デジタルコンテンツのネットワーク配信事業への取り組みが脚光をあびている.ここではその中でも音楽配信におけるコピー管理, ビジネスモデル, コンテンツオーサリングの課題を取り上げ, 特に商用運用を行う場合の問題点と将来の可能性について考察した.

抄録全体を表示

インタネット接続を中心としたケーブルテレビの通信利用が急増している.一方, ケーブルテレビのディジタル化が急務になっているが, 互換性確保による設備投資額低減を目指し, 日本版ケーブルラボという民間の認証機関の設立が進行中である.また, ケーブルテレビにおける高い周波数帯域での無線利用が進められている.

抄録全体を表示

32ビットCPUで制御される, ぬいぐるみ状の対話ロボットを設計し, 音声認識, マルチモーダル対話, ディジタル通信などのソフトウェアを統合的に実装することにより, 高齢者に対してペットのように話し相手になり, かつ, 生活情報を提供する情報端末システムを開発した.本システムは, コミュニケーションという精神的な側面から高齢者を支援する福祉器具である.

抄録全体を表示

Various types of interface devices operated with the hands have been developed and are widely used. However, we use our legs for walking in the real world, so it is natural to use an interface devices operated with the legs for moving around a virtual environment. In this paper, we have developed a new interface device “Waraji (WAlking, Running, And Jumping Interface)”. Using Waraji, users can work at ease in a virtual environment because their hands are free from movement operations. In addition, because Waraji is smaller than other interfaces for movement, it can he easily worn. Experiments have shown that Waraji provides operationality similar to that of a joystick. It will be useful for surround-display virtual-environment systems such as CAVE (CAVE Automatic Virtual Environment).

抄録全体を表示

To provide visually-impaired people access to graphical information, we previously developed a device called a “tactile mouse.” It consists of a computer mouse and a small tactile display. Tactile information is fed back to the fingers according to the mouse's movements, giving the user the feeling of exploring tangible drawings. Because the tactile display is small, however, users tend to deviate from the tangible graphics when they move the mouse too much. To identify conditions for reducing this deviation, we conducted two experiments : one on line following and one on shape discrimination. We found that when the parameter called “hand-movement-to-pin ratio” was too low, the chances of deviation increased, and when it was too high, discrimination failures occurred more frequently. Consequently, intermediate conditions that led to less deviation and fewer discrimination failures were found to be adequate for the efficient use of the tactile mouse.

抄録全体を表示

A sign-language recognition system should use information from both global features, such as hand movement and location, and local features, such as hand shape and orientation. We designed a system that first selects possible words by using the detected global features, then narrows the choices down to one by using the detected local features.
In this paper, we describe an adequate local feature recognizer for a sign-language recognition system. Our basic approach is to represent the hand images extracted from sign-language images as symbols corresponding to clusters by using a clustering technique. The clusters are created from a training set of extracted hand images so that images with a similar appearance can be classified into the same cluster in an eigenspace. Experimental results showed that our system can recognize a signed word even in two-handed and hand-to-hand contact cases.

抄録全体を表示

Available methods for synthesizing facial expressions synthesize basic expressions. Few methods synthesize subtle facial expressions such as a knavish smile or a cold smile. We developed a system for synthesizing subtle facial expressions. It uses an interactive interface based on a genetic algorithm. We use the five main parts of the face : eyebrows, eyes, mouth, and contours. These parts are deformed independently. Using this system, subtle facial expressions can be synthesized by iterating a deformation process for each part. Experiments showed that this system can efficiently synthesize subtle facial expressions.

抄録全体を表示

Cerebral disease is becoming a more serious problem as society ages. Methods are thus needed to evaluate the extent of cerebral disease so that rehabilitation can be provided. In our study, we are trying to detect early sign of cerebral disease by making use of the recognition mechanism of the face and synthesizing facial images using computer technology. We evaluated the extent of cerebral disease by performing two kinds of experiments : parts synthesis and expression synthesis. By detecting cerebral disease early, we start rehabilitation early and reduce the load on medical doctors. Based on the results, we classified prosopagnosia patients as having cerebral disease or being normal. We then identified those in the second group that had a risk of slight cerebral disease by conducting a parts-movement experiment.

抄録全体を表示

手腕に集中するPC操作を分散させるために, 両足で板を傾斜させて操作するポインティングデバイスを試作し, マウスと比較して評価した.ポインティングに要する時間はキーボードとマウス間で手が移動する時間を考慮すればマウスと同等以上, なぞりについては斜めの要素でマウスより精度が劣ることが明らかになった.

抄録全体を表示

本報告では, 人間の自然な手招き動作における身振りと言葉の時間的な関連性に注目し, 手招き動作を撮影した連続動画像のオプティカルフローから算出される手の移動速度と言葉による指示コマンドの発声タイミングを特徴量としてとらえ, 音声をトリガとして用いることにより, 指示者の意図する指示方向を自動認識する手法を提案する.

抄録全体を表示

We developed an adaptive dictionary-selection method for recognizing low-quality characters in scene images. First, the environmental conditions of the input image are estimated using an initial dictionary. Then, a relevant dictionary from several dictionaries reflecting different environmental conditions is automatically selected based on the estimation and used for recognition. Experiments on characters in images of bookshelves showed that this method achieves a higher recognition rate (89.8%) than that using one fixed dictionary (76.4%). Furthermore, recognition accuracy improved from 89.8% to 95.2% when contextual postprocessing was used.

抄録全体を表示

We investigated the effect of using a head-mounted display (HMD) to play game. Thirty young subjects were asked to fill in a questionnaire with 28 items concerning eye strain, motion sickness, and display-work related symptoms before and after 20 minutes of playing TV games. Each subject performed two sessions, in one session, a conventional TV monitor was used, and in the other, a commercially available HMD was used. Using factor analysis, we classified these items into an eye-strain related group and a motion-sickness related group. Almost all symptoms were worsen after game playing. Fourteen items in the TV-monitor sessions and 22 items in the HMD sessions were statistically significant. Statistical tests showed that seven symptoms related to motion sickness were significantly worse only with the HMD.

抄録全体を表示

We tested the image quality of face color for several color reproduction ranges. Image quality was subjectively measured according to a five-grade scale. We prepared test images by using a software program to add white into the image data to decrease the color saturation of the test images. Six different color reproduction ranges, 69%, 50%, 39%, 29%, 19%, and 12% the NTSC color reproduction range, were simulated by changing the amount of white added. Nine portraits were used as the test images. Test participants were asked to judge the quality of the face color and also to judge the amount of perceived unnatural whiteness in the face color.
The results show that the quality of facial coloration decreases and that the perceived unnatural whiteness increases as the color reproduction range is reduced. We found that 50% or more color reproduction was necessary to achieve “fair” to “good” quality color and that no unnatural whiteness was perceived when the color quality was better than intermediate between “fair” and “good”. We also found that the saturation of face color was about 80% that of the original image color.

抄録全体を表示

画像の印象に基づいて数量化4類で画像と印象語とを画面上にブラウジングする感性検索の方法を提案する.同じ方法で複数の製品群を同じ画面上に配置することによって, 印象が似ている製品のコーディネートを視覚的に支援する手法も提示する.また, 印象の中でも好き嫌いの嗜好度だけに着目して, 嗜好に基づいて個人も同時に配置することにより, 各個人の嗜好の違いを把握することができ, 協調フィルタリングを視覚的に行うこともできる.

抄録全体を表示

地上ディジタル放送では, マルチパス妨害に加えアナログ放送波からの妨害に対して耐性の高い受信機が望まれる.本研究では, これらの妨害に対して受信性能を向上する地上ディジタル放送用OFDM受信機を開発し, その性能を評価した.また, オンヱア受信した変調信号を記録再生できる機能を設け, 受信OFDM信号を容易に解析することができるようにした.

抄録全体を表示