Abstract: A method and apparatus are provided for performing entropy encoding on a fine granular scalability layer. A method of entropy encoding on a plurality of current coefficients of a quality layer among a plurality of quality layers of an image block divided into the plurality of quality layers, includes determining a coding pass with respect to each of the current coefficients, selecting a context model with respect to each of the current coefficients using at least one lower coefficient corresponding to each of the current coefficients if the coding pass is a refinement pass, and performing arithmetic encoding a group of coefficients having a same selected context model among the current coefficients by using the selected context model.

Abstract: Bitstream switching for multiple bit rate video streaming uses a first algorithm to generate multiple encoded bitstreams of a piece of video at different bit rates. A different algorithm is used to generate multiple side bitstreams of the video. Each side stream is generated at respective ones of the different bit rates. The side bitstreams comprise switching frames to switch from a first bitstream encoded with the first algorithm at a first bit rate to a second bitstream encoded with the first algorithm at a different bit rate. At non-bitstream switching points, frames from respective ones of bitstreams generated with the first algorithm are streamed for presentation. At a bitstream switching point, a switching frame from a respective side bitstream is transmitted to a decoder to transition streaming from a first of the respective ones of the bitstreams to a second of the respective ones of the bitstreams.

Abstract: A method of and apparatus to efficiently encode and/or decode an image are provided. The method of encoding an image includes: calculating a motion vector of a main reference image by estimating a motion of a current image with reference to the main reference image; deriving a motion vector of at least one auxiliary reference image from this motion vector; generating a prediction image of the current image by using a motion vector of the main reference image or the motion vector of the at least one auxiliary reference image; and encoding a residue image between the current image and the prediction image.

Abstract: The present invention makes it possible to include, when encoding processing is applied to three color components using a 4:0:0 format, data for one picture in one access unit and makes it possible to set the same time information or the same set encoding modes among the respective color components. In an image encoding system for applying compression processing to an input image signal including a plurality of color components, encoded data obtained by independently subjecting an input image signal of each of the color components to encoding processing and a parameter indicating which color component the encoded data corresponds to are multiplexed with a bit stream. In an image decoding system for inputting a bit stream in which an image signal including a plurality of color components is compressed to perform decoding processing, decoding processing of the encoded data of each of the color components is performed using a parameter indicating which color component the encoded data corresponds to.

Abstract: Procedure and device for shaping frames of a video sequence or group of pictures GOP using a mixing table adapted for arranging the frames of the highest importance regularly while leaving intervals between them, filling the intervals with frames of lesser importance, and coding the new sequence thus obtained.

Abstract: A method for temporal decomposition and reconstruction of an input video signal is disclosed. The method can use a prediction process and an update process in the framework of motion compensated temporal filtering (MCTF), the method can employ a block based video codec having an encoder and a decoder. A residual picture signal and a low-pass picture signal are received, the low-pass picture signal having a lesser precision than that of the residual picture signal, and in response to the residual picture signal being decoded in the decoder, the residual picture signal is clipped and reduced in precision, such that the low-pass picture signal, a reconstructed picture, and the residual picture signal have equal precisions. An inverse MCTF process, using a lifting scheme, can be performed on the clipped residual signal.

Abstract: In one embodiment, a method that partitions a first sequence of pictures into plural representations, encodes each of the plural representations independently of each other with a common time base, provides a first portion of the plural encoded representations based on a first network condition, the first portion having a first bit-rate, and provides a second portion of the plural encoded representations having a second bit-rate different than the first bit-rate, wherein a switch from providing the first portion to providing the second portion is responsive to a second network condition different than the first network condition.

Abstract: Image denoising techniques include determining wavelet-domain noise model and a non-parametric multivariate wavelet description from the image signal for raw image data. A noise corrected image may then be determined from the image signal, the wavelet-domain noise model and the non-parametric, multivariate wavelet description and the image signal.

Abstract: An adaptive picture mode selection transcoder selects an encoding mode in a second format for frames of video previously encoded in a first format by determining a magnitude of interlacing phenomenon in the using picture information obtained during decoding of the video from the first format. In one aspect, the picture information includes discrete cosine coefficients for macroblocks in the frame. In another aspect, the picture information includes an encoding mode for the macroblocks in the first format. In yet another aspect, the picture information includes motion vector information for the macroblocks. In still another aspect, the determining is specific to an encoding mode for the frame.

Abstract: Techniques are provided herein to produce encoded video bitstreams and to similarly decode encoded video bitstreams according to a coding standard not supported by an on-chip encoder/decoder. For purposes of encoding, a video sequence is received at a first device. A first bitstream is generated at the first device by encoding the video sequence according to a first coding standard and information associated with the video sequence is generated at the first device according to a second coding standard. The first bitstream and the information are then transmitted to a second device. At the second device the first bitstream is decoded to produce a second bitstream. The second bitstream and the information are combined by removing syntax elements associated with the first coding standard from the second bitstream and adding the information to produce a third bitstream according to the second coding standard. Similar techniques are provided for decoding an encoded bitstream to recover a video sequence.

Abstract: Embodiments include a codec for use in a videoconferencing or similar system includes a video encoder pipeline that has a pre-processor component that is optimized to detect faces and compress the facial video data in an optimum manner. The codec has a pre-processing step that analyzes each frame on a per macroblock basis to determine the mathematical activity level per block. The activity level calculation is used as a parameter to the bitrate control module of the encoder to control the quantization, and thus the fine grained quality of the output data. An object detection module (e.g., a face detector) is placed in the pre-processing step. The object detection data is then combined with the activity level and object detection certainty value through a combinatorial algorithm comprising a weighted average or normalized multiplication process.

Abstract: At least one exemplary embodiment is directed to an image coding apparatus configured to encode moving image data including: a coding unit configured to encode each picture in the moving image data in a unit of a first block; a luminance change detection unit configured to divide the moving image data into a plurality of second blocks and to detect a luminance change block in which a luminance change occurred from the plurality of the second blocks in one picture; and a code amount adjustment unit configured to increase an amount of code allocated to the first block if the first block corresponds to the luminance change block detected by the luminance change detection unit.

Abstract: A scalable video bitstream may have an H.264/AVC compatible base layer (BL) and a scalable enhancement layer (EL), where scalability refers to color bit depth. The H.264/AVC scalability extension SVC provides also other types of scalability, e.g. spatial scalability where the number of pixels in BL and EL are different. According to the invention, BL information is upsampled (TUp,BDUp) in two logical steps in adaptive order, one being texture upsampling and the other being bit depth upsampling. Texture upsampling is a process that increases the number of pixels, and bit depth upsampling is a process that increases the number of values that each pixel can have, corresponding to the pixels color intensity. The upsampled BL data are used to predict the collocated EL. A prediction order indication is transferred so that the decoder can upsample BL information in the same manner as the encoder, wherein the upsampling refers to spatial and bit depth characteristics.

Abstract: In a method and an encoder for encoding a digital video sequence, the digital video sequence having some sets of images including a disparity map, the disparity map being used to reconstruct one image of a set of images from another image of the set of images, the method includes the steps of encoding a type of a disparity map to be used for the reconstruction of an image, and encoding the disparity map.

Abstract: An encoder, a decoder, corresponding encoding and decoding methods, video signal structure, and storage media are provided for scalable video encoding and decoding. A scalable video encoder includes an encoder for encoding a picture to form a base layer bitstream and an enhancement layer bitstream. The base layer bitstream and the enhancement layer bitstream are formed by partitioning the picture into a plurality of image blocks, grouping the plurality of image blocks into one or more slice groups in the base layer bitstream and into two or more slice groups in the enhancement layer bitstream, encoding all of the one or more slice groups in the base layer bitstream and less than all of the two or more slice groups in the enhancement layer such that at least one slice group from among the two or more slice groups is intentionally un-encoded in the enhancement layer bitstream, encoding a syntax element in a header to indicate the at least one intentionally un-encoded slice group in the enhancement layer.

Abstract: Provided is a moving picture decoding device (100) including a direct mode availability judging unit (109), when a decoding target block should be decoded in a temporal direct mode, judging whether or not a reference picture for a decoded picture, which is in a vicinity of the decoding target block in terms of display time, is usable as a forward reference picture for the decoding target block. Further, the moving picture decoding device (100) includes a substitute process performing unit (110) decoding the decoding target block in the temporal direct mode using a picture other than the reference picture, or decoding the decoding target block in other encoding mode.

Abstract: A method to control weighted prediction for video compression. The method comprises (A) generating statistics based upon analysis of a plurality of video frames, (B) detecting a fade condition based upon the statistics, (C) generating one or more estimated weighted prediction parameters based upon the statistics when the fade condition is detected and (D) encoding the plurality of video frames. The encoding takes into account the estimated weighted prediction parameters when generated.

Abstract: The present invention is directed to a method for determining whether a current macroblock and an adjacent macroblock thereof are located in the same slice. The method is used in a predetermined process for a block-based digitally encoded image. The block-based digitally encoded image is represented as an encoded bit-stream and each macroblock therein is assigned a sequence characteristic number. The method includes: providing a memory space for storing and tracing a slice changing point; initializing the slice changing point to a predetermined number; checking the encoded bit-stream, and when the current slice is determined to change, setting the slice changing point to a derived sequence characteristic number derived from the sequence characteristic number of the current macroblock; and determining whether the current macroblock and the adjacent block thereof are in the same slice according to a comparison result between the sequence characteristic number of the adjacent block and the slice changing point.

Abstract: Video decoding innovations for multithreading implementations and graphics processor unit (“GPU”) implementations are described. For example, for multithreaded decoding, a decoder uses innovations in the areas of layered data structures, picture extent discovery, a picture command queue, and/or task scheduling for multithreading. Or, for a GPU implementation, a decoder uses innovations in the areas of inverse transforms, inverse quantization, fractional interpolation, intra prediction using waves, loop filtering using waves, memory usage and/or performance-adaptive loop filtering. Innovations are also described in the areas of error handling and recovery, determination of neighbor availability for operations such as context modeling and intra prediction, CABAC decoding, computation of collocated information for direct mode macroblocks in B slices, reduction of memory consumption, implementation of trick play modes, and picture dropping for quality adjustment.

Abstract: A system and method for generating a number of candidate beamforming vectors for transmission of high definition video data over a wireless medium is disclosed. In one embodiment the system comprises a memory configured to store a plurality of candidate beamforming vectors and a processor for executing software configured to i) determine the number (N) of transmitter antennas and the number (Q) of candidate beamforming vectors to be constructed, ii) provide a discrete Fourier transform (DFT) matrix comprising Q rows and Q columns, iii) select N rows of the DFT matrix, and iv) generate an abridged matrix using the selected N rows, wherein each column of the abridged matrix is configured to be used as a candidate beamforming vector for transmission of a high definition video data signal.

Abstract: A transmission side apparatus selectively inserts video stream data and MPEG data into a video storage area to store video stream data in a frame structure of an SDI system and generates SDI data of which the identification ID indicating a type of data included therein by an SDI data insertion processing device and transmits the SDI data to a coaxial cable from a transmission interface device. A reception side apparatus receives the SDI data transmitted via the coaxial cable by a reception interface device, extracts data to be transmitted from a video storage area of received SDI data and determines whether the data to be transmitted is the video stream data or the MPEG data by an SDI data extraction processing device to output it. Therefore, a high-speed MPEG data transmission assuring frame/field accuracy becomes possible by using an existing technology and device.

Abstract: A method for converting an encoded digital video signal from a decoded relatively lower frame rate to a decoded relatively higher frame rate, including the following steps: deriving, from the encoded video signal, at the relatively lower frame rate, a decoded high resolution component and a decoded low resolution component; increasing the frame rate of the decoded low resolution component by interpolating successive frames of the decoded low resolution component; increasing the frame rate of the decoded high resolution component by duplicating frames of the decoded high resolution component; and obtaining an output relatively higher frame rate video signal by combining the increased frame rate low resolution component and the increased frame rate high resolution component.

Abstract: A distributed video encoder and decoder and a distributed video decoding method using adaptive quantization are provided. Adaptive quantization is performed at the time of encoding and decoding so that limited resources and information can be efficiently used, and a predetermined bit which is previously defined is included in a position of a bit which does not need to be transmitted for channel coding, thereby improving a bit rate-distortion performance as a whole.

Abstract: In one embodiment, a method that includes receiving plural representations of a single video signal, wherein each includes a respective sequence of latticed pictures, wherein each latticed picture in each of the plural representations originates from a corresponding respective picture of the video signal; processing the plural representations; separating the processed plural representations into plural segments, each segment consisting of a respective sequence of processed latticed pictures from one of the processed plural representations; arranging the plural segments in successive segments distributions intervals (SDIs) according to a first temporal order and a second temporal order, the successive SDIs to be provided in a video stream; and promoting one or more of the processed latticed pictures among one or more of the plural segments, the promotion corresponding to an increase in picture importance.

Abstract: Video traffic transmission using video quality metrics may be provided. A video stream comprising a plurality of frames may be received. For each frame, a determination may be made whether or not to drop the frame rather than saving and/or transmitting it to an end user, thereby conserving network bandwidth. This determination may be made on the basis of a threshold metric, such as a size ratio between frames, an average frame size, an average quality metric.

Abstract: A system and method are disclosed for providing improved processing of video data. A multi-instance encoding module receives combined video and audio input, which is then separated into a video and audio source streams. The video source stream is pre-processed and corresponding video encoder instances are initiated. The preprocessed video source stream is split into video data components, which are assigned to a corresponding encoder instance. Encoding operations are performed by each video encoder instance to generate video output components. The video output components are then assembled in a predetermined sequence to generate an encoded video output stream. Concurrently, the audio source stream is encoded with an audio encoder to generate an encoded audio output stream. The encoded video and audio output streams are combined to generate a combined encoded output stream, which is provided as combined video and audio output.

Abstract: A transcoder for decoding a motion picture stream encoded by using a first coding scheme, which provides intraframe coding and interframe predictive coding, and encoding the decoded motion picture stream by using a second coding scheme, the transcoder including: at decoder for decoding an input motion picture stream and detecting sub-information indicating whether an intraframe coding scheme or an interframe predictive coding scheme is used; and an encoder for changing the frame to be referenced at the time of coding depending on whether the sub-information indicates the use of the intraframe coding scheme or the interframe predictive coding scheme.

Abstract: Direct macroblock mode techniques for high performance hardware motion compensation are described. An embodiment includes a hardware motion compensation graphics display device driver. More specifically, an embodiment mitigates a macroblock data parsing bottleneck in the display device driver by directly generating macroblock instructions and storing them in a dedicated buffer. For example, an embodiment includes an independent direct memory access instruction execution buffer for macroblock instructions separate from the direct memory access instruction execution buffer for all other hardware motion compensation instructions. Other embodiments are described and claimed.

Abstract: A method and apparatus for encoding and decoding video performs transformation of at least a portion of a high-resolution video frame into a low resolution image and a plurality of enhancement data sets, encodes the low resolution image as a primary coded picture in a bitstream format and encodes each of the plurality of enhancement data sets as a different redundant coded picture in the bitstream format. For decoding, a decoded low resolution image and a plurality of decoded enhancement data sets are generated and an inverse transform is performed to construct a decoded high-resolution image. The primary coded picture and a redundant coded picture may be formatted according to the ITU-T H.264 Advanced Coding specification. The transform may be a polyphase or a sub-band transform.

Abstract: A method and system for communication uncompressed video over a wireless communication medium is provided. Pixels in an uncompressed video frame are grouped into groups of pixels for wireless transmission from a transmitter to a receiver, wherein the neighboring pixels in each group possess (have) high spatial correlation. Upon receiving the pixel groups, the receiving station reconstructs the video frame from the received pixel groups, wherein upon detecting an erroneous pixel in a group, information from one or more neighboring pixels in that pixel group are used in place of the erroneous pixel. Further, FEC encoding is utilized recovering bit-errors.

Abstract: Disclosed herein is an information processing apparatus configured to process a code stream having a first resolution with data having the first resolution scalably encoded, including setting means, first creating means, authoring means, second creating means, and code stream replacing means.

Abstract: There is provided an image reproduction apparatus for seamlessly reproducing a connected stream which is obtained by connecting plural streams that are respectively coded by different codec methods. An image reproduction apparatus (100) for reproducing a connected stream which is obtained by connecting plural streams of different codec methods such as an MPEG-2 method and an MPEG-4 AVC method is provided with a stream buffer (102) in which the connected stream Bst is stored, and plural decoders Dd1˜Ddn corresponding to the various kinds of codec methods, and a decoder for decoding each stream in the connected stream Bst that is outputted from the stream buffer (102) is selected from among the plural decoders according to the codec method of each stream.

Abstract: The disclosure describes FGS video coding techniques that use cycle-aligned fragments (CAFs). The techniques may perform cycle-based coding of FGS video data block coefficients and syntax elements, and encapsulate cycles in fragments for transmission. The fragments may be cycle-aligned such that a start of a payload of each of the fragments substantially coincides with a start of one of the cycles. In this manner, cycles can be readily accessed via individual fragments. Some cycles may be controlled with a vector mode to scan to a predefined position within a block before moving to another block. In this manner, the number of cycles can be reduced, reducing the number of fragments and associated overhead. The CAFs may be entropy coded independently of one another so that each fragment may be readily accessed and decoded without waiting for decoding of other fragments. Independent entropy coding may permit parallel decoding and simultaneous processing of fragments.

Abstract: Video processing method and means for enhancing a video stream, by computing transform coefficients using a spatio-temporal transform comprising a spatial subband transform and a causal time wavelet transform performing filterings with multiscale causal wavelets, modifying the transform coefficients with a nonlinear processing, and computing a processed video stream from the modified transform coefficients using a spatio-temporal reconstruction transform comprising an inverse subband transform and a short delay inverse time wavelet transform, where the short delay inverse time wavelet transform is implemented with wavelet filters modified with window functions to control the processing delay of the entire video processing method.

Abstract: A method and apparatus for estimating motion and occlusion is disclosed. In one aspect, a method of estimating motion and occlusion between three or more frames in a video stream includes identifying a plurality of motion vector candidates for each of a forward direction and a backward direction, generating a plurality of candidate pairs, determining an energy for at least some of the plurality of candidate pairs using a processor by jointly evaluating the forward motion vector and the backward motion vector of at least some of the plurality of candidate pairs based on interframe difference, spatial motion correlation, temporal motion correlation, and spatial occlusion correlation, and estimating motion and occlusion between the three or more frames by selecting a candidate vector from the plurality of candidate vectors based on the determined energies.

Abstract: A picture processing device is capable of performing a process of superposing a sub picture image on a moving picture image which forms a base. The device includes a detection unit for detecting a region in which a movement amount is small, out of a plurality of regions in which the sub picture image may be superposed on the moving picture image which forms a base, and a picture image superposing unit for superposing the sub picture image on the region in which the movement amount is small, the region being detected by the detection unit.

Abstract: A digital broadcasting system and a data processing method thereof are provided. An apparatus for generating a transport stream includes a packet generator for generating a plurality of packets by compressing video and audio signals; a counter for counting the number of the generated packets and outputting the count result; and an information controller for determining whether the count result of counting reaches a predetermined number of packets, inserting packet start information into the packets according to the count result, and transmitting the packets with the packet start information inserted therein. The packet start information is added to a packet to be transmitted in a transmitter, and the packet start information is extracted from the received packet in a receiver, so that the waste of frequency resources can be reduced.

Abstract: Learning-based image compression is described. In one implementation, an encoder possessing a first set of learned visual knowledge primitives excludes visual information from an image prior to compression. A decoder possessing an independently learned set of visual knowledge primitives synthesizes the excluded visual information into the image after decompression. The encoder and decoder are decoupled with respect to the information excluded at the encoder and the information synthesized at the decoder. This results in superior data compression since the information excluded at the encoder is dropped completely and not transferred to the decoder. Primitive visual elements synthesized at the decoder may be different than primitive visual elements dropped at the encoder, but the resulting reconstituted image is perceptually equivalent to the original image.

Abstract: There are provided methods and apparatus for adaptive geometric partitioning for video encoding and decoding. An apparatus includes an encoder for encoding image data corresponding to pictures by adaptively partitioning at least portions of the pictures responsive to at least one parametric model. The at least one parametric model involves at least one of implicit and explicit formulation of at least one curve.

Abstract: Disclosed are systems and methods used in motion estimation and particularly for data compression. Embodiments of the invention may store and operate on an n-bit value in less than n bits. In one embodiment, if the multi-bit value is less than a threshold, then the multi-bit value is stored in the reduced-bit storage directly, with no loss of precision. If the multi-bit value is greater than the threshold, then the Most Significant Bits (MSBs) of the multi-bit value are shifted into the reduced-bit storage, and a compression flag set. To decompress, if the compression flag was not set, the bits stored in the reduced-bit storage are merely copied back into the multi-bit value directly. If the compression flag was set, then the bits stored in the reduced-bit storage are shifted (left) by the same amount they were shifted (right) during compression, and an error-minimizing value is added.

Abstract: A video coder performs a motion-compensated prediction both in the base layer and in an enhancement layer to determine motion data of the enhancement layer by using the motion data from the base layer and/or to predict sequences of residual error pictures after the motion-compensated prediction in the enhancement layer by using sequences of residual error pictures from the base layer via an intermediate layer predictor. On the decoder side, an intermediate layer combiner is used for canceling this intermediate layer prediction. Thereby, the data rate is improved compared to scalability schemes without intermediate layer prediction with the same picture quality.

Abstract: A video recording system is provided in which reliability of information supplied from a video camera to control recording and the like is improved when video being obtained by a video camera is recorded in one or more VTRs. In the video recording system, the video camera multiplexes: control information to control the VTR in accordance with an operational status of an operation unit of the video camera and a video signal from an imaging unit of the video camera, and transmits the result to the VTR as a transmission serial digital signal. The VTR separates the video signal and the control information from the serial digital signal and controls a video recording system of the VTR based on the control information.

Abstract: A video decoder cache used for motion compensation data may be dynamically reconfigured. In some embodiments, it may be reconfigured on picture or frame boundaries and in other embodiments it can be reconfigured on sequence boundaries. The cache may be flushed on each boundary to enable such reconfiguration.

Abstract: This invention provides a technique of efficiently executing prediction error calculation processing while preventing any increase in the circuit scale. An encoding apparatus of this invention includes a determination unit determining the size of an encoding target pixel block, a plurality of predicted value calculation units, a first error calculation unit calculating, based on a predicted value from a predicted value calculation unit in a prediction mode using one reference pixel value, the prediction error in the determined pixel block size, a second error calculation unit calculating, based on a predicted value from a predicted value calculation unit in a prediction mode for calculating a predicted value based on a plurality of reference pixel values, the prediction error in the determined pixel block size, and a unit determining, based on the sum of the prediction errors from the first and second error calculation units, a prediction mode.

Abstract: System and method of providing improved signal compression using frame decimation through frame simplification and generating an encoded bitstream of video frames therefrom are disclosed. The encoding method comprises zeroing a difference frame generated by an encoder by using a feedback loop that injects a reconstructed frame, generated by the encoder of the difference frame, as a next frame of the video frames to be processed by the encoder. The encoding system comprises an input configured to provide a stream of video frames; a first process configured to generate a difference frame, and a second process configured to generate a reconstructed frame. A feedback loop of the system is configured to inject a generated reconstructed frame from the second process of a generated difference frame from the first process as a next frame of the video frames in the stream to be processed into the encoded bitstream by the encoding system.

Abstract: The present disclosure is directed towards a system and method for time optimized encoding. Time optimized encoding maximizes the usage of the multiple encode nodes or clusters by parallelizing time-intensive encoding with minimal or no sacrifice of encoding quality. Thereby, reducing the time required for accomplishing the time intensive encoding and increasing the efficient throughput of multiple encodes through the system.

Abstract: A method of scanning content comprises receiving an input signal indicating that content being displayed is to be scanned, sensing a duration of the received input signal, and scanning the content at a rate based at least in part on the sensed duration of the received input.

Abstract: A digital broadcasting transmitter, a turbo stream processing method thereof, and a digital broadcasting system having the same. The digital broadcasting transmitter includes a first compressor, forming a normal stream by compressing audio and video signals in a first compression format, a second compressor, forming a turbo stream by compressing the audio and video signals in a second compression format, a transport stream (TS) constructor, compressing the turbo stream in an H.264 format and forming the dual transfer stream by multiplexing the normal stream and the turbo stream, and a TS processor, robustly processing the dual transfer stream transmitted from the TS constructor, thus enhancing the transmission efficiency.

Abstract: An accelerated screen codec technique is described that provides a general screen compression framework, which, in one embodiment, is Graphics Processor Unit (GPU) friendly. In one embodiment, in order to compress screen data, blocks in a compound screen image containing both images and text are segmented into text blocks and pictorial blocks using a simple gradient-based procedure. The text and pictorial blocks are then compressed respectively via different compression techniques. Additionally, a GPU acceleration architecture of one embodiment of the accelerated screen codec technique provides a screen codec that maximally exploits a GPU's high parallelism and reduces the download bandwidth from GPU to Computer Processing Unit (CPU).

Abstract: A moving picture coding apparatus divides each frame of a moving picture into parts and assigns the parts to different coding units, which compressively code their respective parts. The coding process includes motion compensation with respect to a reference frame. Each coding unit has its own reference frame memory. To generate reference frame data, each coding unit receives, decodes, and decompresses the coded data generated by at least one other coding unit, as well as decompressing the data it has coded itself. Consequently, only ordinary coded data have to be passed between different coding units, which saves bandwidth and eliminates the need for special coding hardware and development and testing tools.