Kinect for Natural User Interfaces

All these projects were developed in C#.NET using Microsoft Kinect SDK V1 and V2.

Drawing with Kinect

This is a gesture based free hand drawing application. It can detect a few gestures - rectangle, ellipse, line, free hand drawing, select, delete, etc. One can also choose different fill/border colors, and brush sizes.

Posture Recognition using Kinect

A novel posture classification system using skeletal-tracking feature of the Kinect sensor. Angular representation of the skeleton makes it very robust and avoids problems related to the typical human body occlusions and motion ambiguities.

Kinect Emulator

In this work, we have come forward with a novel design of a complete emulator for the Kinect sensor and its implementation in the .NET platform.

Presentation – Make it Natural

We developed a gesture detection engine on top of the Kinect SDK to detect voice commands and hand movement so that we can control the operation of Microsoft PowerPoint with our gestures.



Publications

Human posture identification for motion controlling applications is becoming more of a challenge. We present a posture classification system using skeletal-tracking feature of Microsoft Kinect sensor. Posture recovery is carried out by detecting the human body joints, its position, and orientation at the same time. Angular representation of the skeleton data makes the system very robust and avoids problems related to human body occlusions and motion ambiguities. The implemented system is tested on a class of relatively common postures comprising hundreds of human pose instances by different people, where our classifier shows an average accuracy of 94.9%, 96.7% and 96.9% for linear, exponential and priority based matching systems respectively.
The Microsoft Kinect sensor has brought a new era of Natural User Interface (NUI) based gaming and the associated SDK has provided access to its powerful sensors, which can be utilized in many ways, especially in research purposes. We have already seen its use in robotics, developing assistive technologies, and augmented reality, aside from gaming. Thousands of people around the world are playing with its built-in multimodal sensors, but still a complete emulator for the Kinect sensor device is lacking, thus requiring a physical device to do any experiments with it. In this work, we have come forward with a novel design of an emulator for the Kinect sensor and its implementation in the .NET platform using the Microsoft Kinect SDK. We have demonstrated the applicability of our system through detailed software design, code descriptions to incorporate this emulator in user's own code, and video demonstration of our proposed system.

Our Team

I started working with my students, but then continued to work with my supervisors - Frank and Bernd

Image

Hasan Shahid Ferdous

PhD Research Student, University of Melbourne
Image

Frank Vetere


Professor, Dept. of CIS, University of Melbourne
Image

Bernd Ploderer


Lecturer, Dept. of CIS, University of Queensland


Image

Samiul Monir


Image

Sabirat Rubya


Image

Asif Salekin




Image

Ridwan Rashid Noel


Image

Mohammad Raihanul Islam


Image

Sazzadur Rahaman




Image

Rakibul Hasan