Discriminating features learning in hand gesture classification

Abstract

The advent and popularity of Kinect provides a new choice and opportunity for hand gesture recognition (HGR) research. In this study, the authors propose a discriminating features extraction for HGR, in which features from red, green and blue (RGB) images and depth images are both explored. More specifically, histogram of oriented gradient feature, local binary pattern feature, structure feature and three‐dimensional voxel feature are first extracted from RGB images and depth images, then these features are further reduced with a novel deflation orthogonal discriminant analysis, which enhances the discriminative ability of the features with supervised subspace projection. The extensive experimental results show that the proposed method improves the HGR performance significantly.

Recognition of complex static hand gestures by using the wristband‐based contour features

“…For hand region segmentation, skin-colour detections [8][9][10] and haar-like features [18] are frequently used by many researchers. Kinect depth camera can be utilised to better locate and extract the hand region [4,[15][16][17]19]. However, these methods require additional cost on sensors.…”

Towards a classification of surgical skills using affine velocity

“…In fact, this kind of characterisation allows to find the similarities and differences and classify between several trajectories in 3D space [13,14]. Most of developed algorithms focuses on simple variables (spatial and temporal) to get a better classification for different kind of trajectories [15][16][17]. In [18], an overview of these techniques, on surgical skills assessment, is provided.…”

Structured entropy of primitive: big data‐based stereoscopic image quality assessment

“…The well-known matching pursuit family (MPF) algorithms have been proposed for solving the sparse decomposition problem [26]. The orthogonal matching pursuit (OMP) [27] is one of the MPFs that works in a greedy fashion.…”