Abstract: A picture coding method for preventing occurrence of a malfunction includes: an adding step (Step S203) of coding a memory management command for the first picture of a predetermined picture number and adding the coded command to the second picture that is different from the first picture; and a re-adding step (Steps S205 to S208) of coding the memory management command again and re-adding re-coded command to the section other than the second picture in a coded picture stream VSt. In the re-adding step, re-addition of the memory management command is prohibited when the first picture is not immediately previous in coding order among the pictures of the predetermined picture number included in the coded picture stream VSt.

Abstract: A method and a device for intra prediction coding and decoding of an image in which the intra prediction according to a selected intra prediction mode based on a direction of an input image is performed are provided. In the method and the device, an input block is frequency transformed, a direction of pixels in the input block is determined, and the intra prediction is performed according to an intra prediction mode that has the most similar direction to the determined direction of the pixels.

Abstract: A method of displaying a shot image on a car reverse video system is described, which is applicable for adjusting an image displaying angle of the car reverse video system with a fixed lens according to a distance from an obstacle. Firstly, a scene image behind the car is shot through using the lens. Then, a distance detector is activated to detect a distance from the obstacle behind the car. If the distance from the obstacle is smaller than a preset distance, a scaling ratio is set according to a ratio of the preset distance to the distance from the obstacle. Finally, according to the scaling ratio and a distance between each pixel in the scene image and a reference point, all pixel pitches are adjusted, such that the scene image is deformed, and thus achieving an effect of switching a viewing angle of the scene image.

Abstract: A compression technique includes storing respective fixed-size symbols for each of a plurality of words in a data block, e.g., a cache line, into a symbol portion of a compressed data block, e.g., a compressed cache line, where each of the symbols provides information about a corresponding one of the words in the data block. Up to a first plurality of data segments are stored in a data portion of the compressed data block, each data segment corresponds to a unique one of the symbols in the compressed data block and a unique one of the words in the cache line. Up to a second plurality of dictionary entries are stored in the data portion of the compressed cache line. The dictionary entries can correspond to multiple ones of the symbols.

Abstract: A scaleable macro block rate control method particularly well-suited for MPEG video. There is provided a method to easily derive a quantization parameter (QP) value using information such as bit usage, previous QP values and SAD values from the past encoded and future frames. The method utilizes quantization estimation techniques based on statistical relationships between different intensity measures, such as distortion intensity, absolute difference intensity and mean of absolute difference intensity. The method is well-suited to applications utilizing MPEG video such as MPEG-1, MPEG-2, MPEG-4, JVT/H.264 standards and so forth.

Abstract: A video decoder, a video decoding method, a video encoder and a video encoding method are disclosed. A video decoder for decoding a video bitstream for an image block includes a motion vector resolution reducer and a motion compensator. The motion vector resolution reducer is for receiving decoded high resolution motion vectors included in the video bitstream and for reducing an accuracy of the high resolution motion vectors to correspond to a low resolution. The motion compensator, in signal communication with the motion vector resolution reducer, is for forming a motion compensated high resolution prediction using the reduced accuracy motion vectors. The video encoder for encoding scalable video comprises a motion compensator for forming a motion compensated full resolution prediction and combining the motion compensated full resolution prediction from an image block to form a prediction residual. The prediction residual is downsampled to form a low resolution downsampled prediction residual and then coded.

Abstract: Techniques and tools for skip modes in encoding and decoding of inter-layer residual video are described. For example, an encoder encodes multiple macroblocks of a picture of inter-layer residual video. For a current macroblock that is skipped, the encoder selects a skip mode from among multiple available skip modes and uses the selected skip mode when encoding the current macroblock. The skip modes can include intra skip mode and predicted-motion skip mode. A corresponding decoder, for the current macroblock, selects and uses the skip mode for the current macroblock during decoding. As another example, an encoder encodes multiple channels of a picture of inter-layer residual video. For each channel, the encoder determines whether to skip the channel. The encoder signals channel skip information to indicate which channels are skipped. A corresponding decoder parses the channel skip information and determines on a channel-by-channel basis whether to skip the respective channels.

Abstract: A moving image converting apparatus includes: a block dividing unit executing block division processing for each of frames that form the moving image data; a motion amount detecting unit detecting a block motion amount corresponding to a motion amount of an object contained in each of blocks obtained; a line-of-sight motion amount detecting unit calculating a line-of-sight motion amount of a viewer who views the moving image data; a sampling point phase change amount determining unit receiving the block motion amount and the line-of-sight motion amount as input, and determining a block-corresponding sampling point phase change amount to be applied to spatial decimation processing on each of the blocks; and a decimation executing unit receiving the blocks obtained as input, executing the spatial decimation processing on the inputted blocks, the spatial decimation processing being executed by applying the block-corresponding sampling point phase change amount determined.

Abstract: The disclosure is directed to techniques for region-of-interest (ROI) video processing based on low-complexity automatic ROI detection within video frames of video sequences. The low-complexity automatic ROI detection may be based on characteristics of video sensors within video communication devices. In other cases, the low-complexity automatic ROI detection may be based on motion information for a video frame and a different video frame of the video sequence. The disclosed techniques include a video processing technique capable of tuning and enhancing video sensor calibration, camera processing, ROI detection, and ROI video processing within a video communication device based on characteristics of a specific video sensor. The disclosed techniques also include a sensor-based ROI detection technique that uses video sensor statistics and camera processing side-information to improve ROI detection accuracy.

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: In an image coding apparatus, a ROI setting unit sets a ROI region in an image. An entropy coding unit entropy-codes the image. A ROI information coding unit encodes information for specifying the ROI region. A codestream generator generates a codestream in a manner that the coded image and the coded information are explicitly included in the codestream. When a plurality of ROI regions are set in the image, the information may include a degree of priority.

Abstract: A method determines a predicted image area for an image area, in which a temporal predictor for the image area is determined based on a reconstructed image that precedes said image, a local predictor for the image area is determined within a reconstructed image area of the image, a margin of error between the image area and the image area predicted image area is determined by the local predictor using the reconstructed image area and by the temporal predictor using one of the preceding images. A predicted image area can be reconstructed. Also disclosed are an establishing device for carrying out the method for determining a predicted image area as well as a reconstructing device for carrying out the reconstruction method.

Abstract: Scalable Video Coding is recently attracting attentions due to its high flexibility. The current H.264/AVC scalable extension has adopted the Motion Compensated Temporal Filter (MCTF) framework to provide temporal scalability. In this paper, described is another fast motion estimation (ME) algorithm based on the MCTF framework. Simulation results show that the herein described algorithm can reduce the encoding complexity significantly while maintaining similar bit rate and PSNR, comparing with existing fast ME algorithms implemented in the reference software.

Abstract: An image processing device (10) removes noise from an input image by adding the input image to a reference image, and includes: a motion estimation unit (1) which generates motion information for each of blocks of the input image; a motion compensation unit (3) which generates a motion-compensated reference image; a block distortion amount calculation unit (5) which calculates a block distortion amount for each of blocks of the motion-compensated reference image; an addition ratio calculation unit (6) which calculates an addition ratio of a block of the motion-compensated reference image so that the addition ratio decreases as the block distortion increases; and an addition unit (4) which perform weighted sum on a pixel value of the block of the motion-compensated reference image and a pixel value of a block of the input image, using the addition ratio.

Abstract: An image processing apparatus includes a unit configured to calculate a sum of absolute differences between pixel values of pixels in a block of interest and pixel values of pixels in a reference block, a unit configured to compare sums of absolute differences calculated and to determine a minimum sum of absolute differences thereamong, a unit configured to estimate a motion vector for the pixel of interest using the reference pixel having the minimum sum of absolute differences and the pixel of interest, a unit configured to generate a motion-compensated image by using pixel values of the pixels as pixel values of corresponding pixels in the motion-compensated image, a unit configured to cumulatively add sums of absolute differences, and a unit configured to generate an interpolation image by mixing the pixels and corresponding pixels in the motion-compensated image according to an obtained cumulative-sum result.

Abstract: In an image acquisition apparatus, RAW data obtained in image acquisition with an imager in which a color filter is disposed in a front face is divided into each identical color by a RAW data dividing unit, a compression coding processing unit produces coded data of each color by compressing the RAW data at a compression rate controlled for each color, and the coded data of each color is recorded in a recording unit. A motion compensation unit compensates a relative positional relationship between frames by estimating a motion of a subject between frames of plural images, using specific color division coded data which is decoded by the compression coding processing unit and a RAW format data reproducing unit, the specific color division coded data being coded data of a color close to a peak of spectral sensitivity of human vision in the coded data of each color.

Abstract: To provide an image processing apparatus which reduces, using a temporal noise removal scheme, image block distortion or noise and temporal flickering and fluctuation phenomena. The image processing apparatus (1) removing noise in an input image by adding a reference image to the input image includes a motion estimation unit (20) which generates motion information indicating a motion estimated for the input image; a motion compensation unit (30) which generates a after motion compensation second image corresponding to a before motion compensation first image by performing motion compensation using the generated motion information; a flatness degree calculation unit (40) which calculates a flatness degree of the input image; a blending rate calculation unit (53) which calculates addition rates for the input image, first image, and second image, using the calculated flatness degree; and a pixel addition unit (60) which adds the respective images according to the addition rates.

Abstract: A method of compressed picture reconstruction using a plurality of post-processed reference pictures. The method generally includes the steps of (A) generating a first of a plurality of reconstructed pictures by decoding a first of a plurality of compressed pictures using at least one of a plurality of non-post-processed reference pictures buffered in a reference memory, wherein the compressed pictures are received in an input bitstream, (B) generating a first of a plurality of processed pictures by artifact processing the first reconstructed picture to remove artifacts, (C) buffering in the reference memory both (i) the first reconstructed picture as one of the non-post-processed reference pictures and (ii) the first processed picture as one of the post-processed reference pictures and (D) generating a second of the reconstructed pictures by decoding a second of the compressed pictures using at least one of the post-processed reference pictures buffered in the reference memory.

Abstract: An optimal interpolation frame is achieved by estimating optimal motion vectors both for a coding process and an interpolation frame generating process. A video codec apparatus coding and decoding a video and generating an interpolation frame in the decoding includes a motion vector estimation unit calculating an evaluation value to evaluate a correlation between images using a calculating formula and estimates a motion vector using the evaluation value, a motion vector coding unit coding video data using the motion vector and outputs a coded stream, video decoding unit decoding the coded stream and outputs video frames generated by the decoding, a frame interpolation unit generating interpolation frame using the motion vector, a cost calculating formula selecting unit selecting the calculation method depending upon whether the coding by the motion vector coding unit or the generating an interpolation frame by the frame interpolation unit is being executed.

Abstract: The present invention relate generally to digital watermarking. One claim recites an apparatus comprising: electronic memory for buffering a color video signal comprising a first frame and a second frame; and one or more electronic processors programmed for: transforming a first color channel and a second color channel of the first frame by hiding a digital watermark in the first color channel such that the first digital watermark includes a first signal polarity, and hiding the digital watermark in the second color channel such that the digital watermark includes a second signal polarity that is inversely related to the first signal polarity; transforming a first color channel and a second color channel of the second frame by hiding a digital watermark in the first color channel such that the first digital watermark includes the second signal polarity, and hiding the digital watermark in the second color channel such that the digital watermark includes the first signal polarity.

Abstract: The present invention relates to a video decoder for decoding a bit stream corresponding to pictures of a video signal. The invention is such that, motion vectors having values coded at a nominal resolution in the bit stream, said video decoder includes a rounding unit for rounding decoded nominal motion vectors to a resolution different from the nominal resolution in a way that minimizes the error accumulation along prediction path by assigning to rounded successive motion vectors along the prediction path, successive values such that their average value is equal to the value of the nominal motion vector.

Abstract: A method for context-modeling coding information of a video signal for compressing or decompressing the coding information is provided. An initial value of a function for probability coding of coding information of a video signal of an enhanced layer is determined based on coding information of a video signal of a base layer.

Abstract: An image processing method includes generating a block matching result by performing a block matching operation according to a plurality of image blocks in the horizontal direction without referring to an image block in the vertical or other directions and performing an image processing operation according to the block matching result of the block matching operation.

Abstract: A picture coding method for preventing occurrence of a malfunction includes: an adding step (Step S203) of coding a memory management command for the first picture of a predetermined picture number and adding the coded command to the second picture that is different from the first picture; and a re-adding step (Steps S205 to S208) of coding the memory management command again and re-adding re-coded command to the section other than the second picture in a coded picture stream VSt. In the re-adding step, re-addition of the memory management command is prohibited when the first picture is not immediately previous in coding order among the pictures of the predetermined picture number included in the coded picture stream VSt.

Abstract: A moving picture coding apparatus for performing inter-picture predictive coding on pictures constituting a moving picture is provided with a coding unit for performing predictive error coding on image data; a decoding unit for performing predictive error decoding on an output from the coding unit; a reference picture memory for holding output data from the decoding unit; and a motion vector detection unit for detecting forward and backward motion vectors on the basis of decoded image data stored in the memory. When coding a B picture, a picture that is timewise closest to the target picture is used as a candidate picture for forward reference, and an I or P picture that is timewise closest to the target picture is used as a candidate picture for backward reference. The moving picture coding apparatus so constructed can improve coding efficiency of a B picture to be subjected to bidirectional predictive coding.

Abstract: A motion estimation method is provided for processing successive images in an image sequence, with a motion vector being associated with each of the processed images. For a current image, motion vectors associated with images that precede the current image in the sequence are selected. Candidate motion vectors are generated from the motion vectors that are selected. A motion vector is elected from among the candidate motion vectors. Information that associates the elected motion vector with the current image is stored in memory. At least one of candidate motion vectors is an acceleration vector generated from the acceleration between first and second motion vectors averaged relative to a first and second images, with the first and second images being distinct and preceding the current image in the image sequence. A motion vector averaged relative to a given image is obtained from selected motion vectors associated with images preceding the given image.

Abstract: An interpolation frame generation device that generates an interpolation frame that interpolates image frames that are obtained by decoding a coded image signal that is coded by motion compensation, includes a motion vector deriving unit and an interpolation frame generating unit. The motion vector deriving unit acquires a motion compensation vector of a coded block that forms the coded image signal. The interpolation frame generating unit generates the interpolation frame in accordance with the motion vector of the image block that forms an image frame by using the motion compensation vector of the coded block as the motion vector of the image block.

Abstract: Provided is an adaptive motion search range determining apparatus and method for encoding UHD-class high-resolution images. The adaptive motion search range determining apparatus includes an MVD average/standard deviation calculation unit calculating an value average and a standard deviation of MVDs of neighboring macroblocks of a current macroblock, and a motion search range determination unit determining a motion search range of the current macroblock using the value average and the standard deviation. According to the adaptive motion search range determining apparatus, it is possible to enable each macroblock to have an adaptive search range by variably adjusting a motion vector search range of a current macroblock with reference to motion vectors of neighboring macroblocks of the current macroblock.

Abstract: A method and apparatus for an adaptive systolic array structure is initially configured for motion estimation calculations and optionally reconfigured as the motion estimation algorithm progresses. A scheduling map of the processing element (PE) calculations for a given motion estimation algorithm is generated. A systolic array structure may then be generated from the scheduling map, whereby the size and shape of a processing element array is configured to generate the search pattern that is to be used during the search. In addition, delay elements may be implemented within the systolic array structure, so as to preserve the pixels of a current macroblock that are reused in accordance with the scheduling map. The systolic array structure may also be adapted by the motion estimation algorithm during subsequent search stages to accommodate refinements required by the search strategy.

Abstract: The invention provides a method of encoding a multi-view video and an encoder capable of calculating global disparity between pictures at different viewpoints in the multi-view video and forming a hierarchical B picture in both view and time directions on the basis of the calculated global disparity, thereby improving encoding efficiency. According to an embodiment of the invention, there is provided an encoder for encoding data for pictures captured at different viewpoints in a multi-view video. When a picture at one of the first and last viewpoints is encoded into an I picture, the encoder encodes a picture that is captured at the other viewpoint but is disposed in the same time direction into an VP picture.

Abstract: According to one embodiment, an interpolated frame generator comprises first detector detects potential motion vector by block matching between input frame images, using first block of fixed size, second detector detects motion vector by block matching between the input frame images, using second block having fixed size larger than that of the first block, and generator generates interpolated frame by using the potential motion vector when first detector detects only one potential motion vector, and generates interpolated frame by using a potential motion vector closest to the motion vector detected by the second detector when first detector detects a plurality of motion vectors, wherein first detector includes extractor compares SAD of motion vector with SAD of motion vectors adjacent to the motion vector, and extracts potential motion vector having SAD smaller than any of the SAD of adjacent motion vectors, as a potential motion vector used in generator.

Abstract: An image processing apparatus includes the following elements. An obtaining unit obtains a first still image, a first difficulty level indicating a complexity of the first still image, a second still image, and a second difficulty level indicating a complexity of the second still image. A combining unit combines the first and second still images at a combination ratio which is changed along a time axis to generate a moving image. An encoding unit encodes the moving image. A moving-image difficulty-level calculating unit calculates a moving-image difficulty level indicating a complexity of the moving image according to the first and second difficulty levels and the combination ratio changed along the time axis. A control unit controls a characteristic of a process of encoding the moving image according to the combination ratio changed along the time axis and the moving-image difficulty level.

Abstract: The present invention relates to a method and apparatus for encoding a multi-view video and a method and apparatus for decoding a multi-view video, and more particularly, to a method and apparatus for encoding a multi-view video and a method and apparatus for decoding a multi-view video that can perform encoding and decoding in consideration of global disparity between pictures captured at two different viewpoints to remove spatial redundancy. An embodiment of the invention provides an encoder for compressing data for pictures captured at different viewpoints using global disparity in a multi-view video. The encoder includes: an input unit that receives pictures captured at two or more viewpoints; a reference picture generating unit that generates a new reference picture on the basis of the global disparity between a target picture and a reference picture; and a motion predicting unit that calculates motion vectors of the target picture on the basis of the reference picture.

Abstract: A method for controlling an ambient lighting element including receiving a content signal, analyzing the content signal to determine a motion vector of an object (120A) depicted in the content signal, presenting the content signal on a display device, and adjusting an ambient lighting effect provided by the ambient lighting element as determined by the motion vector The presented content signal may be portioned into macro-blocks (110A) and sub-blocks (230). A motion vector of each sub-block (230) may be resolved into components that are parallel and perpendicular to an outside edge of the display device. An average color of each sub-block (230) depicted within a macro-block (110A) in proximity to an outside edge of the display device may be weighted by the motion vector of the corresponding sub-block (230) for determining an average color of the macro-block (110A). The average color of the macro-block (110A) may be used to adjust the ambient lighting element.

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 system, method, and computer program product are provided for refining motion vectors. In operation, a plurality of motion vectors associated with a current frame and a first resolution are created. Furthermore, the motion vectors are refined utilizing information including at least one of first information describing motion vectors associated with a previous frame and second information describing motion vectors associated with the current frame and a second resolution.

Abstract: Multi-resolution images of a reference image and a target image are generated. Then, whole-range matching is performed on an image of a lower resolution to detect a two-dimensional displacement between the images. Block matching is performed on an image of a higher resolution to detect a displacement at each feature point. The accuracy of motion data is increased by correcting the motion data with an image of a higher resolution by using the previously calculated motion data of the lowest resolution through higher resolutions as an initial value.

Abstract: A method and apparatus for encoding and decoding multi-view images. The multi-view image encoding method selects a block corresponding to a current block from another picture having a view-point which is different from a view-point of a current picture to which the current block belongs, on the basis of a global disparity vector representing a global disparity between the current picture and the other picture; and encodes the current block on the basis of block information of a block from among the selected block and blocks adjacent to the selected block. Accordingly, multi-view images can be encoded in consideration of the individual differences between the appearances of objects as well as global disparities between view-points.

Abstract: An image processing apparatus which performs processing on an input moving image including a plurality of access units arranged every first period, includes a motion vector calculation unit which calculates a motion vector of an object included in the input moving image every second period, a motion vector conversion unit which, converts the motion vector by multiplying the calculated motion vector by a predetermined gain, a frame compensation unit which generates an output moving image including a plurality of access units arranged every third period by performing frame compensation processing of converting or compensating for the access units on the input moving image so that the object moves in accordance with the motion vector which has been subjected to the conversion, and a gain calculation unit which calculates the gain in accordance with brightness in a user environment and supplies the obtained gain to the motion vector conversion unit.

Abstract: The present invention relates to an image processing device, method, and program, enabling processing efficiency to be improved. In the event that a block B1 of a sub macro block SMB0 is a block which is the object of prediction, pixel values of a decoded image are set to be used for pixel values of adjacent pixels included in an upper region U and an upper left region LU, out of templates adjacent to the block B1 of the sub macro block SMB0. On the other hand, pixel values of a prediction image are set to be used for pixel values of adjacent pixels included in left region L, out of templates adjacent to the block B1 of the sub macro block SMB0. The present invention can be applied to an image encoding device performing encoding with the H.264/AVC format, for example.

Abstract: Data of both MPEG-2 and MPEG-4 is generated simultaneously with a small circuit scale and a small power consumption. A moving picture encoding apparatus for encoding a moving picture through motion-compensated inter-frame prediction has: a MPEG-2 encoding unit including a motion vector estimator, a frame memory, a forward prediction circuit, a bidirectional prediction circuit, a prediction selection circuit, an intra-frame encoding circuit and a local decoding circuit; a MPEG-4 encoding unit including a frame extraction circuit for extracting a predetermined MPEG-2 frame and a transcoder for encoding the extracted frame; a motion vector calculator calculating a motion vector to be used for MPEG-4 prediction from a motion vector to be used for MPEG-2 prediction; and a prediction mode controller controlling the prediction mode of the MPEG-2 encoding unit in such that the MPEG-2 prediction mode becomes coincident with the MPEG-4 prediction mode.

Abstract: This disclosure describes apparatus and systems for encoding and decoding signals from a host signal such as audio, video or imagery. One claim recites an apparatus comprising: electronic memory for storing a media signal representing audio or video; and an electronic processor. The electronic processor is programmed for: extracting data representing at least some features of the media signal; using the extracted data as a key to select coefficients in a transform domain, the coefficients representing the media signal; and modifying selected coefficients to hide a steganographic signal in the media signal. Of course, other claims and combinations are provided as well.

Abstract: Systems and methods for a capsule camera having on-board storage or wireless transmission to control image capture using motion information and to encode captured images based on motion compensated video compression are disclosed. In order to conserve the precious storage and power resources, the capsule camera measures motion metric to select a skip mode or a capture mode. In order to reduce computations related motion metric derivation, the motion metric is based on sub-images corresponding to a current frame and a previous captured frame instead of full-size images. When the skip mode is selected, the capsule camera exercises storage space/power conservation by skipping unnecessary image capture and/or reducing luminous energy of a light source or other means for power conservation. When the capture mode is selected, a full-size full-resolution frame is captured and compressed using motion-compensated video compression to achieve high compression ratio.

Abstract: A process performs a search on a base reference frame in a video signal to determine a first motion vector for each of the possible partitions of a current macroblock. The first motion vector for a partition provides the most optimal match for the partition in the base reference frame. An optimal partition for the current macroblock, which gives a minimum motion estimation cost among all the possible partitions in the base reference frame, is determined. The process performs the search on the optimal partition over a non-base reference frame in the video signal to determine a second motion vector that provides the most optimal match for the corresponding partition in the non-base reference frame. A simplified search is performed on each partition other than the optimal partition over the non-base reference frame to determine the second motion vector for the corresponding partition.

Abstract: Provided is a method of and apparatus for compressing/decompressing data, in which lossless compression is applied to a raw data transfer mode in order to decompress data into high-quality data in a encoder/decoder (CODEC) system. A method of encoding data includes performing intraprediction within a current frame and performing interprediction between the current frame and a reference frame, applying lossless compression in input data in a raw data transfer mode in which raw data is transferred within an input frame, and comparing a predetermined parameter generated by the intraprediction and the interprediction with a predetermined parameter generated by the lossless compression, in order to select one of a prediction mode and a lossless compression mode.

Abstract: A feature vector is encoded into a sparse binary vector. The feature vector is retrieved, for example from storage or a feature vector generator. The feature vector represents a media object or other data object. One or more permutations are generated, the dimensionality of the generated permutations equivalent to the dimensionality of the feature vector. The permutations may be generated randomly or formulaically. The feature vector is permuted with the one or more permutations, creating one or more permuted feature vectors. The permuted feature vectors are truncated according to a selected window size. The indexes representing the maximum values of the permuted feature vectors are identified and encoded using one-hot encoding, producing one or more sparse binary vectors. The sparse binary vectors may be concatenated into a single sparse binary vector and stored. The sparse binary vector may be used in the similarity search, indexing or categorization of media objects.

Abstract: A quantizing matrix selecting part 3 for calculating the average and variance of brightness values in a prediction image created by a motion-compensated prediction unit 1, and selecting a quantizing matrix corresponding to the average and variance of brightness values in the prediction image from among a plurality of quantizing matrices which are prepared in advance is disposed, and a quantizing part 6 quantizes orthogonal transformation coefficients outputted from an orthogonal transformation part 5 with reference to the quantizing matrix selected by the quantizing matrix selecting part 3.

Abstract: A method for transcoding from an MPEG-2 format to an H.264 format is disclosed. The method generally comprises the steps of (A) decoding an input video stream in the MPEG-2 format to generate a plurality of macroblocks; (B) determining a plurality of indicators from a pair of the macroblocks, the pair of the macroblocks being vertically adjoining; and (C) coding the pair of the macroblocks into an output video stream in the H.264 format using one of (i) a field mode coding and (ii) a frame mode coding as determined from the indicators.

Abstract: The present invention is directed to an image information encoding apparatus, used in receiving compressed image information through network media when processing of such compressed image information is performed on storage media. A picture sorting buffer delivers information of picture type of frame Picture_type to a picture type discrimination unit. The picture type discrimination unit transmits command to a motion prediction/compensation unit on the basis of that information. The motion prediction/compensation unit generates predictive picture by using filter coefficients having the number of taps lesser than that of P picture with respect to B picture for which operation quantity and the number of memory accesses are required to more degree as compared to P picture on the basis of that command.

Abstract: Methods and apparatuses for encoding and decoding a multi-view image are provided. The method of encoding the multi-view image can use a difference in views between pictures of the multi-view image, add an image area obtained from a picture at one view at a first time to a picture at another view at a second time, thereby generating a reference picture, and perform prediction encoding using the generated reference picture, thereby increasing prediction encoding efficiency.