i-m-Flashlight
Enabling Beyond-Surface Interactions for Interactive Surface with An Invisible Projection
Authors - Li-Wei Chan, Hsiang-Tao Wu, Hui-Shan Kao, Ju-Chun Ko, Home-Ru Lin,
Mike Y. Chen, Jane Hsu, Yi-Ping Hung
Authors Bios - Li-Wei Chan is a PhD student at the National Taiwan University and researches with the Image and Vision Lab and iAgent.
Hsiang-Tao Wu is now a researcher at Microsoft Research Asia but previously attended the National Taiwan University.
Hui-Shan Kao, Ju-Chun Ko, Home-Ru Lin were students at the National Taiwan University during the time of this paper.
Mike Y. Chen is an assistant professor at the National Taiwan University specializing in human-computer interaction and mobile computing.
Jane Hsu is a professor at the National Taiwan University and focuses on multi-agent systems and web technologies.
Yi-Ping Hung is a professor at the National Taiwan University and researches human-computer interaction among other things.
Venue - This paper was presented at the UIST '10 Proceedings of the 23rd annual ACM symposium on User interface software and technology.
Summary
Hypothesis - In this paper, researchers propose using Infrared (IR) projectors to allow 3D interactions with traditional 2D tabletop surfaces. The hypothesis is that through the use of IR projectors, invisible markers can be created to allow for an entirely new set of interactions with touch surfaces interacting in the 3rd dimension.
Content - The researchers first had to build the devices to be used seeing as most of what is needed is not currently available on the market. The IR projector was built by upgrading a DLP projector and the table had to be made with the diffuser layer above the touch-glass because the touch-glass would reflect on observers and degrade the light. Detecting multi-touch was implemented using a combination of touch recognition and the Diffused-Illumination(DI) method in a clever way by simulating backgrounds, inspecting suspected objects in a region of interest (ROI) by projecting white regions onto it, and smoothing out feedback with a Kalman filter. Due to the camera and projectors being independent systems, it was necessary to use software synchronization to keep the cameras from feeding old data to the projectors.
Methods - 3 applications were developed including:
1) i-m-Lamp - Ordinary lamp structure with IR camera and pico-projector instead of a light bulb. Users move the lamp to point at an area of interest like a particular part of a map. The tabletop surface masks its projection at that location and the pico-projector takes over by projecting the same data shown on the tabletop beforehand only with more detail than before.
2) i-m-Flashlight - Mobile version of the i-m-Lamp allowing for quicker examination of certain areas of interest like those found on paintings.
3) i-m-View - A 3D viewer that lets a tablet, or tablets, view a 3D model of whatever 2D image is being shown on the tabletop surface. The boundaries of the tabletop surface are clearly denoted on these tablet devices to keep users aware of the system they are using.
Results - 5 users were asked to use the 3 systems to perform certain actions and provide feedback:
1) The i-m-Lamp was found to be the most stable of the devices and proper use was described as moving the lamp to a desired location to then search nearby for information.
2) The i-m-Flashlight was used more quickly and was found to be effective in finding information from a variety of locations in quick succession but a problem was encountered if users moved the device too close or too far from the table.
3) The i-m-View became lost when users tried to look up at tall buildings focusing the device away from the tabletop. Users also wanted touch actions enabled on the tablet for easy manipulation as well as a portrait mode. The i-m-View also immersed the users too much in the tablet, as expected, and made them unaware of the real scene on the table.
Conclusion - The researchers conclude by noting that they had accomplished the goal of creating a 3D interactive environment through the use of IR cameras and projectors. They also discussed future work in fixing problems found during testing and adding more features requested by users.
Discussion
I think the researchers accomplished their goal of creating a 3D interactive surface using invisible markers and the applications helped prove that but where this would fit in in the general public remains a mystery. This system is too bulky by itself and would need a good reason to exist. I can see this applied to Air Traffic Control systems. The table could display general airplane symbols and show direction but to get more data, the controller can highlight it with a device similar to the i-m-Flashlight. Because of things like this, I can see this technology developing and maybe even finding a niche somewhere.
No comments:
Post a Comment