We propose a new method to recover the 3D structures and motions of rotating rigid bodies from a moving image. We assume that the image is taken under the perspective condition and each axis of rotation does not move. A rotation in 3D space becomes a circular trajectory. A circular trajectory projected into a 2D image plane becomes a conic section, for instance an ellipse. A recent work by Sawhney et al. showed that the elliptic trajectory in a 2D image plane could be used to recover the 3D space motions and structures. We emphasized the time factor to recover them. We employ the Hough transform method and also devide the transform into several stages. The division of the transform stage reduced the computation cost. Our method enables separation of two trajectories which cannot be separated by Sawhney's method.

View full abstract

This paper presents a new type of AM transmitter which can produce AM waves directly from digitized audio signals by shifting the phase of the signals digitally using a frequency twelve times as high as the carrier. The transmitter consists of six switching-type RF power amplifiers with a maximum peak output power of 330 W, a 12-bit A/D converter, a ROM for waveform compensation (arc sin etc.), phase shifters/selectors for producing carrier pulse width signals, and a power combiner. The transmitter displays high overall efficiency of 67% (RF output/DC input), an SN ratio of 68 dB, and about 0.9% total harmonic distortion at 80% modulation.

View full abstract

Several kinds of stereoscopic display systems have been developed. The back-light distribution method has generated a lot of interest recently. This method permits the viewing of stereo images by several persons simultaneously without wearing of special glasses. This paper describes the method of measuring the stereo viewing area of this system. Using the experimental results, we analyze the stereo viewing area including the properties of the lenses that control the directivity of back-light, as they have a significant effect upon the stereo viewing area.

View full abstract

This paper describes a control scheme for the multi-encoder system that transmits multiple video signals under the unified control of each encoder. First, the rate-distortion characteristics of the MC+DCT (MPEG) encoder are analyzed. We show that the rate-distortion characteristics can be represented by a simple power function. Based on this function, a bit-rate distribution rule for multiple encoders and a method of applying this to moving pictures is proposed. A computer simulation using real moving pictures shows that the proposed scheme reduces the total distortion of the system as well as probability of severe distortion which occurs in the case of a “difficult” picture, as compared with the non bit-rate distribution method.

View full abstract

For digital video compression, motion-compensated interframe prediction is one of the most effective technique and many kinds of predictor have been studied. In this paper, we propose a method to calculate efficiency of the predictors. The method consists of image models and descriptions of the predictors. The image models are band-limited and sampled functions : step function and impulse. The descriptions of the predictors specify the way to construct predictive pixels. Three descriptions are introduced in this paper : simple prediction, half-pel prediction and average type prediction. The image models are processed according to the description and prediction efficiencies are derived as reciprocal of average prediction error. Theoretical results obtained by the proposed method are also compared to the results of computer simulation with a video sequence. The simulation results show good agreement.

View full abstract

[in Japanese]

View full abstract