An eigenspace-based method with a user adaptation scheme for human gesture recognition by using Kinect 3D data

Abstract

A Kinect sensor has been used in several applications and is considered a novel interaction interface between humans and computers because of the simplicity of the acquired image data. The sensing image data obtained using the Kinect sensor includes depth measurement information and, therefore, the (x,y,z)-coordinates of 3D objects. This study proposes an eigenspace-based method for recognizing human gestures by using Kinect 3D data. The developed approach, called Eigen3Dgesture, involves employing principal component analysis to derive significant eigen3Dgestures possessing the most critical information from Kinect 3D data. To facilitate establishing the eigenspace of Kinect 3D data, this study developed a frame filling algorithm before performing eigenspace construction. To enhance Eigen3Dgesture and the performance of the constructed eigenspace of Kinect 3D data, a user adaptation method for Eigen3Dgesture was developed. Fourteen classes of active gestures were subjected to human gesture recognition experiments. The experimental results demonstrate the superiority of Eigen3Dgesture with user adaptation.

Kinect based Gesture Command Recognition for Vending Machine Operations Using an Android Platform as the Machine Simulator

“…The Kinect sensor and its software development kit (Kinect SDK) can be efficiently used to simplify the conventional gesture recognition technique. Recently, studies on Kinect-based gesture recognition have been frequently seen [3][4][5][6], and Kinect-based gesture recognition applications mainly include robot automation, health-care, entertainment and surveillance. For the vending machine operation application using Kinect-based gesture recognition, extremely rare studies are done.…”

“…For the categorization technique of gesture recognition computations without the utilization of the Kinect software development kit, the Kinect-extracted human skeleton data is then efficiently utilized for the self-developed feature extraction calculations and the corresponding gesture pattern recognition estimations. In the work of [3], the gesture recognition computation is performed using an eigenspace approach where the principal component analysis to the three dimensional location data of all kernel joints is done for deriving the eigenspace of Kinect-extracted human action gestures. A hidden Markov model (HMM) approach for Kinect-based gesture recognition is proposed in [4] where the recognition accuracy of the HMM gesture model could be improved by properly adjusting HMM model parameters.…”

Three-layered hierarchical scheme with a Kinect sensor microphone array for audio-based human behavior recognition

“…Furthermore, despite the Kinect sensor's highly functional gesture recognition performance [11], video-based studies have rarely employed the Kinect sensor as the data acquisition receiver. Moreover, the Kinect sensor can be employed in audio-based human behavior recognition by embedding four microphones in a well-arranged microphone array [9,10].…”

“…With advances in pattern recognition methods, artificial intelligence algorithms, and hardware sensing devices, a satisfactory smart recognition system without a personal viewer will be realized in the near future. Microsoft Kinect, a popular sensor device [9,10], performs outstandingly in human gesture recognition [11]; it is expected to accelerate the development of automatic human behavior recognition for smart home and smart office applications.…”

Gesture Recognition by Using Depth Data: Comparison of Different Methodologies

“…A frame-filling algorithm is proposed in Ref. [27] to first align gesture data, then an eigenspace-based method (called Eigen3Dgesture) is applied for recognizing human gestures.…”