When distributing 3D contents real-time over a network with a narrow bandwidth such as a telephone line, methods for streaming and data compression can be considered indispensable.In previous work, we made possible the real-time streaming of 3D animation data on a network with a narrow bandwidth such as a telephone line by partitioning motion data for humanoid characters (data obtained by motion capture, for example full frame data at 30 frames/sec) into packets and then carrying out compression by culling data along the time axis.However, as a 3D scene becomes more complex, the number of humanoid characters also increases. Accordingly, the transmission rate also increases, becoming greater than the available bandwidth and making real-time distribution impossible.In this paper, we concentrate on the problem of real-time distribution, describing a new data packet format which allows flexible scalability of the transmission rate, and a data compression method, SHCM, which maximizes the features of this format using a 3D scene structure.Because compression using a 3D scene structure aims to obtain the optimal overall compression rate by altering the compression rate for each object, based on information on the position in 3D space relative to the behavior (motion) data of each object, its application to MPEG4 can be expected.Using this method the real-time distribution of 3D contents becomes possible despite the bandwidth restrictions of an ordinary telephone line.

In VRML, a modeling language for describing 3D objects on the internet, the specification to realize lifelike movement of a 3D character with a skeletal structure (such as a human) has been standardized as VRML Humanoid Animation Ver. 1 .O (H-Anim Ver. 1 .O) in the H-Anim WG of the VRML Consortium. To extend this specification, we suggest a method that makes it possible to send/receive motion data in real time on a network with narrow bandwidth such as a telephone line. Moreover, by sending the motion data with streaming data from server to client, the time required before playback can be greatly reduced. This technology uses basic techniques that can be applied widely to webbased 3D applications, broadcasting contents etc.

A proposal is made for the use of contextual information in the machine translation of Japanese and English. This paper describes the use of a Context Monitor to nlaintaill contextual infortmttion dyn,'unically and the ~tugmenlalion of appropriate features to ~t semantic network to enahh~ simple inference. The al}proach taken is that {}t" "best gucs~?' processing with the cont~extual information being hal~dled with semantic inf{~rmalion on ~ shallow level. 2 Introduction Current Machine qh'anslatiou (MT) systems proc,~ss input sentence by sentence. I[owever, experience wil.h English and Japanese has shown that some languages difl'er to such a degree that sentential translation yiehls poor results, l,eL us first compare the results of a conventional MT sysl.em with those we expect, t,o get for MT with context: t. J::q'lJf]~-._-i~¢{:¢/,?)[ b v,-)-1/~3-'.~E~gfi £" 2)i L v, K 2. K-7: ~t-laJaS'd?,l~ ~a ~t~.I;~y b t:: o 4. k-c g ];! (5'~ko This might be translated by a current, machine tl'anslation system as shown in Figure 11: It can clearly l)e seen that meaning in IHally seI/tenees is obscured. Let us compare this with I.he resuits of a system using simple cont.exl.ual informal,ion ms shown in Figure 2: This secol/d translation is i-tlllch Ill(H'(? CO]lO['{~ll{. ;IH{I better preserves the meaning of the original se,lten{'o. An attempt has therefore I}een made to solve some of tile problems of translal.ing languages SllCIt sis Japanese and English using contextual information. Due to [.he consideral.ions of wanting to produce a high quality small-sized MT system, lhe approach taken is to use tile resources awdlahle in an exisl;ing MT system and to process the contextu;d i,l['orlmd.ion l There is obviously n great difference in results Imtweet, systems, hnl, l.hese translat.icms relweSent tyl}iCal {uHe, llted) r~.stdts fi'om a numher of systems, a) and I}) options,hq}end on the default settings of individual systems