Showing papers by "Mitsubishi Electric published in 1999"

Voice puppetry

Abstract: We introduce a method for predicting a control signal from another related signal, and apply it to voice puppetry: Generating full facial animation from expressive information in an audio track. The voice puppet learns a facial control model from computer vision of real facial behavior, automatically incorporating vocal and facial dynamics such as co-articulation. Animation is produced by using audio to drive the model, which induces a probability distribution over the manifold of possible facial motions. We present a lineartime closed-form solution for the most probable trajectory over this manifold. The output is a series of facial control parameters, suitable for driving many different kinds of animation ranging from video-realistic image warps to 3D cartoon characters. CR Categories: I.3.7 [Computer Graphics]: Three-Dimensional Graphics and Realism—Animation; I.2.9 [Artificial Intelligence]: Robotics—Kinematics and Dynamics; I.4.8 [Image Processing and Computer Vision]: Scene Analysis—Time-varying images; G.3 [Mathematics of Computing]: Probability and Statistics—Time series analysis; E.4 [Data]: Coding and Information Theory—Data compaction and compression; J.5 [Computer Applications]: Arts and Humanities—Performing Arts

The VolumePro real-time ray-casting system

Abstract: This paper describes VolumePro, the world’s first single-chip realtime volume rendering system for consumer PCs. VolumePro implements ray-casting with parallel slice-by-slice processing. Our discussion of the architecture focuses mainly on the rendering pipeline and the memory organization. VolumePro has hardware for gradient estimation, classification, and per-sample Phong illumination. The system does not perform any pre-processing and makes parameter adjustments and changes to the volume data immediately visible. We describe several advanced features of VolumePro, such as gradient magnitude modulation of opacity and illumination, supersampling, cropping and cut planes. The system renders 500 million interpolated, Phong illuminated, composited samples per second. This is sufficient to render volumes with up to 16 million voxels (e.g., 256) at 30 frames per second. CR Categories: B.4.2 [Hardware]: Input/Output and Data Communications—Input/Output DevicesImage display; C.3 [Computer Systems Organization]: Special-Purpose and ApplicationBased Systems—Real-time and embedded systems; I.3.1 [Computer Graphics]: Hardware Architecture—Graphics processor;

A Winnow-Based Approach to Context-Sensitive Spelling Correction

Abstract: A large class of machine-learning problems in natural language require the characterization of linguistic context. Two characteristic properties of such problems are that their feature space is of very high dimensionality, and their target concepts depend on only a small subset of the features in the space. Under such conditions, multiplicative weight-update algorithms such as Winnow have been shown to have exceptionally good theoretical properties. In the work reported here, we present an algorithm combining variants of Winnow and weighted-majority voting, and apply it to a problem in the aforementioned class: context-sensitive spelling correction. This is the task of fixing spelling errors that happen to result in valid words, such as substituting to for too, casual for causal, and so on. We evaluate our algorithm, WinSpell, by comparing it against BaySpell, a statistics-based method representing the state of the art for this task. We find: (1) When run with a full (unpruned) set of features, WinSpell achieves accuracies significantly higher than BaySpell was able to achieve in either the pruned or unpruned conditions (2) When compared with other systems in the literature, WinSpell exhibits the highest performances (3) While several aspects of WinSpell‘s architecture contribute to its superiority over BaySpell, the primary factor is that it is able to learn a better linear separator than BaySpell learnss (4) When run on a test set drawn from a different corpus than the training set was drawn from, WinSpell is better able than BaySpell to adapt, using a strategy we will present that combines supervised learning on the training set with unsupervised learning on the (noisy) test set.

MPEG-4 rate control for multiple video objects

Abstract: This paper describes an algorithm which can achieve a constant bit rate when coding multiple video objects. The implementation is a nontrivial extension of the MPEG-4 rate control algorithm for single video objects which employs a quadratic rate quantizer model. The algorithm is organized into two stages: a pre- and a post-encoding stage. In the pre-encoding stage, an initial target estimate is made for each object. Based on the buffer fullness, the total target is adjusted and then distributed proportional to the relative size, motion, and variance of each object. Based on the new individual targets and rate-quantizer relation for texture, appropriate quantization parameters are calculated. After each object is encoded, the model parameters for each object are updated, and if necessary, frames are skipped to ensure that the buffer does not overflow. A preframeskip control is exercised to avoid buffer overflow when the motion and shape information occupies a significant portion of the bit budget. The rate control algorithm switches between two operation modes so that the coder can reduce the spatial coding accuracy for an improved temporal resolution. A shape-coding control mechanism is also proposed, which provides a tradeoff between texture and shape coding accuracy. Overall, the algorithm is able to successfully achieve the target bit rate, effectively code arbitrarily shaped objects, and maintain a stable buffer level. These techniques have been adopted by the MPEG committee in July 1997 as part of the video verification model (VM8).

Java-based mobile agents

Abstract: The mobile agent concept grows out of three earlier technologies: process migration [5], remote evaluation [7], and mobile objects [3]—all developed to improve on remote procedure calling (RPC) for distributed programming. Early systems supporting process migration allowed an entire address space to be moved from one computer to another. One goal of this mechanism was to reduce network bandwidth (compared to RPC) when multiple RPC calls are needed to execute an application. While process migration allowed an entire process to be transferred to a remote host, this mechanism did not allow an easy way to return data back to the source node without the entire process returning as well (see Figure 1). Next came remote evaluation programming, allowing one computer to send another computer a request in the form of a program (rather than an entire process address space). The remote computer receiving such a request executes the program referenced in the request within its own local address space and returns the results to the sending computer. Remote evaluation systems improved on he Internet and the World-Wide Web have become

Mining features for sequence classification

Abstract: Classi cation algorithms are di cult to apply to sequential examples, such as plan executions or text, because there is a vast number of potentially useful features for describing each example. Past work on feature selection has focused on searching the space of all subsets of the available features which is intractable for large feature sets. We adapt data mining techniques to act as a preprocessor to select features for standard classi cation algorithms such as Naive Bayes and Winnow. We apply our algorithm to the task of predicting whether or not a plan will succeed or fail, during plan execution. The features produced by our algorithm improve classi cation accuracy by 10-50% in our experiments. Submitted to IJCAI'99. This work may not be copied or reproduced in whole or in part for any commercial purpose. Permission to copy in whole or in part without payment of fee is granted for nonpro t educational and research purposes provided that all such whole or partial copies include the following: a notice that such copying is by permission of Mitsubishi Electric Information Technology Center America; an acknowledgment of the authors and individual contributions to the work; and all applicable portions of the copyright notice. Copying, reproduction, or republishing for any other purpose shall require a license with payment of fee to Mitsubishi Electric Information Technology Center America. All rights reserved. Copyright c Mitsubishi Electric Information Technology Center America, 1998 201 Broadway, Cambridge, Massachusetts 02139 Rensselaer Polytechnic Institute University of Rochester Publication History:{ 1. First printing, TR-98-22, December 1998

Using plan recognition in human-computer collaboration

Abstract: Human-computer collaboration provides a practical and useful application for plan recognition techniques. We describe a plan recognition algorithm which is tractable by virtue of exploiting properties of the collaborative setting, namely: the focus of attention, the use of partially elaborated hierarchical plans, and the possibility of asking for clarification. We demonstrate how the addition of our plan recognition algorithm to an implemented collaborative system reduces the amount of communication required from the user.

Power loss and junction temperature analysis of power semiconductor devices

Abstract: A newly developed electrothermal calculation method is implemented to estimate the power loss and working temperature of insulated gate bipolar transistor (IGBT) devices. Based on the measurement of the IGBT's characteristics, the exact estimation of power loss considering the junction temperature is introduced. Then, the thermal network is used to calculate the working temperature. The comparison between experimental and calculation results shows that this method is effective as a designing step with only the time-domain voltage and current data obtained from simulation results.

System for classifying an individual's gaze direction

Abstract: A system is provided to classify the gaze direction of an individual in which frequently occurring head poses of the individual are automatically identified and labelled according to their association with the surrounding objects. In conjunction with processing of eye pose, each observed head pose of the individual is automatically associated with a bin in a "pose-space histogram". This histogram records the number of occurrences of different head poses over an extended period of time. Given observations of a car driver, for example, the pose-space histogram develops peaks over time corresponding to the frequently viewed directions of toward the dashboard, toward the mirrors, toward the side window, and straight-ahead. Each peak is labelled using a qualitative description of the environment around the individual, such as the approximate relative directions of dashboard, mirrors, side window, and straight-ahead in the car example. The labelled histogram is then used to classify the head pose of the individual in all subsequent images.

Analysis of ozone generation from air in silent discharge

Abstract: A new theoretical model for ozone generation from air is proposed. This model considers the dependency of the ozone dissociation rate by electron impact on the electric field strength, in addition to ozone decomposition processes through the reactions with nitrogen oxides. Assuming a stationary and uniform discharge in time and space, ozone concentrations obtained experimentally under various discharge gap widths and gas pressures are in good agreement with the analytical results. It is concluded that the discharge condition of a narrow gap and accordingly an optimized high gas pressure is one of the most effective means of efficient ozone generation for an air-fed ozone generator.

Method and apparatus for detecting and correcting errors and erasures in product ECC-coded data arrays for DVD and similar storage subsystems

Abstract: A method and apparatus for detecting and correcting errors and erasures in product-coded data arrays by iterative syndrome processing array data in row major order and column major order A first dense map is formed for classifying each row containing location indicia of random errors, their correction patterns, and pointers to rows containing erasure errors This map is used to effectuate row array random error corrections in place in memory A second dense map is formed of location indicia and correction patterns for each pair adjacent position within a column containing erasure errors as indexed by a counterpart row pointer The second map is used to effectuate column array erasure corrections and random error corrections in place in memory

Method for formatting and routing data between an external network and an internal network

Abstract: A method for formatting and routing data between an external network and an internal IEEE 1394 network is provided. The method comprises receiving a data packet at a gateway device, separating data information from the data packet, reformatting the separated data information from a first digital format into a second digital format, selecting a transmission mode for communicating the data information in the second digital format to a particular node residing on the internal network, preparing a portion of the data information in the second digital format for transmission in the selected transmission mode, and transmitting the portion of the data information in the second digital format to the particular node via the selected transmission mode.

Plasma processing method and plasma processing apparatus

Abstract: A plasma processing apparatus mainly comprises a processing chamber ( 10 ) formed by a vacuum vessel, a magnetic field forming coil ( 80 ) arranged around the processing chamber for forming a rotating magnetic field and gas supply means ( 101 ) supplying various gases to the processing chamber ( 10 ). The processing chamber ( 10 ) is divided into a reaction chamber ( 44 ) forming plasma with a partition wall ( 43 ) and a buffer chamber ( 45 ) discharging externally supplied gases with pressure difference. The reaction chamber ( 44 ) includes a high-frequency electrode arranged oppositely to the buffer chamber ( 45 ). The gas supply means ( 101 ) includes pulse gas valves ( 63 a and 63 b ) for pulsatively supplying gases to the processing chamber ( 10 ). Thus provided are a plasma processing method and a plasma processing apparatus capable of uniformly processing a wafer having a large diameter and reducing RIE lag with respect to a fine etching pattern.

Investigation of Acidic Texturization for Multicrystalline Silicon Solar Cells

Abstract: A new texturing method especially for multicrystalline silicon solar cells was investigated. This technique is based on a maskless acidic etching of silicon. It can texturize surfaces quite homogeneously in a short time and eliminate a saw damage removal process. The authors showed phosphoric acid (H{sub 3}PO{sub 4}) is preferable as a catalytic agent to moderate etching rates without effecting texturing. A simulation was carried out in an attempt to achieve lower reflectance. It was found that the reflectance is dependent on the quotient of the depth and the width of texture (h/D). The authors found that surface active agents are effective in increasing the h/D value. The reflectance decreases corresponding to h/D, and measured reflectance is in good agreement with the simulation. Comparing the electrical characteristics of an acidic textured multicrystalline solar cell and an alkaline textured one, the short-circuit current density of acidic texturing is higher than the other.

Communication apparatus and communication method

Abstract: An object of the present invention is to make it possible to utilize information service provided by a carrier without designating a standard protocol as a communication network A plurality of carriers provide information services to a communication network to which a called terminal unit (10) is connected through a communication line (20) At the time of termination, a carrier control signal (6) transmitted from the communication network includes a calling line number (6a) for identifying an originator, carrier information (6b) for identifying a carrier, and carrier service information (6c) for identifying a service to be provided When a plurality of terminals (15, 16) are connected to the called terminal unit (10) as slave units, a function of inter-calling a terminal-number provided by a carrier can be utilized Carrier information identifying means (21) identifies a carrier on the basis of the carrier information (6b), and carrier-service-information-identifying means (22) controls a terminal selecting circuit (14) on the basis of the carrier service information (6c)

High-frequency module

Abstract: Heretofore, a plurality of packages were used in a high frequency module in which a plurality of waveguide terminals were positioned resulting in problems, such as degradation of characteristics in the connection lines between packages, lower ease of assembly when mounting and connecting the connection lines, increased cost, and so forth. To solve these problems, a plurality of cavities having a part or the entire side metallized is formed in a multi-layer dielectric substrate. The multi-layer dielectric substrate is provided with a plurality of waveguide terminals, microstrip line-waveguide converters, RF lines, bias and control signal wiring, and bias and control signal pads. A high frequency circuit is mounted within the cavity and sealed with a seal and cover. This is intended to reduce the number of package, improve performance, improve fabrication, and lower the cost.

208-W TEM00 operation of a diode-pumped Nd:YAG rod laser.

Abstract: 208-W average-power TEM00-mode operation from a diode-pumped Nd:YAG rod laser was demonstrated. The side-pumping method of generating a high-gain aberration-free rod, the bifocusing-compensation technique with two identically pumped rods, and a suitable choice of beam spot size were employed in the design of this laser. At the maximum pump power of 1.1??kW the fundamental transverse-mode operation M2<1.1 was characterized by 7.6% electrical efficiency. The extraction efficiency was almost 60% of full multimode operation. Stable operation was obtained within a pump-power range of 15% of the maximum pump power.

Temporal and spatial coherent ray tracing for rendering scenes with sampled and geometry data

Abstract: A method traces rays through graphical data. The method partitions the graphical data into a plurality of blocks according to a scheduling grid. For each block, a ray queue is generated. Each entry in the ray queue representing a ray to be traced through the block. The ray queues are ordered spatially and temporally using a dependency graph. The rays are traced through the blocks according to the ordered list.

Image display and light-emitting device

Abstract: A liquid crystal projector having a simple optical system and operating at a high efficiency of use of light, comprises a light-emitting device (50), a parallel transforming optical system (60), an optical switch (70), and a display optical system (80). The light-emitting device (50) includes LD arrays (51, 52, 53) where semiconductor lasers (54) are arranged two-dimensionally. The output beams from the semiconductor lasers (54) are transformed into parallel beams by means of the lens arrays (61, 62, 63). Since the output beams from the semiconductor lasers (54) are linearly P- or S-polarized, no polarization transforming optical system is needed. By adequately combining semiconductor lasers (54), the shapes of the LD arrays (51, 52, 53) can be the same as those for light utilization of a red-light transmission liquid crystal panel (71), a green-light transmission liquid crystal panel (72), and a blue-light transmission liquid crystal panel (73).

A precise on-chip voltage generator for a gigascale DRAM with a negative word-line scheme

Abstract: A precise on-chip voltage generator for gigascale DRAM's with a negative word-line scheme is described. It combines a charge-pump regulator and a series-pass regulator, and it also includes a positive and negative offset voltage generator that uses a bandgap generator with a differential amplifier. The proposed circuit was experimentally evaluated with a test device fabricated using a 0.3-/spl mu/m process. The simulation results show that the series-pass regulator suppresses the noise on a word-line low voltage (negative) to below 30 mV for the word-line transient and V/sub BB/ bouncing. A dc-voltage error of less than 6% without trimming is confirmed for the positive and negative offset voltage generator through the test device. These results show that the described scheme can be used in future low-voltage gigascale DRAM's.

Charge pump circuit

Abstract: A charge pump circuit comprises a plurality of booster stages Each booster stage has a first node, a second node, a first charge pump and a second charge pump Both charge pumps operate in a complementary manner, and raise potential of the second node relative to potential of the first node by transferring charge from the first node to the second node Each charge pump comprises a pumping capacitor, an NMOS transistor and a PMOS transistor In each charge pump, the NMOS transistor is used for charging the pumping capacitor with charge input through the first node, and the PMOS transistor is used for discharging the pumping capacitor to send charge to the second node

Substrate-bias effect and source-drain breakdown characteristics in body-tied short-channel SOI MOSFET's

Abstract: The substrate-bias effect and source-drain breakdown characteristics in body-tied short-channel silicon-on-insulator metal oxide semiconductor field effect transistors (SOI MOSFET's) were investigated. Here, "substrate bias" is the body bias in the SOI MOSFET itself. It was found that the transistor body becomes fully depleted and the transistor is released from the substrate-bias effect, when the body is reverse-biased. Moreover, it was found that the source-drain breakdown voltage for reverse-bias is as high as that for zero-bias. This phenomenon was analyzed using a three-dimensional (3-D) device simulation considering the body-tied SOI MOSFET structure in which the body potential is fixed from the side of the transistor. This analysis revealed that holes which are generated in the transistor are effectively pulled out to the body electrode, and the body potential for reverse-bias remains lower than that for zero-bias, and therefore, the source-drain breakdown characteristics does not deteriorate for reverse-bias. Further, the influence of this effect upon circuit operation was investigated. The body-tied configuration of SOI devices is very effective in exploiting merits of SOI and in suppressing the floating body-effect, and is revealed to be one of the most promising candidates for random logic circuits such as gate arrays and application specific integrated circuits.

Adsorption equilibria of krypton, xenon, nitrogen and their mixtures on molecular sieve 5A and activated charcoal

Abstract: The adsorption equilibria of Kr, Xe and N2, which are constituents of the off-gas from nuclear reprocessing processes, on representative adsorbents (Molecular Sieve 5A (MS5A) and activated charcoal) were studied. Adsorption experiments were conducted in the temperature range of 77 to 323 K using a packed bed column. The adsorption isotherms for the activated charcoal adsorbent were successfully correlated by the vacancy-solution model. The adsorption isotherms for the MS5A adsorbent were properly correlated by the Langmuir model and the vacancy solution model. The adsorption experiments for the binary component systems (Kr—Xe, Kr—N2 systems) were also performed, and the results suggest that the coexistence of Xe greatly inhibits the adsorption of Kr. The coexistence of large amounts of N2 was also found to inhibit the adsorption of Kr. The experimental results for the adsorption equilibrium of binary component systems on the activated charcoal adsorbent were well reproduced by the vacancy solution model w...

Effects of very fast transient overvoltages on transformer

Abstract: An analytical solution is found using the Laplace transform to describe the very fast transient overvoltage (VFTO) in the gas insulated switchgear (GIS) connected directly to a transformer. The oscillation frequencies, and other properties, are explicitly given in terms of system parameters. The magnitude of VFTO at the transformer is suppressed due to the transformer capacitance, Magnitudes of the initial abrupt voltage change and oscillatory voltage are assessed. The abrupt voltage is larger, but not enough to induce a harmful voltage in the transformer. The oscillatory components may damage the winding insulation at the occurrence of resonance.

Part pose statistics: estimators and experiments

Abstract: Many of the most fundamental examples in probability involve the pose statistics of coins and dice as they are dropped on a flat surface. For these parts, the probability assigned to each stable face is justified based on part symmetry, although most gamblers are familiar with the possibility of loaded dice. In industrial part feeding, parts also arrive in random orientations. We consider the following problem: given part geometry and parameters such as center of mass, estimate the probability of encountering each stable pose of the part. We describe three estimators for solving this problem for polyhedral parts with known center of mass. The first estimator uses a quasistatic motion model that is computed in time O(n log n) for a part with n vertices. The second estimator has the same time complexity but takes into account a measure of dynamic stability based on perturbation. The third estimator uses repeated Monte Carlo experiments with a mechanics simulation package. To evaluate these estimators, we used a robot and computer vision system to record the pose statistics based on 3595 physical drop experiments with four different parts. We compare this data to the results from each estimator. We believe this is the first paper to systematically compare alternative estimators and to correlate their performance with statistically significant experiments on industrial parts.

Transmitting method and transmitting device

Abstract: Out of devices connected to the same power line (7), a device which has judged based on the results of noise measurement that it is difficult to continue communication in the case of the current tone-set/tone position comes to serve as a virtual master for tone change. The device serving as a virtual master transmits a tone change request command and waits for a response from another device for a predetermined time. If the virtual master device receives even one tone change rejection command during the time, the virtual master device abandons the change and ends the virtual master processing. Conversely, if the virtual master device receives no tone change rejection command in the predetermined time, the virtual master device transmits a tone change instruction command. On receiving the tone change instruction command, all the other devices currently connected change the tone at a time.

Effect of channel direction for high performance SCE immune pMOSFET with less than 0.15 /spl mu/m gate length

Abstract: A high performance CMOSFET with a channel along the crystallographic axis has been developed. Current drivability of the pMOSFET is improved by about 15% by changing the channel direction from to due to an increase in hole mobility and high immunity against short channel effects (SCE). As a result, a drive current of 810 /spl mu/A//spl mu/m for nMOS and of 420 /spl mu/A//spl mu/m for pMOS with 0.14 /spl mu/m gate length has been achieved under 1 nA//spl mu/m off current at 1.8 V operation.

Method for estimating volumetric distance maps from 2D depth images

Abstract: A volumetric distance map of an object is generated from one or more depth images of the object. Each depth image is projected onto the object by casting parallel rays to the object. The parallel rays are cast perpendicular to the depth image. Sample points in a projected distance volume represent distances from the distance map to a surface of the object. The magnitude of a local gradient at each sample point of the projected distance volume is determined, and each distance at each sample point is divided by the magnitude of the corresponding local gradient at each sample point to obtain a scalar distance to a closest surface of the object.

An electroabsorption modulator module for digital and analog applications

Abstract: This paper presents an electroabsorption modulator (EAM) module for digital and analog (D/A) applications. Optically broad-band operation of the EAM module is studied for such digital application as wavelength division multiplexing (WDM) systems. Utilizing anisotropic electroabsorption of a multiple quantum-well (MQW) EAM, 40- and 100-nm bandwidth operations in 2.5-Gb/s digital signal transmission over 200-km standard fiber are confirmed by the experiments and the simulations, respectively. For analog applications, low distortion and high link gain characteristics of the EAM module are investigated at the wavelength of 1535 nm. High spurious-free dynamic range of 123 dB/spl middot/Hz/sup 4/5/ and high link gains of -10.3 dB with matching circuit and -20.6 dB without matching circuit are obtained using the EAM module.

Building virtual structures with physical blocks

Abstract: We describe a tangible interface for building virtual structures using physical building blocks. We demonstrate two applications of our system. In one version, the blocks are used to construct geometric models of objects and structures for a popular game, Quake II™. In another version, buildings created with our blocks are rendered in different styles, using intelligent decoration of the building model.