Home | Forums | Article Search | Subscribe & Shop | Contact Us | Log Out Under Development Email This Print This View My Personal Library What's Cooking June 2005 • Vol.5 Issue 6 Page(s) 106-107 in print issue Add To My Personal Library Under Development A Peek At What's Brewing In The Laboratory A Picture Worth A Thousand (Pass)words We increasingly rely on cell phones and PDAs to store sensitive information and data, and not just phone numbers and addresses but financial records, passwords, and documents. Text passwords and key locks provide some basic security, but they are generally used inconsistently, leaving information vulnerable if the device falls into nefarious hands. Enter the OKAO Vision Face Recognition Sensor developed by Japan’s OMRON Corporation. Designed to be incorporated into camera-equipped handhelds, the biometric technology secures a device by requiring users to photograph themselves before using it. The photo is compared against faces registered in an onboard database. If a match is authenticated, the device unlocks for use in about one second. The OKAO Sensor’s requirements are modest. No additional hardware is needed beyond a camera-equipped device that is capable of capturing images at 320 x 240, and each registered photo requires only 1.5KB of storage. The software will be available for Symbian OS, BREW, embedded Linux, and ITRON OSes. Sure, the cool factor is high, especially if you enjoy photographing yourself, but does the technology really work with notoriously lame camera phone imaging? In OMRON’s testing the OKAO Vision Face Recognition Sensor produced a 99% success rate in identifying an authorized user under various lighting and focus conditions, including the ability to recognize a rotated image. The technology was recently demonstrated at the 2005 Security Show in Tokyo, but OMRON is still mum on a specific timeline when it might be widely available to consumers, claiming that availability will depend entirely on cell phone manufacturers. Is It Live Or . . . ? The blending of live-action video and computer-generated images is so commonplace these days, it’s almost cliché. Adding computer effects to film in the studio is a no-brainer, but a live broadcast is a much more difficult prospect. Researchers at Oxford’s Department of Engineering Science are developing a new camera navigation technology that will allow computer-generated images to be added to live video footage in real time. Existing systems that superimpose graphics over live-action video require a lot of calibration ahead of time, with external markers placed on the scene to ensure accurate positioning of the overlaid graphics (a la SportsVision’s yellow first-down line seen during football broadcasts). In contrast, the Oxford system requires only one object of known size to be placed in the scene for calibration. From there it can build a detailed visual map on the fly, using existing landmarks as reference points. The system uses a handheld camera connected to a computer. As the camera moves, software measures how the distance from the camera to the reference points changes, letting it accurately estimate the camera’s location and movement in relation to the surrounding environment, a process called SLAM (Simultaneous Localization and Mapping). With this data graphics can be accurately overlaid onto the live scene. “This localization and mapping technology turns a camera into a flexible, real-time position sensor,” says Dr. Andrew Davidson, a principal researcher on the project. In addition to TV and video entertainment—think of adding graphics to broadcasts of sporting events, concerts, and other live events—the technology could be used to virtually redecorate a room, adding furniture and more to the image your DV camcorder captures. The researchers are also working with Japan’s AIST Research Institute to use the technology in a navigation system for humanoid robots. Take A Look Around Wish you could browse a favorite Swedish furniture store without having to deal with hundreds of other shoppers doing the same? Ever arrived at your vacation resort only to find that what looked quaint and quiet in the brochures was actually right next to a noisy six-lane highway? If so, take heart. Researchers at the Fraunhofer Institute for Telecommunications in Berlin are developing a system that could put you virtually in the middle of these scenes, ending vacation planning mishaps and letting you choose your Billy bookcase in peace. The Virtual Reality Showroom can display physical environs as high-resolution, cylindrical, 360-degree panoramas, providing a complete view of a virtual scene all around its axis. Although such panoramas are nothing new—witness the “virtual reality home tours” now seen on most realty sites—the Fraunhofer Institute’s system expands on existing technology, using image data that allows for the embedding of video, interactive scene elements, and 3D audio. The system includes several algorithms to optimize virtual scenes. One algorithm can stitch together video and digital images seamlessly, adjusting lighting and exposure as needed. Another can insert still images in scenes, thus creating action, say of a crowd of people or cars passing by. Using MPEG-4 BIFS (Binary Format for Scenes) architecture allows for each of the multimedia elements to be processed as separate, individual objects, thus decreasing access time as only the elements being currently viewed are loaded. The addition of these elements brings the environments to life and lets users interact with virtual scenes in ways not possible before. For example, in a showroom scene, a shopper could filter out those items that are not of a certain brand or style, zoom in on an item of interest and view it from several angles, or access additional product information that the vendor has embedded. “The customer can interactively create a personalized shopping atmosphere among the goods that are on show,” said Fraunhofer’s Peter Eisert in a recent release. The travel industry could also benefit from application of this technology. An agent might use such a system to give customers a walkthrough tour of their intended destination, with 3D audio letting them hear just how loud the surf (or interstate) is from their suite. Tabula Digital Imagine if you and a colleague could sit at an electronic tabletop, transfer a document to it from your laptop, edit it on the table’s surface using fingers or a stylus, and then transfer the document back to your laptop and be on your way. Making this a reality is the focus of the UbiTable project at MERL (Mitsubishi Electric Research Laboratories) in Cambridge, Mass. Chia Shen, senior research scientist and associate director at the lab, and her team are developing tools that will let users work with digital docs and images on an interactive tabletop as they might with hard copies on a traditional table. Shen says one challenge in bringing a digital tabletop to fruition is in the area of user interface and computer/human interaction. “Once people can sit or stand around a tabletop, and when everyone around the table can interact [with digital content] in parallel and simultaneously, the rules of the game change drastically.” To that end, the team has developed the UbiTable (Ubiquitous Table). Still in the prototype stage, the Ubi-Table software provides an interface that lets users move content fluidly and easily from personal devices to the shared workspace of the interactive tabletop. Once on the table, a finger slide can move and rotate the content. The content can also be marked up with digital ink, resized, and copied. The Mitsubishi Diamond-Touch, a “simultaneous, multi-user, touch-sensitive input device” scheduled to be commercially available in May 2005, runs the UbiTable. Four users can connect their notebooks via USB, and the table tracks which user is touching where via embedded antennas that transmit a unique signal to a receiver on the user’s chair. Beyond workplace applications the team sees a time when interactive tabletops could be available for walkup use in public spaces, such as cafes and airports or anyplace people meet for impromptu face-to-face collaboration. by William Van Winkle Want more information about a topic you found of interest while reading this article? Type a word or phrase that identifies the topic and click "Search" to find relevant articles from within our editorial database. Enter A Subject (key words or a phrase): ALL Words (‘digital’ AND ‘photography’) ANY Words (‘digital’ OR ‘photography’) Exact Match ('digital photography'- all words MUST appear together) Home      Copyright & Legal Information      Privacy Policy      Site Map      Contact Us Copyright © 2010 Sandhills Publishing Company U.S.A. All rights reserved.