scispace - formally typeset
Search or ask a question

Showing papers on "Residual frame published in 1990"


Patent
11 Apr 1990
TL;DR: In this article, a correlation surface is generated for each block in a first field or frame of a video signal, representing the difference between the content of the block in the first frame and the content in the following frame with which it has been compared.
Abstract: In, for example, a motion compensated video standards converter (Figure 1) blocks in a first field or frame of a video signal are each compared (3) with a plurality of blocks in the following field or frame of the video signal for deriving motion vectors representing the motion of the content of respective blocks between the first field or frame and the following field or frame, a correlation surface is generated (3) for respective blocks in the first field or frame, the correlation surface representing the difference between the content of the block in the first field or frame and the content of each block in the following field or frame with which it has been compared, the correlation surface is tested (21, 22, 23) for a clear minimum, the size of the blocks is increased (20) to generate new correlation surfaces, each new correlation surface is tested (21, 22, 23) for a clear minimum, and motion vectors are derived (27) in dependence on the clearest minimum so found

267 citations


Patent
22 Jun 1990
TL;DR: In this paper, a continuous speech prefiltering system for use in continuous speech recognition computer systems is presented, where the speech to be recognized is converted from utterances to frame data sets, which frame data set are smoothed to generate a smooth frame model over a predetermined number of frames.
Abstract: A continuous speech prefiltering system for use in continuous speech recognition computer systems. The speech to be recognized is converted from utterances to frame data sets, which frame data sets are smoothed to generate a smooth frame model over a predetermined number of frames. A resident vocabulary is stored within the computer as clusters of word models which are acoustically similar over a succession of frame periods. A cluster score is generated by the system, which score includes the likelihood of the smooth frames evaluated using a probability model for the cluster against which the smooth frame model is being compared. Cluster sets having cluster scores below a predetermined acoustic threshold are removed from further consideration. The remaining cluster sets are unpacked for determination of a word score for each unpacked word. These word scores are used to identify those words which are above a second predetermined threshold to define a word list which is sent to a recognizer for a more lengthy word match. A controller enables the system to initialize times corresponding to the frame start time for each frame data set, defining a sliding window.

213 citations


Patent
Hsueh-Ming Hang1
27 Sep 1990
TL;DR: In this paper, a buffer/quantizer controller in a video coder utilizes the average quantization step size employed for the previous frame, the average number of bits produced per frame, and current frame buffer fullness level to obtain an indication of image complexity.
Abstract: A buffer/quantizer controller in a video coder utilizes the average quantization step size employed for the previous frame, the average number of bits produced per pel by the video coder for the image contained in the previous frame and current frame buffer fullness level to obtain an indication of image complexity. A target current frame quantization step size to be employed by the video coder in encoding a current frame is based on the indication of image complexity. Optionally, the target frame quantization step size is adjusted by a mapping procedure, at a predetermined intervals within each frame period, to account for the current buffer fullness. The map employed by the mapping procedure is adjusted in accordance with the target frame quantization step size.

169 citations


Patent
10 May 1990
TL;DR: In this article, a frame interpolating circuit for obtaining an interpolated frame between the.. encoded.!..Iadd.frames, and a circuit for getting an error formed by frame interpolation is presented.
Abstract: A moving image signal encoding apparatus includes: a frame decimating circuit for extracting . .encoded.!. frames from an input moving image signal at specified intervals; a frame interpolating circuit for obtaining an interpolated frame between the . .encoded.!. .Iadd.extracted .Iaddend.frames, and a circuit for obtaining an error formed by frame interpolation. A moving image signal decoding apparatus includes: a receiving circuit for extracting a frame code from an inputted signal; a frame decoding circuit for decoding the frame code to obtain a reproduced frame, and a frame interpolating circuit for obtaining an interpolated frame between the reproduced frames. By transmitting an error of the interpolated frame from the encoding apparatus to the decoding apparatus and correcting the error of the interpolated frame with the decoding apparatus, the error of the interpolated frame is eliminated. Alternatively, depending on the value of the error of the interpolated frame obtained with the encoding apparatus, a circuit determines the operation mode as to whether the frame interpolating circuit of the decoder carries out frame interpolation or preceding value holding and sends a flag to show the operation mode to the decoder, so that improvement occurs when the error of the interpolated frame is large.

146 citations


Patent
11 Oct 1990
TL;DR: In this paper, a motion vector from a frame which is frames away from a reference frame is converted to motion vector per one frame between the reference frame and the frame, is optimized, and is then transmitted.
Abstract: Video signal transmitting system in which digital video signal are divided into groups of predetermined frames. Digital video signals of at least one frame of them are intraframe coded and then transmitted. Remaining digital video signals are interframe coded with reference to the digital video signals, intraframe coded, and the digital video signals, intraframe coded, of a subsequent group of frames and are then transmitted. A motion vector from a predetermined reference frame is detected. Video signals are interframe coded by the motion vector to transmit the video signals. A motion vector from a frame which is frames away from the reference frame is converted to motion vector per one frame between the reference frame and the frame, is optimized, and is then transmitted.

134 citations


Patent
18 Jan 1990
TL;DR: In an inter-frame predictive encoding system, reference frames are set at equal intervals as discussed by the authors, and a prediction signal for a dependent frame between the reference frames is generated on the basis of a signal of the reference frame which precedes and follows the dependent frame respectively.
Abstract: In an inter-frame predictive encoding system, reference frames are set The reference frames are separated at equal intervals The reference frames are selected from successively inputted frames of a video signal Each of the reference frames is encoded A prediction signal for a dependent frame between the reference frames is generated on the basis of a signal of the reference frames which precedes and follows the dependent frame respectively A signal of the dependent frame is generated on the basis of the prediction signal corresponding thereto An error of the predicting of the signal of the dependent frame is encoded IaddA complementary decoding system reconstructs the video signal from the encoded reference frames and the encoded error Iaddend

115 citations


Patent
29 May 1990
TL;DR: In this paper, the number of information packets (B) in a frame is determined as follows: if P in the formula is an integer, where BR is the bit rate of the second digital signal, ns is the total number of samples of the wideband digital signal.
Abstract: The transmitter (1) in a digital transmission system derives from a wideband digital signal SBB having a sample frequency Fs a second digital signal appearing on its output (7), which signal comprises successive frames, each frame being made up of information packets (IP) each having a length of N bits. The number of information packets (B) in a frame is determined as follows: If P in the formula is an integer, where BR is the bit rate of the second digital signal, ns is the number of samples of the wide-band digital signal, SBB in which the information, after having been converted into the second digital signal, is present in one frame of the second digital signal, the number of information packets B in a frame is equal to P. If P is not an integer, B is equal to P′ for a number of frames, P′ being the next lower integer following P, whereas B for the other frame is equal to P′+1, in such a way that the average frame rate of the second digital signal is substantially equal to Fs/ns. A frame comprises a first frame portion (FD1), a second frame portion (FD2), and a third frame portion (FD3). The first frame portion contains synchronising information and system information. The second frame portion contains allocation information, and the third frame portion contains samples of and, if applicable, scale-factor information for the second digital signal.

70 citations


Patent
25 Jul 1990
TL;DR: In this paper, the similarity between the frames is used to decide whether to encode one of them with low or high information content, i.e., as a low resolution frame or a high resolution frame.
Abstract: Continuously monitoring successive frames of a video signal and deciding, based on the similarity between the frames, whether to encode one of them with low or high information content, i.e., as a "low-resolution" frame or a "high-resolution" frame, and whether to encode the next transmitted frame as a low- resolution frame or as a high-resolution frame, and, when sending a high-resolution frame, determining whether to transmit an indication to re-display the past frame or to transmit an indication to display the high-resolution frame as a representation of the present frame. The compression method provides efficient compression of a video signal. A single future frame embodiment allows up to a doubling of the number of pixels transmitted through a channel. With multiple frame systems, even higher compression ratios are possible. Further improvements in compression ratios may be obtained by compressing the individual frames.

60 citations


Patent
20 Nov 1990
TL;DR: In this paper, a method and apparatus for detecting an object of interest against a cluttered background scene is presented, in which a sensor is movable on a platform such that each frame of the video representation of the scene is aligned, i.e., appears at the same place in sensor coordinates.
Abstract: A method and apparatus for detecting an object of interest against a cluttered background scene. In a first preferred embodiment the sensor tracking the scene is movable on a platform such that each frame of the video representation of the scene is aligned, i.e., appears at the same place in sensor coordinates. A current video frame of the scene is stored in a first frame storage device (14) and a previous video frame of the scene is stored in a second frame storage device (20). The frames are then subtracted by means of an invertor (24)and a frame adder (28) to remove most of the background clutter. The subtracted image is put through a first leakage reducing filter, preferably a minimum difference processor filter (32). The current video frame in the first frame storage device (14) is put through a second leakage-reducing filter, preferably minimum difference processor filter (36). The outputs of the two processors are applied to a two quadrant multiplier (42) to minimize the remaining background clutter leakage and to isolate the moving object of interest. In a second preferred embodiment the sensor does not track the scene and therefore the frames of video from the two storage devices (140, 200) must be aligned before the subtraction process can occur. The alignment function is performed by a background registration correlator (150) and a frame shifter (130). Once the two frames of video data are aligned they are then subtracted and preferably applied to the minimum difference processor filters (320, 360) as with the first embodiment. The outputs of the two minimum difference processors (320, 360) are again applied to a two quadrant multiplier (420) to eliminate the remaining background clutter.

55 citations


Patent
03 May 1990
TL;DR: In this article, a moving image signal decoding apparatus comprises a receiving circuit for extracting a frame code from an inputted signal, a frame decoding circuit for decoding the frame code to obtain a reproduced frame, and a frame interpolating circuit for obtaining an interpolated frame between the reproduced frames.
Abstract: A moving image signal encoding apparatus comprises a frame decimating circuit for extracting encoded frames from an input moving image signal at specified intervals, a frame interpolating circuit for obtaining an interpolated frame between the encoded frames, and a circuit for obtaining an error formed by frame interpolation. A moving image signal decoding apparatus comprises a receiving circuit for extracting a frame code from an inputted signal, a frame decoding circuit for decoding the frame code to obtain a reproduced frame, and a frame interpolating circuit for obtaining an interpolated frame between the reproduced frames. By transmitting an error of the interpolated frame from the encoding apparatus to the decoding apparatus and correcting the error of the interpolated frame with the decoding apparatus, the error of the interpolated frame is eliminated. Alternatively, depending on the value of the error of the interpolated frame obtained with the encoding apparatus, circuit determines the operation mode as to whether the frame interpolating circuit of the decoder carries out frame interpolation or preceding value holding and sends a flag to show the operation mode to the decoder, so that improvement is given when the error of the interpolated frame is large.

51 citations


Patent
Tanoi Toshiyuki1
11 May 1990
TL;DR: In this article, the second motion vector is derived at the transmit end from successive frames using a larger block size than that used in the first motion vector, and at the receive end, original frames are recovered from the predicted error signal as well as from the first and second motion vectors, and evaluated whether it is valid or not.
Abstract: At the transmit end of a video communications system, a first motion vector is derived from successive frames during a frame transmit mode and a second motion vector is derived during or immediately following a frame discard mode. An interframe predicted error signal is generated which is representative of the difference between each input frame and a motion-compensated, previous frame during the frame transmit mode, the difference being zero during the frame discard mode. The predicted error signal and the vectors are transmitted to the receive end of the system. In a first embodiment, the second motion vector is derived at the transmit end from frames spaced apart by a discarded frame, and at the receive end, original frames are recovered from the predicted error signal as well as from the first and second motion vectors, and the second motion vector is down-scaled and evaluated whether it is valid or not. During frame discard mode, motion compensation is performed on the recovered frame using the down-scaled vector in response to a valid evaluation, but no compensation is performed if invalid evaluation is made. In a second embodiment, the second motion vector is derived at the transmit end from successive frames using a larger block size than that used in the first motion vector. At the receive end, the first motion vector as well as the error signal are used in recovering original frames. During the frame discard mode, the second vector is simply used for motion compensation.

Patent
13 Aug 1990
TL;DR: Intraframe coding and interframe coding are selectively employed for generating a compressed digital video frame in order to record a motion video signal on a small capacity recording medium such as a CD ROM as mentioned in this paper.
Abstract: Intraframe coding and interframe coding are selectively employed for generating a compressed digital video frame in order to record a motion video signal on a small capacity recording medium such as a CD ROM. Motion compensation is also employed for effective compression, and motion compensation in only one direction is applied to an area of the frame in which interframe coding is applied that corresponds to an area of a previous frame in which the intraframe coding was applied.

Patent
19 Jan 1990
TL;DR: In this article, reference frames are separated at equal intervals, and each of the reference frames is encoded in an inter-frame predictive encoding system, where a prediction signal for a dependent frame between reference frames was generated on the basis of a signal of reference frames which precedes and follows the dependent frame respectively.
Abstract: In an inter-frame predictive encoding system, reference frames are set. The reference frames are separated at equal intervals. The reference frames are selected from successively inputted frames of a video signal. Each of the reference frames is encoded. A prediction signal for a dependent frame between the reference frames is generated on the basis of a signal of the reference frames which precedes and follows the dependent frame respectively. A signal of the dependent frame is generated on the basis of the prediction signal corresponding thereto. An error of the predicting of the signal of the dependent frame is encoded.

Patent
Kurenai Murakami1, Tutomu Murase1
14 Sep 1990
TL;DR: In this paper, a frame aligner for frame aligning an input time-division multiplexed (TDM) signal to an output frame synchronous signal is proposed, where an input frame signal of the TDM signal is separated into a transport overhead carrying a message pointer and a subframe carrying data signal.
Abstract: In a frame aligner for frame aligning an input time-division multiplexed (TDM) signal to an output frame synchronous signal, an input frame signal of the TDM signal is separated into a transport overhead carrying an input frame synchronous signal and a message pointer and a subframe carrying data signal. A fresh overhead having a fresh pointer is made corresponding to a phase difference between said input and said output frame synchronous signals and said subframe is sequentially written into and read from a buffer memory. The fresh overhead and the subframe read are multiplexed to form an output TDM frame signal which is synchronized with the output frame synchronous signal. The buffer memory is permitted to have a reduced memory capacity storable a number of channel signals equal to that of time slots carrying the overhead. When the input frame signal is asynchronous with an output clock signal, the input frame signal is converted to a converted frame signal which is synchronized with the output clock signal before the frame alignment is performed.

Patent
25 May 1990
TL;DR: In this paper, the contour data stored in two frame buffers are compared with each other so that a transition between corresponding portions of the two buffer buffers are detected, and additional pixel transitions are interpolated at pixels having adjacent transitions.
Abstract: An apparatus for to extracting a video signal corresponding to a moving object in a video signal representing moving and stationary object comprises means for detecting a contour in the video signal and storing the contour data of successive frame in two frame buffers. Contour data stored in the two frame buffers are compared with each other so that a transition between corresponding portions of the two frame buffers are detected. This transition image is also stored in another memory and additional pixel transitions are interpolated at pixels having adjacent transitions. Thereafter, an area encompassed by the pixels representing the transition is combined with the video signal so that a motion video signal can be extracted.

Patent
23 Feb 1990
TL;DR: In this article, a 60 frames per second progressively scanned video signal is transmitted at a 30 frames-per-second rate accompanied by decoding information, where every other frame is deleted and a respective nearest neighbor is selected to represent the deleted frame.
Abstract: A method and apparatus for reducing the amount of video information which is transmitted to a receiver, while still enabling the receiver to reconstruct a representation of the original video source material. In a preferred embodiment, a 60 frames per second progressively scanned video signal is transmitted at a 30 frames per second rate accompanied by decoding information. In short, every other frame is deleted and a respective nearest neighbor is selected to represent the deleted frame. This selection is represented by a digital selection information. Thus the transmitted frame information can be reconstructed into the original video source material after decoding with the digital selection information.

Patent
20 Aug 1990
TL;DR: In this paper, a pass-band reduction decoder is used to decode high-definition video signals at a desired frame frequency and with a desired number of lines, and a video processing circuit is connected to the outputs of the decoder.
Abstract: A device for conversion of frame frequency and number of lines for a high-definition television receiver which receives input video signals corresponding to different standards and to output a high-definition signal at a desired frame frequency and with the desired number of lines. The device uses a pass-band reduction decoder processing the high-definition input 625/2:1/50 Hz video signals. It includes: a memory, a decoder, and a processing circuit. The memory stores the input video signal frames at a frequency which is a function of the frame frequency of the input signal. This memory is read at the frame frequency desired at the output. The decoder is a pass-band reduction decoder connected to the memory and gives separately at its outputs even and odd frames, respectively, of a decompressed high-definition video signal output at the frame frequency desired. The video processing circuit is connected to the outputs of the pass-band reduction decoder and delivers a display video signal at the required frame frequency and number of lines.

Patent
Michael Emmerich Nagy1
31 Dec 1990
TL;DR: In this paper, a modified High-level Data Link Control (HDLC) control mechanism is incorporated into a normal data communication header thus eliminating the requirement for control messages to acknowledge or reject transmissions.
Abstract: A protocol and data packaging format optimized for transmissions over noisy communications links is disclosed. The data packaging format provides a modified High-level Data Link Control (HDLC) control mechanism incorporated into a normal data communication header thus eliminating the requirement for control messages to acknowledge or reject transmissions. An eight bit value is used to encode a toggle bit status for each of eight possible transmit sequence numbers. The eight bit value when used alone, constitutes a null frame indicating data-free feedback. The null frame is used to confirm receipt of data frames. This eight bit value is also used as the first byte of every data frame. The second byte of the data frame consists of a four-bit logical channel index, a single bit transmit toggle bit and a three-bit transmit sequence number. The second byte of the data frame may be followed by zero or more eight-bit data bytes. When a data frame arrives at a receiver, the bit value of the transmit toggle bit is copied into one of the eight bits of the first byte according to the transmit sequence number. The receiver sends the newly modified first byte back to the sender as part of a data frame transfer or as a null frame to confirm receipt of the data frame.

Patent
31 Aug 1990
TL;DR: In this article, the correlation coefficient of the feature amount by a frame unit with the last frame in the moving image composed of continuing plural images and deciding a point where the change rate of correlation coefficient is beyond the prescribed value as the change point of a cut was determined.
Abstract: PURPOSE:To perform high-speed change point detection by determining the correlation coefficient of the feature amount by a frame unit with the feature amount of the last frame in the moving image composed of continuing plural images and deciding a point where the change rate of the correlation coefficient is beyond the prescribed value as the change point of a cut. CONSTITUTION:A picture input processing 21 fetches the moving image by one frame in a frame memory. Further, the address on the medium of the frame is read as a time code. An image preprocessing 22 performs the thinning of the image and the compression of a color picture element. A frame correlation coefficient calculation processing 23 performs the calculation of the frame corre lation coefficient with the last frame to store the result in an array. A change rate calculation processing 24 calculates the change rate of the correlation coefficient to be stored in a variable DDT. A judgement processing 25 decides the change of the cut when the change rate DDT is higher than the prescribed value. An image change point registration processing 26 prepares the index information of the moving image cut. A frame update processing 27 updates the frame.

Patent
Robert Kutka1
21 Mar 1990
TL;DR: In this article, a 16×16 picture block of an HDTV television picture is transformed with a discrete cosine transformation as a 16 ×16 frame block or is separately transformed in two 16×8 field blocks according to a decision method.
Abstract: A 16×16 picture block of an HDTV television picture is transformed with a discrete cosine transformation as a 16×16 frame block or is separately transformed in respectively two 16×8 field blocks according to a decision method. Picture value differences between pairs of field picture elements of two picture lines of the same field are absolutely summed up to form a field sum. Likewise, picture value differences between pairs of frame picture elements of two picture lines of the frame are absolutely summed up to form a frame sum. The frame sum multiplied by a frame weighting factor is subtracted from the field sum to form what may be a positive decision result. Given a positive decision result, the frame is transformed; otherwise, the two fields are separately transformed.

Patent
14 Dec 1990
TL;DR: In this paper, the encoding mechanism produces a preliminary frame check sequence by encoding the information in an encoder using a generator polynomial G₄₈(x), which is a combination of the generator poynomials G�₆(x) and G ₃₂(x).
Abstract: A node operating in a network using the International Standard Organization (ISO) High-Level Data Link Control (HDLC) network protocol includes a mechanism for encoding information such that frames including the encoded information can be correctly interpreted by nodes operating in either of the standard 16-bit or 32-bit ISO-HDLC operating modes. The encoding mechanism produces a preliminary frame check sequence by encoding the information in an encoder using a generator polynomial G₄₈(x), which is a combination of the generator polynomials G₁₆(x) and G₃₂(x) which are used to produce frame check sequences for nodes operating in 16-bit or 32-bit modes, respectively. Before the information is encoded, the encoding mechanism sets the encoder to an initial condition using an initializing polynomial I₄₈(x). The preliminary frame check sequence is further encoded by adding to it a complementing polynomial C₄₈(x). The result is a 48-bit frame check sequence. The encoding mechanism appends the 48-bit frame check sequence to the information, and transmits the information and the appended 48-bit frame check sequence over the network as part of a frame.

Patent
15 Jun 1990
TL;DR: In this article, a time information generator for generating time information in an arbitrary frame of video information varying with changes in time on the basis of time information delivered at a predetermined time interval from a video information forming device for forming video information is employed.
Abstract: In a time information generator for generating time information in an arbitrary frame of video information varying with changes in time on the basis of time information delivered at a predetermined time interval from a video information forming device for forming video information, there is employed a scheme to discriminate between operational modes of the video information forming device to form, by an interpolative operation, time information in an arbitrary frame of video information from time information delivered from the video information forming device every n (n=2, 3, 4, . . .) frames of video information on the basis of the result indicative of the discriminated operational mode.

Journal ArticleDOI
TL;DR: A low bit-rate video codec based on motion vector replenishment that has a comparable compression efficiency with that of the frame dropping method, but does not introduce any picture ‘jerkiness’.
Abstract: A low bit-rate video codec based on motion vector replenishment is described. Motion vectors are used to update pictures at full frame rate. In addition, part of each frame is conditionally updated with a strip of interframe video data. The video data fill the remaining channel capacity not used for motion vectors. Thus under most conditions, each frame is fully updated by motion vectors and partially with interframe video data. This method has a comparable compression efficiency with that of the frame dropping method, but does not introduce any picture ‘jerkiness’. Finally the application of the proposed method to packet video networks is examined.

Patent
Koji Wada1
25 Jun 1990
TL;DR: In this paper, a control unit includes a first frame buffer and a second frame buffer as the memory area, a microprocessor, a shift circuit and a mask pattern generating circuit.
Abstract: A control unit includes a first frame buffer and a second frame buffer as the memory area, a microprocessor, a shift circuit, a shift number register and a mask pattern generating circuit An image data signal is output from the microprocessor Shifting of respective words of the image data is performed by the shift circuit in compliance with a shift number signal output from the shift number register, for writing the words of the image data into the first and second frame buffers from desired bit positions corresponding to the shift numbers Mask pattern signals nullifying bits, except the bits corresponding to the image data in output data of the shift circuit, are output from the mask pattern generating circuit Selection between a high order side portion and a low order side portion of each word of the image data to be written into the frame buffers is performed by the least significant bit of an address signal The address value input into the first buffer is incremented by "1" in an adder The first frame buffer is accessed by address line output from the adder from which the least significant bit is removed, while the second frame buffer is accessed by address line output from the microprocessor Access to address N or N+1 in the first frame buffer and address N+1 or N in the second frame buffer is thereby achieved

Patent
15 Mar 1990
TL;DR: In this paper, the transmission side picture discrimination section 14 judges to which scene stored in the frame memories 2-1 to 2-3 an inputted moving picture signal M1 is similar, uses a frame memory giving the similar scene for coding and informs the frame memory section signal S representing which frame memory is selected to the reception side via a Frame Memory section signal line 15.
Abstract: PURPOSE:To reduce code quantity by using a signal stored in a frame memory giving a most similar signal among signals in plural frame memories so as to apply coding based on a difference signal of a frame memory to be coded. CONSTITUTION:Plural memories 2-1 to 2-3 storing plural scenes are provided to both a transmission side and a reception side, and a same referenced picture is inputted in advance respectively. Then the transmission side picture discrimination section 14 judges to which scene stored in the frame memories 2-1 to 2-3 an inputted moving picture signal M1 is similar, uses a frame memory giving the similar scene for coding and informs a frame memory section signal S representing which frame memory is selected to the reception side via a frame memory section signal line 15. Thus, when plural scenes are sent alternately, the increase in the coding information quantity when the scene is changed is suppressed.

Patent
31 Aug 1990
TL;DR: In this article, the authors propose to improve the resolution of a moving vector by obtaining not only an inter-frame moving vector but also the moving vector for the unit of M frame and executing motion compensation by an estimated value from these vectors.
Abstract: PURPOSE:To improve the resolution of a moving vector by obtaining not only an inter-frame moving vector but also the moving vector for the unit of M frame and executing motion compensation by an estimated value from these vectors. CONSTITUTION:The first frame is read from a frame memory FM 1 through a selector SEL 2 and the decoded picture of the 0-th frame is read from a decoded picture frame memory DFM. Then, the pictures are respectively supplied to a motion compensation circuit MCO. The detection of a moving vector MVO is executed in the circuit MCO and the vector is supplied to a moving vector decision circuit SEL-MV. A residual moving vector MVe corresponding to the vector MVO is inputted from a buffer BMVe for residual moving vector to a division and subtraction circuit CAL-MVei. At such a time, when a scene change is not detected, the vector is remained from a residual frame counter FMCNT to the circuit CAL-MVei and a frame count is supplied. An estimated moving vector MVei is supplied from the circuit CAL-MVei to a moving vector select circuit SEL-MV.

Patent
Bernhard Dohmann1
07 Mar 1990
TL;DR: In this paper, a method for reproducing received video signals which represent images of various phases of movement of recorded objects, motion vectors being received in addition to the video signals and several frames being represented within one frame period of the received video signal is presented.
Abstract: In a method for reproducing received video signals which represent images of various phases of movement of recorded objects, motion vectors being received in addition to the video signals and several frames being represented within one frame period of the received video signals, one of the frames to be represented for each received frame corresponds to a received frame. Signals for other frames to be represented (intermediate frames) are obtained from the signals of two successively received frames by the fact that the picture elements of one received frame are displaced from intermediate frame to intermediate frame by a fraction of the motion vector valid for the respective picture element, the fraction being obtained from the number of frames to be represented per received frame, and that the amplitudes of the video signals of the picture elements of one received frame and of the picture elements of the subsequent received frame which is distant by the motion vector are averaged between the corresponding picture elements of the received frames taking into consideration the position of the respective picture element of the frame to be represented.

Patent
20 Mar 1990
TL;DR: In this paper, the authors propose a device for converting frame rate for a high definition television receiver able to receive as input chrominance signals or luminance signals with different characteristics but with the same frame rate, and yield at output a signal with a different frame rate.
Abstract: Device for converting frame rate for a high definition television receiver able to receive as input chrominance signals or luminance signals with different characteristics but with the same frame rate, and to yield at output a signal with a different frame rate. This device comprises: - at least two frame memories (12, 12') for storing at a sampling rate dependent upon the rate of sampling the input signal, the even frames in one memory and the odd frames in the other memory; - a device for processing the input signals of static type (30) connected to the output of the frame memories and delivering a signal with the desired frame rate and desired number of lines; - a device for processing the input signals of dynamic type (21) connected alternately to the output of the even and odd frame memories through a first switching device (210), and delivering a signal with the desired frame rate and desired number of lines; - a second switching device (23) connected respectively to the devices of static type (30) and of dynamic type (21) as a function of the DATV signal. Application to high definition television.

Patent
31 Aug 1990
TL;DR: In this paper, the sketch plane data is overlaid on the video image data and may be assigned a transparency value (α), in this overlaying process a system for displaying a plurality of frames of a sequence of video images includes a mass storage unit (10) for the frames of the video images.
Abstract: Each one of a plurality of frames of video image data (12, 26) in a mass storage unit (10) has a frame of sketch plane data (14, 30) associated with it. The sketch plane data can be selectively output via a switch (32) to preview proposed modifications to a sequence of video images. The sketch plane data may be overlaid on the video image data and may be assigned a transparency value (α) in this overlaying process a system for displaying a plurality of frames of a sequence of video images includes a mass storage unit (10) for the frames of the video image. This outputs individual frames to a plurality of frame stores (1-5). The output of each frame store is multiplied by a fraction (α1-5) in a multiplier (71-75). The outputs of all the multipliers (71-75) are combined in an adder (76) to give a resultant video image which includes proportions of the frames stored in the frame stores (1-5) in dependence upon the fractions by which they are multiplied.

Proceedings ArticleDOI
01 Aug 1990
TL;DR: A new frame content independent stripping (FCIS) algorithm for token rings is discussed and it is demonstrated that the robustness of the algorithm to errors and its reliability is as good as the inherent mechanisms of the token ring.
Abstract: Token rings have the property that a station that transmits a frame on the ring is responsible for removing the frame after it has been delivered to the destination stations. The algorithm to perform the frame removal is called 'Frame Stripping'. Most existing algorithms strip frames based on their content. This is not always adequate. The need for a new algorithm arises from the fact that frames transmitted by a station need not have the station's own address as the source address for a variety of reasons - such as when a bridge transmits a frame or when another address is used as the source address by a station instead of its original station address. This paper discusses a new frame content independent stripping (FCIS) algorithm for token rings.The FCIS algorithm counts the number of frames transmitted by the station after capturing the token. In addition, the station places a special delimiter frame at the end of the transmission of frames, before releasing the token. The station then strips all received frames until either the number of frames stripped equals the number of frames transmitted or when either the delimiter frame or a token is received. We demonstrate that the FCIS algorithm has a minimal impact on the performance of the ring. We study the robustness of the algorithm to errors and demonstrate that its reliability is as good as the inherent mechanisms of the token ring.The algorithm studied here is very simple to implement and interoperates with other stations not implementing this algorithm. The algorithm places no topological restrictions on the network and has the attractive feature of removing large fragments and no-owner frames.