Abstract: A single channel stereoscopic monitor receives a stereoscopic TV image and an Internet stereoscopic image. Left and right eye images are simultaneously received to display the stereoscopic image by a monitor or a TV through a single channel, are polarized by polarizing plates having left and right deflection angles symmetrical to each other, and are automatically combined into a single stereoscopic image by a reflecting mirror and a half mirror, so that a viewer can watch the stereoscopic image.

Abstract: The general technical field of the invention is that of viewing devices mounted on vehicles and comprising a display whose image is reflected by the windscreen of the vehicle, making it possible to superimpose the displayed image on the outside world. One of the major problems with this type of display is that the windscreen gives, by reflection, two non-superimposed doubled images. The basic principle of the invention is that the initial image is such that the combination of the first reflected image and the second image reflected by the windscreen gives a “clean” or “ideal” perceived image without any offsets. To perform this function, an image is generated on the display that comprises, obviously, the “ideal” image, but also a succession of offset images comprising negatives of the “ideal” image so that the resultant sum of the first reflected image and of the second reflected image includes no more than the “ideal” image.

Abstract: Systems and methods for streaming video information to a hand-held device include dynamically modifying video image properties of captured video frames to generate video data packets of a size suitable for transmission over a low bit-rate channel to a hand-held device for viewing. The systems and methods may dynamically and automatically control image properties via a hardware capture card device driver to produce a video data packet of a desired maximum data size such that subsequent compression and coding strategies produce a data packet suitable for transmission to the hand-held device for decoding and decompression to display the streaming video in near real-time.

Abstract: A motion vector estimating device estimating a motion vector having high accuracy is provided. An eliminating unit eliminates each of pictures in a set of input pictures for each picture. A picture reducing unit reduces left pictures after the elimination, and generates reduced pictures. An overall motion vector estimating unit reads the reduced pictures, stored in the frame memory, as either reduced inputted pictures or reduced reference pictures, and estimates an overall motion vector between reduced pictures. Based on the overall motion vector estimated between the reduced pictures, a correction motion vector calculating unit calculates a motion vector between two reduced pictures as a correction motion vector. The correction motion vector calculating unit calculates the motion vector in the order of the inputted pictures in a pre-eliminated set of input pictures. A coding unit estimates the motion vector for each macroblock, using a determined search range, and performs cording processing.

Abstract: A first prediction unit performs intra-prediction of an original image corresponding to an inter-frame coded frame to produce a first prediction residual and a prediction parameter. A second prediction unit performs intra-prediction of local decode data corresponding to the inter-frame coded data with the prediction parameter to produce a second prediction residual. A subtractor calculates difference data between the first prediction residual and the second prediction residual. A third prediction unit performs intra-prediction of data corresponding to a decoded frame with the prediction parameter to produce a third prediction residual, when the inter-frame coded data is decoded and transformed into the intra-frame coded data. An adder calculates addition data of the third prediction residual and the differential data. An intra-frame coding unit performs intra-frame coding to the addition data.

Abstract: The present invention relates to a method of assembling a first data stream (ds1) with a second data stream (ds2), each data stream comprising coded frames (I,P), each coded frame being associated with a relative time value, specific coded frames of a data stream being preceded by an absolute time value. Said method comprises the steps of: calculating a current reference time value for a current coded frame on the basis of a previous reference time value and on an integer part of the current relative time value, calculating an offset time value on the basis of the reference time value of the last frame in the display order of the first data stream, modifying the absolute time values of the specific coded frames of the second data stream by adding the offset time value to said absolute time values.

Abstract: In an electronic endoscope system, an RF signal is produced through quadrature modulation of a picture signal that is representative of an image taken through an electronic endoscope. When control signals are entered by an operator through a control section of the electronic endoscope, a data superimposing section superimposes the entered control signals on the RF signal in horizontal scanning intervals within a vertical blanking interval. The RF signal having the control signals superimposed thereon is sent as an electric wave of a single frequency band to a processor. In the processor, a data analyzer carries out sampling to extract the entered control signals if they are superimposed on the picture signal, and analyzes the contents of the entered control signals. Based on the results of analysis, a CPU controls respective components of the signal processor.

Abstract: Time stamped streams respectively generated by an AV stream generating means and data stream generating means are combined by a time-stamped contents generating means. Consequently, a time-stamped contents stream is generated which synchronizes the processing of an AV stream and multimedia data.