Handshape Recognition Using Skeletal Data

Abstract

In this paper, a method of handshapes recognition based on skeletal data is described. A new feature vector is proposed. It encodes the relative differences between vectors associated with the pointing directions of the particular fingers and the palm normal. Different classifiers are tested on the demanding dataset, containing 48 handshapes performed 500 times by five users. Two different sensor configurations and significant variation in the hand rotation are considered. The late fusion at the decision level of individual models, as well as a comparative study carried out on a publicly available dataset, are also included.

Data Augmentation with Suboptimal Warping for Time-Series Classification

“…In future work, deep learning architectures for time-series classification [32], data augmentation near class boundaries, or augmentation addressing a specific domain (e.g., captured human actions or sign language gestures [33,34]) will be considered.…”

Special Issue on Visual Sensors

“…During the past few years, many proposals have been made in this area including different methods that allow the recognition of the objects present in an image. In this way, Kapuscinski et al [20] presented a method for hand shapes recognition based on skeletal data. It encodes the relative differences between vectors associated with the pointing direction of the particular fingers and the palm normal.…”

Hand Posture Recognition Using Skeletal Data and Distance Descriptor

“…It is expected that sensors for the acquisition of hands skeletal data will be improved soon. Therefore, work is underway to use them to recognize sign languages: American [15][16][17][18][19][20][21][22][23][24], Arabic [25][26][27][28], Australian [13], Indian [29][30][31], Mexican [32], Pakistani [33], and Polish [34].…”

“…In [34], 48 static hand shapes from the Polish Finger Alphabet and Polish Sign Language were recognized. Gestures were shown 500 times by five users.…”

“…• Most authors do not provide data used in experiments, which makes verification and comparative analysis difficult. Only three datasets available in works [15,18,24,34] are known to the authors. • Many cited works omit tests using the leave-one-subject-out protocol.…”