Jing Yang

Inertial sensor based recognition of 3-D character gestures with an ensemble classifiers

We present a 3-D input medium based on inertial sensors for on-line character recognition and an ensemble classification scheme for the recognition task. The system allows user to write a character in the air as a gesture, with a sensor-embedded device held in hand. The kinds of sensors used are 3-axis accelerometer and 3-axis gyroscope generating acceleration and angular velocity signals respectively. For character recognition, we used the technique of FDA (Fisher discriminant analysis). We tried different combinations of sensor signals to test the recognition performance. It is also possible to estimate a 2-D handwriting trajectory from the sensor signals. The best recognition rate of 93.23%, in case we use only raw sensor signals, was attained when all 6 sensor signals were combined. The recognition rate of 92.22% was reached if the estimated trajectory was used as input. Finally we tested the ensemble method and the generalization rate of 95.04% was attained on the ensemble recognizer consisting of 3 FDA recognizers based on acceleration-only, angular-velocity-only and handwriting trajectory respectively.

A novel hand gesture input device based on inertial sensing technique

In this paper, we present a novel gesture-based input device by using inertial sensing technique. The trajectories of users hand-drawn gestures in 3D space are captured and recognized by this device to fulfill user interaction task. The proposed device employs gyro-free inertial measurement unit (IMU) to track hand motions without requiring any external reference sensors or signals. Since the unbounded growing error of trajectory estimation, as a major drawback of IMU-based motion tracking technology, is carefully solved by using zero velocity compensation. Here, a deliberately-designed motion detection scheme is proposed to capture accurate hand motion period. Finally, the recognition algorithm based on Bayesian networks takes estimated trajectories and finds the corresponding gesture model with the maximum probability. Because the IMU provides outstanding capability of self-contained positioning, the proposed device is extraordinary simple and effective, comparing with the devices using other tracking technologies such as vision-based system. Experimental results also show its effectiveness and feasibility. Currently, after employing the trajectory estimation method provided in this paper, the recognition rate of 95.51% for 14 gestures has been achieved when this device is implemented as a TV remote controller. It can be used as a powerful, flexible interface for ubiquitous computing device.

Gesture based Input Device: An All Inertial Approach

In this paper, we develop a gesture-based input device equipped with accelerometers and gyroscopes. The sensors measure the inertial measurements, i.e., accelerations and angular velocities produced by the movement of the system when a user is inputting gestures on a plane surface or in a 3D space. The gyroscope measurements are integrated to give orientation of the device and consequently used to compensate the accelerations. The compensated accelerations are doubly integrated to yield the position of the device. With this approach, a users gesture input trajectories can be recovered without any external sensors. Three versions of motion tracking algorithms are provided to cope with wide spectrum of applications. Then, a Bayesian network based recognition system processes the recovered trajectories to identify the gesture class. Experimental results convincingly show the feasibility and effectiveness of the proposed gesture input device. In order to show practical use of the proposed input method, we implemented a prototype system, which is a gesture-based remote controller (Magic Wand).