Kinected fitting room, robot fish, evaluating the balance of the elderly and more from the Kinect for Windows China Workshop 2011
- Posted: Feb 14, 2012 at 6:00AM
A few months ago we highlighted a series of inspiration Kinect projects from students and faculty in China, The Kinect knows no national boundaries, as shown by Chinese University students pushing the Kinect's boundaries.
Yesterday in, Students, now is your chance to join the Kinect Effect!, I asked if the you were ready to use the Kinect to help change the world.
Today we're going to show you some ideas and thinking about how to do just that...
On December 2, 2011, Microsoft Research Asia held the Kinect for Windows Workshop 2011 in Beijing, China. The event, which drew more than 100 participants, including faculty and students from Chinese universities, provided a forum for exploring research that utilizes Kinect for Windows. It not only offered a great opportunity for faculty members and students to showcase their Kinect-based research and exchange creative ideas, it also fostered enhanced cooperation between Chinese academic institutions and Microsoft Research Asia.
The student group attended a number of sessions tailored specifically for them, including a speech on computer art, delivered by Tsinghua University Professor Yingqing Xu, and an explanation of the operating principles of Kinect, presented by DJ Lan of Microsoft Asia R&D. Two Microsoft Research Asia interns also shared their Kinect application development experience with the student participants and joined them in hands-on projects.
The demo session generated the most excitement, and featured 15 booths of posters, videos, and demo programs for Kinect projects. The booths were organized by professors and students who delivered detailed demonstrations of their projects. The demos attracted many attendees, including Microsoft Research Asia staff members who participated in discussions with professors and students, and were inspired by their innovative ideas.
The workshop also inspired faculty and students, who left with a better understanding of the possibilities for Kinect-based research. In addition, the workshop bolstered opportunities for future collaboration between Microsoft Research Asia and the Chinese academic community. As Lolan Song, the senior director of Microsoft Research Asia observed, “It’s a great opportunity for Microsoft Research Asia to strengthen communication and collaboration with faculty and students. Microsoft Research Asia is committed to exploring more qualified research projects with Chinese universities and academic institutions, as we believe such collaboration will have long-term social benefits.”
“Writing in the Air by Hands- recognition of virtual handwritten characters using Kinect” is developed by the South China University of Technology.
With no keyboard, no mouse, no handwriting screen or white board, the users of Kinect are faced with a problem when there is a need to input text.
This project tries to solve this problem through the mode of human-computer interaction by giving the users a way to write in the air. Through the natural use of our figures to write in the air, Kinect’s depth sensor, RGB image sensor, we were able to gain dual-channel information and track the finger movements of people in real time. Then we could identify the Chinese character through trajectory analysis. This provides the users with a type of new handwriting input interactive experience which is controlled by body sensory.
“A system for evaluating the balance function of the elderly” is developed by BeiHang University. Instead of make Kinect into equipment for human-computer interaction, this project introduces it into the field of health assessment.
Compared with professional equipment, Kinect has the advantages of low price, stability and convenience. This created the possibility that some of the motion analysis that used to carry out in specialized laboratories, could be carried out in the community or at home. This project is just one preliminary trail of using Kinect in the health field. In the future, Kinect will surely play a more important role in health assessment and assisted rehabilitation.
“Robot Fish Controlled by Kinect” is developed by Peking University.
At present, many systems have complicated controlling operations, and it is very hard for common people to learn how to control these things. After the release of Kinect, we can take very natural human movements and turn it into robot and system directives.
This “control system” is what this subject combines together. In addition, because Kinect just takes human gestures and reflects them, this project provides an idea that we can provide people with feedback on the environment through the robot controlled fish.
We believe that in the future robots will have applications that are very useful in various aspects which will at the same time make our lives better.
“Kinect based Fitting Room” is developed by the Beijing Institute of Clothing Technology. This project is about the use of virtual fitting-room system.
Virtual fitting-room research has been conducted for a long time and Kinect had greatly benefited the work.
In this project, Kinect is mostly used in two places; the first is to make virtual coat-hangers and fitting-room avatars. The second is to collect human stances and movement for interactive virtual fittings. On the virtual platform this project designed you could enjoy the same experience as in the real stores.
“Scanning 3D Full Human Bodies using Kinects” is developed by Zhejiang University. This project is focused on using Kinects to create a three-dimensional full body scan.
Kinect could obtain in-depth data while one has to be far away if he wants to capture data from the entire body and the data collected in this way is relatively poor. And if one wants to be closer and use multiple Kinects, there will be interference. This project provides a kind of set-up that could allow one to be close to the equipment, but at the same time gather relatively good quality data.
This project also solves the problem of collecting data from a non-rigid body. Registration for two non-rigid substances can be realized easily, however, when the subject turns around there will be a cycle consistency problem and some parts of the body will be blocked. And with the global registration, we can create a good quality reconstruction of the human body.