Toppan's entry into packaged multimedia includes CD-ROM and DVD products. Inan area flooded by technical standards Toppan's efforts are aimed at improvingproduct quality within the confines of the standards. For example, a troublingaspect of MPEG encoding is that viewed motion pictures suffer from "jitter," inwhich the image jumps in position slightly from frame to frame. Toppandemonstrated a system to remove jitter from MPEG-encoded video. It also hastechniques for improving the shading and tone of JPEG images to make them morerealistic.
The network library system being developed at NTT provides multimediaservices based on a broadband ATM network. The network is served by hi-fimusic, MPEG-1, MPEG-2 and digital library servers. Processing engines for voicerecognition, search, Japanese/English translation and text-to-speech areprovided. A key component in this network is a super-high definition display,at a resolution of 2048 x 2048 pixels, 24 bits/pixel operating at 60 frames/secfor video. The network library is being used for doctors' viewing of medicalimages, sightseeing tours, teleconferences and on-the-fly machine translationbetween Japanese and English.
Electronic commerce is viewed as being one of the promising opportunities inthe 21st century. Major concerns in making this feasible are guaranteeingsecurity, copyrights, and maintaining the timeline of transactions. The WTECteam saw two especially interesting demonstrations illustrating how electronicmoney can be securely moved around between interested parties and howcopyrights can be protected in the sale and distribution of digital objects. Inthe demonstration of moving electronic money around, a smart card is used formaking purchases from anywhere as long as one is connected to the network. Whendigital objects are marketed over the network, the sellers need to ensure thattheir copyrights are protected. NTT's InfoProtect project demonstrates thesecure distribution of images. The owner of the digital content first creates apartial image (semi-disclosed) and its descrambling key. The descrambling keyis registered with the system center and the partial image is transmitted tothe potential buyer. The buyer decides to purchase by inspecting the scrambledimage and buys the descrambling key via a secure key transmission protocolknown as InfoKey developed at NTT. The key is used to descramble the image. Thebuyer ID is embedded using digital watermarking, providing protection againstcopyright violation.
The high presence video teleconference system demonstrated at NTT iscentered around two large projection displays (each 110 inches long along thediagonal). The resolution is four times that of high definition TV and enablesinteraction with life-sized humans. The quality of display performance wasdemonstrated using 2D monocular and stereo still images. The monocular imageswere viewed at a resolution of 6 million pixels/frame and the stereo pairs eachhad about 3 million pixels/image, giving excellent quality to the stereoimages. Although this system as a whole is expensive, key components of thedisplay technology have been commercialized. Using sound localization, anenhanced multimedia presentation is possible with applications to remotemuseums and education.
The WTEC team saw a demonstration of a virtual tour of the monastery atSande Marco in Florence, Italy was shown at Keio University. The tour wasdisplayed on three flat screens using back projection. About 1,000 photographstaken at the monastery were used with a 3D modeling package to create the tour.The building and surroundings were all synthesized, whereas the artwork was allphotographed. This required 200 MB of storage. On the virtual tour it ispossible to zoom in on the many works of art. The tour is controlled using ajoystick.
The team also saw Toppan's Virtual Reality Gallery, which consists of aportion of a spherical screen in an auditorium giving a horizontal visual rangeof 150 degrees, so the viewer is enveloped by the image being displayed by adigital projection system of resolution 3,500 x 1,000 lines. Toppandemonstrated a virtual reality tour through the Sistine Chapel that was createdby taking still photographs from 50 different vantage points throughout thechapel, digitizing them and using them to create a 3D digital model. The vieweris able to move around within the chapel by means of a hand-held joystick. ATRdemonstrated an agent interface that serves as a tour guide. This genericapproach will be useful for exploring cyberspace. Other ongoing researchefforts in this laboratory are communication by mental images including kansei processing, virtual reality, art and technology, includinginteractive environments, understanding of emotions from music and peoples'voices, and human communications science. This chapter will now brieflyelaborate on the concept of kansei computing, which appears to besimilar to the concept of affective computing put forward by Picard (1997).
Kansei is defined as computing that relates to, arises from, or isinfluenced by human characteristics such as sensibility, perception, affectionor subjectivity. A national research project at ATR sponsored by the Ministryof Education during 1992-1995 gave impetus to giving computers human-likeresponsiveness. One of the media of Kansei information is the face, asit is able to express subtle emotions. There are two aspects to what isperceptible from a facial expression. One deals with image engineering issuesof accurate, robust face recognition and expression algorithms. The other dealswith human-science aspects of mimicking human information processing. Forexample, to respond to queries such as who looks similar here? or who is themost senior? one should incorporate ideas from human information processing.The emphasis of this group is on combining psychophysics and image engineering.This will guide the design of human computer interface (HCI) systems. Forfurther information refer to the HCI report.