaching and Its Performance Analysis

“…Then the Euclidean distance [16][17][18] between the j-th performance analysis indicator of the i-th performance analysis object and the maximum ideal value * can be expressed as:…”

Two-Stage RGB-Based Action Detection Using Augmented 3D Poses

“…In this section, we propose to use the estimated 3D skeleton sequences P 3D as in (6) for carrying out action detection. To overcome view-point variability, a pre-processing of skeleton alignment is performed by estimating a linear transformation matrix between the absolute coordinate system and a local coordinate system defined using the spine and the hip joints of the skeleton as in [5,8]. To validate the use of 3D skeletons estimated from RGB images, a 3D skeleton-based approach for action detection is introduced.…”

Temporal 3D Human Pose Estimation for Action Recognition from Arbitrary Viewpoints

“…As our focus is to infer the action, we hence normalize the estimated 3D skeletons of the same action in order to relate them to the same reference frame. To that end, we follow the same normalization process proposed in [21] to eliminate the anthropometric variability.…”