A Seminar On Accelerometer Based Gesture Recognition 6r254c

ACCELEROMETER BASED GESTURE-RECOGNITION

Seminar By: Sourav Kumar Jena 0911015158

COURSE OF ACTION Introduction To Accelerometers Brief Idea On Working Of Accelerometers Example Of Accelerometer Applications In Cellphones And Tablets / Vehicle Crash Detection / Gesture Control Introduction To uWave Algorithm Design  Quantization Of Acceleration Data  Dynamic Time Warping  Template Adaption

 Authentication 3D Mobile Interface Conclusion Reference

INTRODUCTION What is an accelerometer? An accelerometer is an electromechanical device that will measure acceleration forces. These forces may be static, like the constant force of gravity pulling at your feet, or they could be dynamic caused by moving or vibrating the accelerometer.

HOW DO ACCELEROMETERS WORK?  There are many different ways to make an accelerometer! Some accelerometers use the piezoelectric effect .Other ways to do this is by using resistive, capacitive, servo and other mechanism that get stressed by accelerative forces, which causes a voltage to be generated.

Original Accelerometer Patent

Piezo-based Accelerometers

Accelerometer - Example  DE-ACCM6G Buffered ±6g Accelerometer

• • •

Dimension Engineering Has ±6g sense range 222 mV/g sensitivity

VARIOUS APPLICATIONS

Accelerometers find applications in fields like  In Cellphones And Tablets  Vehicle Crash Detection  GESTURE CONTROL

An accelerometer as a sensor measures the tilting motion and orientation of a mobile phone thus facilitating playing of games on our phones and tabs.

One of the most common uses for accelerometers is in airbag deployment systems for modern automobiles. In this case the accelerometers are used to detect the rapid negative acceleration of the vehicle to determine when a collision has occurred and the severity of the collision.

Airbag Sensor Accelerometers

Gesture recognition helps in interpreting human gestures via mathematical algorithms. Gestures can originate from any bodily motion or state but commonly originate from the face or hand. Gesture recognition enables humans to communicate with the machine (HMI) and interact naturally without any mechanical devices. Using the concept of gesture recognition, it is possible to point a finger at the computer screen so that the cursor will move accordingly. So we will be learning how to control devices using gestures and using them as s.

Introduction to uWave  uWave is an efficient gesture recognition method based on a single accelerometer using dynamic time warping (DTW). It requires a single training sample per vocabulary gesture.  Two applications of uWave namely gesture-based authentication and gesture-based manipulation of threedimensional interfaces on mobile phones have been discussed  With uWave, s can create simple personal gestures for frequent interaction. Its simplicity, efficiency, and minimal hardware requirement (a single accelerometer) make uWave have the potential to enable personalized gesture-based interaction with a broad range of devices.

UWAVE ALGORITHM DESIGN The premise of uWave is that human gestures can be characterized by the time series of forces applied to the handheld device. Therefore,

uWave bases the recognition on the matching of two time series of forces, measured by a single three-axis accelerometer. The key technical components of uWave discussed are: - Acceleration Quantization - Dynamic Time Warping - Template Adaption

The input to uWave is a time series of acceleration provided by a three-axis accelerometer. Each time sample is a vector of three elements, corresponding to the acceleration along the three axes. uWave first quantizes acceleration data into a time series of discrete values. The same quantization applies to the templates too. It then employs DTW to match the input time series against the templates of the gesture vocabulary. It recognizes the gesture as the template that provides the best matching.

QUANTIZATION OF ACCELERATION DATA uWave quantizes the acceleration data before template matching. Quantization reduces the length of input time series for DTW in order to improve computation efficiency. It also converts the accelerometer read. Quantization improves recognition accuracy by removing variations not intrinsic to the gesture, e.g. accelerometer noise and minor hand tilting into a discrete value thus reduces floating point computation.

Dynamic Time Warping Dynamic time warping (DTW) is an algorithm for measuring similarity between two sequences which may vary in time or speed. is a classical

algorithm based on dynamic programming to match two time series with temporal dynamics . uWave employs the Euclidean distance for matching quantized time series of acceleration.

Template Adaptation uWave’s objective is not to explore the most effective adaptation methods but to demonstrate the template adaptation can be easily implemented and effective in improving recognition accuracy over multiple days. template adaptation works as follows. uWave keeps two templates generated in two different days for each vocabulary gesture. It matches a gesture input with both templates of each vocabulary gesture and take the smaller matching cost of the two as the matching cost between the input and vocabulary gesture. Two updating schemes are adapted POSITIVE UPDATE and NEGATIVE UPDATE. If both templates for a vocabulary gesture in the library are at least one day old and the input gesture is correctly recognized, the older one will be replaced by the newly correctly recognized input gesture this is used in positive update. In negative update we replace the older template with the input gesture when it is incorrectly recognized.

POSITIVE UPDATE

NEGATIVE UPDATE

This figure summarizes the recognition results averaged across all data collected. It shows an accuracy of 97.4%(avg.) for Positive Update and 98.6%(avg.) for Negative Update.

Gesture-based Light-Weight Authentication

Outside the traditional realm of -based strong authentication, there is a need for light-weight authentication techniques that prioritize ease of use over hard security. For privacy-insensitive -specific data, this manner of lightweight, ‘soft’ authentication provides a mechanism for a to personalize the device. The objectives are: 1) accurate recognition of a and 2) to be friendly, easy to and easy to perform. uWave enables authentication based on physical manipulation of the device with low cost and high efficiency. It is particularly suitable for implementation on resource-constrained devices, such as mobile phones and TV remotes.

Gesture-based 3D Mobile Interface

One of the strengths of uWave is that it can recognize three dimensional hand movement. It has been shown that it is intuitive and convenient to navigate a 3D interface with 3D hand gestures. In order to explore this, we used a 3D-mobile application(video-sharing service under development within Motorola) and integrated uWave with it to enable gesture-based navigation. The interface shows a rotating ring that contains thumbnails of various videos. uWave to navigate this interface using a series of specific movements such as tilting and slight shaking, which are more appropriate for a mobile device when the is focused on the screen. Even when the 3D rendering consumes a significant amount of memory, uWave works smoothly with it, without introducing any human perceptible performance degradation.

CONCLUSION We present uWave for interaction based on personalized gestures and physical manipulations of a consumer electronic or mobile device. uWave employs a single accelerometer so it can be readily implemented on many commercially available consumer electronics and mobile devices. Its simplicity and efficiency allow implementation on a wide range of devices.

REFERENCES  Popova et al ,Micromechanical gyros & accelerometers for digital navigation & control systems, IEEE V24, I5 Year: 2009 ,Page(s): 33- 39  I. J. Jang et al ,Signal processing of the accelerometer for gesture awareness on handheld devices IEEE Int. Wkshp. Ed., 2003, pp. 139-144  Matt Knapp, Accelerometer-based Personalized Gesture Recognition, Worcester Polytechnic Institute

Thank You