Abstract: A method of processing a digital television (DTV) signal is disclosed. Herein, the DTV signal is generated by performing Reed-Solomon (RS) encoding on additional data, multiplexing the RS-encoded additional data with main data, RS encoding the multiplexed additional and main data, interleaving the RS-encoded additional and main data, trellis encoding the interleaved additional and main data, and transmitting a Radio Frequency (RF) DTV signal including the trellis-encoded additional and main data. The method to process the DTV signal includes receiving the DTV signal including the additional data multiplexed with the main data through an antenna, in which signaling information is periodically inserted in the additional data. The received DTV signal is demodulated including performing channel equalization on the demodulated DTV signal. Trellis decoding is performed on the channel-equalized DTV signal.

Abstract: There is provided an addition ratio learning apparatus including a noise adding unit that adds noises to data of an image input as a teacher image, a motion compensating unit that sets an image where time addition noise reduction processing is executed as an NR screen and performs motion compensation with respect to the NR screen, a differential feature amount calculating unit that sets an image as an input screen and calculates a differential feature amount, a circulation history specifying unit that counts a circulation history in the time addition noise reduction processing and specifies the circulation history, an addition ratio computing unit that computes an addition ratio on the basis of pixel values, and a time adding unit that performs multiplication by a coefficient determined according to the computed addition ratio to perform weighted addition and executes the time addition noise reduction processing with respect to the input screen.

Abstract: A system that incorporates teachings of the present disclosure may include, for example, obtaining regression coefficients that quantify a relationship between premises feedback and first network and premises performance indicators, obtaining second network performance indicators for the network elements, obtaining second premises performance indicators for the customer premises equipment, and predicting customer complaints by applying the obtained regression coefficients to at least the second network performance indicators and the second premises performance indicators. Other embodiments are disclosed.

Abstract: A method and device for tracking error propagation and refreshing a video stream is provided. The proposed subject matter comprises of an error propagation tracking method that works in the sub-sampled domain to reduce computational cycles and memory bandwidth. Further, the tracking based update of the error propagation metric is done differently for static and non-static regions to avoid unnecessary refresh of static areas. Through suitable thresholding of the metric at a macroblock (MB) level, a set of refresh MBs are selected for each frame. These refresh MBs are coded either as an intra MB or as an inter MB that is predicted from one or more reliable reference frames (—frames that are known to be available at the decoder with negligible errors—). Such inter coding of refresh MBs improves the compression efficiency when compared to pure intra coding of refresh MBs.

Abstract: A method of transmitting a broadcast signal includes performing Reed-Solomon (RS) frame encoding and Cyclic Redundancy Check (CRC) encoding on first mobile service data to form a primary RS frame and on second mobile service data to form a secondary RS frame; encoding on at least the first mobile service data or the second mobile service data, in serial concatenated convolution code (SCCC) block units; encoding signaling information including transmission parameters, the transmission parameters including SCCC encoding information and RS frame encoding information; formatting a data group including the encoded first mobile service data and second mobile service data, wherein the first mobile service data are included in a first region within the data group and the second mobile service data are included in a second region within the data group, the second region being different from the first region; and transmitting the broadcast signal including the formatted data group.

Abstract: The minimum Sum of Absolute Differences obtained by a motion vector search roughly judges the magnitude of quantization error by whether or not exceeding a predetermined threshold value. When the quantization error is lower, whether or not visually noticeable noise exists in some of the pixels of the current macroblock is judged based on the amount of flatness and noise detected in each of the 4×4 pixel blocks of the current macroblock partitioned into 16 sub-macroblocks. If there is visually noticeable noise, intra-frame coding is selected. When the quantization error is higher, whether or not visually noticeable noise exists in the current macroblock is judged while considering the magnitude of the motion vector. If there is visually noticeable noise, intra-frame coding is selected.

Abstract: The reception method is implemented in a receiver apparatus receiving images of a video sequence having undergone losses during an exchange of data with a sender apparatus connected to said receiver apparatus via a communication network. The receiver apparatus comprises at least a first module implementing a transport layer and a second module implementing an application layer, the application layer being subdivided into a first sub-module implementing the decoding of the video and a second sub-module implementing the display of said video. The method comprises the following steps: in the transport layer, transmitting a data loss detection signal intended for the application layer, as soon as the transport layer detects a loss of data; and in the application layer, implementing a loss concealment mechanism, on reception of said loss detection signal and without waiting for said application layer itself to detect said data loss so detected.

Abstract: A method of noise filtering of a digital video sequence is provided that includes computing a motion image for a frame, wherein the motion image includes a motion value for each pixel in the frame, and wherein the motion values are computed as differences between pixel values in a luminance component of the frame and corresponding pixel values in a luminance component of a reference frame, applying a first spatial noise filter to the motion image to obtain a final motion image, computing a blending factor image for the frame, wherein the blending factor image includes a blending factor for each pixel in the frame, and wherein the blending factors are computed based on corresponding motion values in the final motion image, generating a filtered frame, wherein the blending factors are applied to corresponding pixel values in the reference frame and the frame, and outputting the filtered frame.

Abstract: A method for compressing video operates by first deleting one or more video packets from binary train to provide a reduced binary train. Thereafter, error masking is performed on the reduced binary train to yield a corrected degraded local decoded image which is the image reconstructed on the basis of the reduced binary train and of the masking algorithm. The reduction in the binary train is validated as a function of the quality of the corrected degraded local decoded image. Lastly one of the reduced binary train or of the binary train, is selected depending on whether or not there is validation.

Abstract: In some embodiments, a video decoder is capable of recovering skipped video data (e.g. motion vectors, prediction modes, quantization parameters, selected frequency-domain coefficients such as DC and 5-lowest-frequency AC coefficients, and/or entire video data blocks) using multiple alternative recovery modes such as spatial interpolation, temporal interpolation, and motion search. To decide whether to skip a particular data type for a block, the encoder evaluates the effect of skipping the data on rate and distortion by simulating the decoder data recovery using the multiple recovery modes. The encoder transmits indicators of skipped data types and associated recovery modes, if different from decoder baselines. The skipped data and recovery mode indicators may be included in macroblock and/or slice headers, and/or as part of sequence, group-of-picture (GOP), or picture coding parameter data. Skipped data recovery decoder resources (e.g.

Abstract: A media processor having a controller operable to recognize a portion of a video stream in an interactive television network having video compression artifacts corresponding to a stored model and perform model-based video correction of the portion recognized using synthetically generated images of objects in a captured video scene. Other embodiments are disclosed.

Abstract: A digital television system performing modulation/demodulation by VSB (vestigial side band) is provided. The invention includes a VSB transmitter including an additional error correction encoder designed such that a signal mapping of a TCM encoder is considered, a multiplexer (MUX), a TCM encoder operating in correspondence with state transition processes of the additional error correction encoder, and a signal transmission part including an RF converter. The invention further includes a VSB receiver including a signal receiver part receiving a signal transmitted from the transmitter, a TCM decoder, a signal processing part including a derandomizer, and an additional error correction decoder part.

Abstract: An apparatus for facilitating reception of multiple representations of a video signal. In one embodiment, the apparatus includes a mechanism for receiving plural representations of the video signal corresponding to plural decimated versions of the video signal, associating pictures of the received plural representations of the video signal, and outputting pictures corresponding to information from associated pictures in accordance with a relative temporal order.

Abstract: Systems and methods for error resilient transmission and for random access in video communication systems are provided. The video communication systems are based on single-layer, scalable video, or simulcast video coding with temporal scalability, which may be used in video communication systems. A set of video frames or pictures in a video signal transmission is designated for reliable or guaranteed delivery to receivers using secure or high reliability links, or by retransmission techniques. The reliably-delivered video frames are used as reference pictures for resynchronization of receivers with the transmitted video signal after error incidence and for random access.

Abstract: Systems and methods of transcoding video bitstreams that employ look-ahead approaches to enhance the overall perceptual quality of transcoded video information, communications, and entertainment delivered to an end user. The disclosed systems and methods of transcoding video bitstreams take into account the scene characteristics and the local coding complexity of video frames in a video sequence before performing bit allocations for the video frames, thereby significantly improving the perceptual quality of transcoded video delivered to the end user.

Abstract: An image processing apparatus according to the present invention includes an encoding means for encoding image data on a tile-by-tile basis, each tile being constituted by a plurality of blocks, and each block having a predetermined size, and a storing means for determining whether or not each tile includes a row or column of pixels designated to be deleted in the image data, and if the tile includes a row or column of pixels designated to be deleted, storing deletion information that indicates a corresponding row or column of pixels designated to be deleted in the tile, in header information of the tile that has been encoded.

Abstract: Systems and methods of objective video quality measurement based on support vector machines. The video quality measurement systems can obtain information pertaining to features of a target training video, obtain corresponding information pertaining to features of a reference version of the target training video, and employ the target training features and/or the reference training features to build video quality models using such support vector machines. Based on the target training features and/or the reference training features used to build such video quality models, the video quality models can be made to conform more closely to the human visual system. Moreover, using such video quality models in conjunction with target features of a target video whose perceptual quality is to be measured, and/or reference features of a reference video, the video quality measurement systems can be employed to predict measurements of the perceptual quality of such a target video with increased accuracy.

Abstract: An apparatus and method are provided for compensating a block tutor in an image frame. This may include a video codec decoder for decoding an inputted image frame, and outputting a decoded image frame. An error concealment block may detect an error-generated block in the decoded image frame and compensate the detected error block through a median filter, and output the compensated image frame.

Abstract: A method for estimating video quality on bit-stream level, wherein the video quality refers to a video after error concealment and the method is performed on bit-stream level before said error concealment, comprises extracting and/or calculating a plurality of global condition features from a video bit-stream, extracting and/or calculating a plurality of local effectiveness features at least for a lost MB, calculating a numeric error concealment effectiveness level for each (or at least for each lost) MB by emulating an error concealment method that is used in said error concealment, and providing the calculated error concealment effectiveness level as an estimated visible artifacts level of video quality.

Abstract: A method and apparatus to decode audio data constructed with a plurality of layers. An error concealment method of process a decoded bitstream selects one of a frequency domain and a time domain in order to conceal the errors, detects a position where the errors exist in a frame when the error concealment method in the frequency domain is selected, and conceals the errors only in a segment after the detected position.

Abstract: A method of transmitting a high definition (HD) image and an ultra high definition (UHD) image, the method including obtaining a first HD image by down-scaling an original UHD image, obtaining at least one of a second HD image and a half-resolution UHD image by down-scaling the original UHD image, encoding the first HD image, encoding at least one of the second HD image and the half-resolution UHD image, generating a first transmission stream by multiplexing and synchronizing the encoded first HD image, generating a second transmission stream by multiplexing and synchronizing at least one of the encoded second HD image and the encoded half-resolution UHD image, transmitting the first transmission stream via a first transmission channel, and transmitting the second transmission stream via a second transmission channel differing from the first transmission channel is provided.

Abstract: A receiving apparatus comprises a receiver for receiving a video signal, an output for outputting the video signal to a screen, storage device or further terminal; and a concealment module for applying a concealment algorithm to a lost area of the video signal. The concealment module is configured to determine an estimate of concealment quality selectively directed toward a region of interest within the area in question, and based on this estimate to determine whether or not to apply the concealment algorithm.

Abstract: An auxiliary information map (10) is upsampled to form an upsampled auxiliary information map (20). Multiple reference pixels (23) in the upsampled auxiliary information map (20) are selected for a current pixel (21) in the upsampled auxiliary information map (20) based on texel values of texels in an associated texture (30). An updated pixel value is calculated for the current pixel (21) based on the pixel values of the selected reference pixels (23).

Abstract: A video transport system is provided for transporting as 8B/10B coded video stream across a 64B/66B coded link, wherein forward error correction is provided without the overhead of the prior art. The system also provides a system with the ability to recover 64B/66B Encoded blocks that have corrupt Sync bits.

Abstract: An embodiment improves the operation of a H.264 and Joint Scalable Video Codec (e.g., JSVC/H.264 Amendment 3) video decoder by managing neighboring block data during the decoding process. An embodiment pre-computes neighboring block tables to efficiently locate the neighboring block data required to decode a current macroblock. In particular, the pre-computed most probable joint neighboring block tables disclosed herein handle both macroblock adaptive frame field (MBAFF) coding and non-MBAFF coding. An embodiment is further capable of managing variable block sizes. Other embodiments are described and claimed.

Abstract: A method includes decompressing first compressed video data to provide uncompressed video data in a first order. The method includes compressing the uncompressed video data to provide second compressed video data in a second order. The decompressing and compressing are based on first and second compression rates, respectively. The first order may be based on fundamental blocks of a frame of video data and the second order is based on lines of the frame of video data. The compressing may include alternating compression of partial portions of a first line of uncompressed video data with compression of partial portions of at least a second line of uncompressed video data to thereby generate a first line of compressed video data corresponding to the first line of uncompressed video data and at least a second line of compressed video data corresponding to the second line of uncompressed video data.

Abstract: Apparatuses and methods for improving coding processes and coding parameters for coding video data are provided for. A coder may select coding parameters for video data according to a default coding policy. The default coding policy may include selection of prediction modes (e.g., intra-coding or inter-coding) for each pixel group in each frame. A video coder may select some pixel groups in a frame to be coded as refresh pixel groups as an exception to the default assignment policies. The selection of refresh pixel groups may be based on prediction relationships among multiple frames of source video data. The default coding of the refresh pixel groups is then modified to enhanced the coding of the refresh pixel groups. The refresh pixel groups may permit fewer intra (I) frames be sent and/or may improve the quality of the recovered video.

Abstract: A system for encoding a video stream into a processed video signal that includes at least one image. The system includes a downscaling module, a partitioning module, a rate control module, and an encoder section. The downscaling module receives the video stream and produces a downscaled video stream. A partitioning module, including a region detection module, receives the downscaled video stream and detects a pattern of interest in the at least one image. The partitioning module is operable to partition the at least one image based on the detected pattern of interest. The rate control module that receives an output from the partitioning module and produces an encoder control signal dependent on the output from the partitioning module. The encoder section, coupled to the rate control module, receives the video stream and generates the processed video signal.

Abstract: A method is provided for adapting the throughput of a video stream comprising a plurality N of compressed images with a plurality of spatial resolutions and/or a plurality of quality layers, the images being compressed separately from one another.

Abstract: The present invention provides a data processing method, the data processing method including receiving a broadcasting signal where mobile service data are multiplexed with main service data, extracting transmission-parameter-channel signaling information and fast-information-channel signaling information from a data group within the received mobile service data, parsing first program table information describing virtual channel information of an ensemble and a service provided by the ensemble using the fast-information-channel signaling information, the ensemble corresponding to a virtual channel group of the received mobile service data, obtaining information representing that second program table information, which includes a descriptor describing event information of the mobile service data, is received, from the first program table information, parsing the second program table information, and storing the event information and displaying service guide information including the event information.

Abstract: A sequence of digital images encoded according to a predictive format is decoded using encoded auxiliary data which is representative of at least part of the sequence of digital images. A current encoded image, received with at least one detected error, is decoded. The decoding includes applying an error concealment decoding on at least part of the current encoded image containing such a detected error to obtain a first decoded image. An item of information representative of reconstruction errors is obtained based upon a reconstruction confidence of the error concealment decoding for at least one pixel of the error-containing part of the current image. The encoded auxiliary data and the first decoded image are processed based upon the item of information representative of reconstruction errors to obtain corrected auxiliary data. The first decoded image is modified to obtain a second decoded image based upon the corrected auxiliary data.

Abstract: A method, comprises monitoring a encoding process of a source video file performed by an encoder; obtaining an encoding decision parameter used to encode a picture of the source video file during the encoding process; comparing the encoding decision parameter to a threshold; based on the step of comparing, identifying the picture as a candidate picture for a visual defect or coding error; and storing a timestamp of the candidate picture.

Abstract: In an image processing device, the processor performs: (a) setting, as target pixel data, each set of pixel data; (b) selecting, from among the plurality of sets of pixel data, a plurality of first sets of pixel data corresponding to a plurality of first pixels positioned within a prescribed range of the image; (c) determining whether the plurality of first sets of pixel data represents a white background; (d) determining whether a variation in color difference components among the plurality of first sets of pixel data satisfies a prescribed condition; and (e) determining that target pixel data includes mosquito noise when the plurality of first sets of pixel data represents the white background and when the variation in the color difference component among the plurality of first sets of pixel data satisfies the prescribed condition. The prescribed range includes a pixel corresponding to the target pixel data.

Abstract: The present invention relates to an encoding device and a method, a decoding device and a method, an editing device and a method, a storage medium, and a program which can perform encoding and decoding so that buffer failure does not occur. Information, such as a minimum bit rate, a minimum buffer size, and a minimum initial delay time, is contained in a random access point header contained in an accessible point in a bitstream. A bitstream analyzing unit 72 analyzes an input bitstream, sets the above-mentioned information, and outputs the resulting information to a buffer-information adding unit 73. The buffer-information adding unit 73 adds the input information to the input bitstream and outputs the resulting bitstream. The present invention is applicable to an encoding device and a decoding device which process bitstreams.

Abstract: A method of processing video data includes providing a bit-stream; and decoding the bit-stream. The step of decoding the bit-stream includes extracting a data portion from the bit-stream; extracting a first backup copy of the data portion from the bit-stream; and determining correct values of syntaxes in the data portion from the data portion itself and the first backup copy of the data portion. A second backup copy of the data portion may be extracted from the bit-stream and cross-examined with the data portion and the first backup copy of the data portion.

Abstract: One technique for implementing error control in response to a video data error comprises receiving an indication of a video data error from a video decoder, determining whether the video data error occurred on a reverse link of a wireless network between a video encoder and a network device, and applying error control in response to the video data error if the video data error did not occur on the reverse link. For example, the indication of the video data error may include a first sequence number (SN) of a packet comprising lost data, and determining whether the video data error occurred on the reverse link may include comparing first SN with a second SN of a packet associated with a most recent RL error.

Abstract: In a DTV transmitter the bits of shortened BCH codewords that exhibit undesirably low densities of ONEs are ONEs' complemented before being further coded, and used to modulate carrier waves. In a DTV receiver the further coding is decoded after demodulation. The results of such decoding are processed to recover successive shortened BCH codewords, some of which are in TRUE form and others of which have had their bits ONEs' complemented. Each shortened BCH codeword is extended to full length with ZEROs, and decoding is attempted. Successful decoding confirms that the shortened BCH codeword was received in TRUE form. If decoding is unsuccessful, the bits of the shortened BCH codeword as received are ONEs' complemented, extended to full length with ZEROs, and decoding is attempted. Successful decoding confirms that the shortened BCH codeword was received in ONEs' complemented form and has subsequently been converted to TRUE form.

Abstract: An image transmission device compresses and transmits image data to be sent and cannot correct errors occurring on the transmission path when image data to be transmitted is larger than a currently prescribed image size. In an image receiving device which receives compressed image data, data to be inputted is inputted by switching between first periods, during which the amount of data transmitted per prescribed time interval is a first data transmission amount, and second periods, during which the amount of data transmitted per prescribed time interval is less than the first data transmission amount. Compressed image data is inputted during the first periods, and error correction codes are inputted during the second periods, and a control unit expands the output of an error detection unit by means of an expansion unit and outputs the same.

Abstract: A noise estimator device of a video decoder includes a reception end, a low pass filter coupled to the reception end, a delay unit coupled to the reception end, a minimum level estimation unit coupled to the low pass filter for estimating a minimum of a specific number of low-pass results, a difference level estimation unit coupled to the low pass filter, the delay unit, and the minimum level estimation unit for determining a synchronization signal according to a plurality of low-pass results and the minimum of the specific number of the low-pass results, and for estimating a noise level according to the synchronization signal and signals outputted from the delay unit, and an output end coupled to the difference level estimation unit.

Abstract: In one embodiment, a method that includes receiving plural representations of a video signal, the video signal comprising plural sequenced pictures corresponding to at least a portion of a video program, wherein two or more of the plural representations of the video signal (PROTVS) includes a respective sequence of latticed pictures and one or more of the other PROTVS includes a respective sequence of non-latticed pictures; and providing in plural successive non-overlapping segments distribution interval (SDIs) compressed versions of the PROTVS in a single video stream, wherein each SDI consists of plural non-overlapping, consecutive segments, each of the plural non-overlapping consecutive segments originating from a respective one of the collective PROTVS.

Abstract: An adaptive Wiener filter may be applied to improve coding efficiency because of information lost during quantization of the video encoding process. The Wiener filter may be selectively applied globally to an entire picture or locally to portions of the picture. Histogram segmentation may be used to select pixels for Wiener filtering in some embodiments. The Wiener filter may be adaptively applied to histogram bins, improving coding efficiency in some cases.

Abstract: A method is described for efficiently determining total end-to-end distortion of a pre-compressed data stream, such as video streams or other media streams, at the time of delivery over a lossy-network, and for providing adaptive error-resilient delivery schemes based on distortion estimates. The methods can be utilized with single or multilayer packet streams and are particularly well suited for video streams. By way of example, distortion estimates are performed by generating side-information at the time of data stream compression, wherein the side-information is used in conjunction with information about the network status to determine an estimated distortion for the group of packets when the data stream is transported over the network to a destination end. This estimation may be utilized within described resiliency techniques in which the error correction mechanism is selected in response to the estimated distortion, which may be additionally refined in reference to cost factors.

Abstract: A decoding device includes: a determination unit that determines whether or not a decoding ending condition is satisfied at an interval shorter than an interval of one decoding process in repeated decoding and ends the process in the middle of the one decoding process in a case where the decoding ending condition is satisfied.

Abstract: The present invention discloses a method for media file compression, which includes: extracting the encoding parameters from an input media file, separating and decoding the audio and video stream from the input media file, and extracting an original audio stream and an original video stream; computing the transcoding parameters required for compression according to the encoding parameters; encoding the original audio stream to output a new compressed audio stream, and encoding the original video stream to output a new compressed video stream according to the transcoding parameters; synthesizing the new compressed audio stream and the new compressed video stream to create a new media file. The present invention also provides a system for media file compression.

Abstract: The subject matter of this specification can be implemented in, among other things, a method of decoding video data that includes decoding compressed input video data using a first entropy coding technique to obtain first decoded data. The method further includes re-encoding the first decoded data using a second entropy coding technique that is different from the first entropy coding technique to obtain re-encoded data. The method further includes storing the re-encoded data in a storage device. The method further includes decoding the re-encoded data using the second entropy coding technique to obtain second decoded data.

Abstract: The invention relates to a method of concealing errors attributed to missing frames in a Motion Picture Expert Group-2 video stream, including the steps of: receiving a new frame for decoding and a “frame missing” flag that is set to a value associated with the occurrence of a missing frame, parsing the new frame to recover a picture type and a frame structure of the new frame, retrieving a picture type and a frame structure of a previous frame decoded immediately prior to the new frame, assigning a picture type and a frame structure to the missing frame based on the picture type and frame-structure values of the previous frame and the new frame and then applying an error concealment technique based upon the assigned picture type and frame structure of the missing frame.

Abstract: A video image processing method for use in a terminal, including: detecting a packet loss rate in a current video image transmission; determining that the detected packet loss rate is higher than a predetermined threshold; and responsive to the determining, processing a first part of a video image corresponding to a target region based on a first code rate, and processing a second part of the video image corresponding to a non-target region based on a second code rate.

Abstract: A system and method for recovering from data loss are described including monitoring a first bit rate video data bit stream to determine if there is frame loss or damage, multiplexing decoding parameters of a second bit rate video data bit stream, the second bit rate video data bit stream and the first bit rate video data bit stream, if there is frame loss or damage, demultiplexing the first bit rate video data bit stream and the second bit rate video data bit stream, decoding the first bit rate video data bit stream, removing damaged frames from the first bit rate video data bit stream, decoding the second bit rate video data bit stream, up-sampling frames from said processed second bit rate video data bit stream if said processed second bit rate video data bit stream has a lower resolution than said processed first bit rate video data bit stream and merging frames from the second bit rate video data bit stream and the first bit rate video data bit stream.

Abstract: A mutliview video decoding apparatus including: a decoding unit which decodes one of coded videos without reference to a coded video to generate one of decoded videos, and decodes an other one of coded videos with reference to the one of the coded videos to generate an other one of decoded videos; an error detecting unit which detests an error-source image in the decoded videos; and a decoded image replacing unit which replaces the error-source image with an image generated using a previous image that is decoded before the error-source image, without using the second decoded video that is other than the first decoded video that includes the error-source image, and replaces an associated error image that is associated with the error-source image with an image generated using an associated previous image that is associated with the error-source image, without using the first decoded video.

Abstract: Techniques and tools for video coding/decoding with sub-block transform coding/decoding and re-oriented transforms are described. For example, a video encoder adaptively switches between 8×8, 8×4, and 4×8 DCTs when encoding 8×8 prediction residual blocks; a corresponding video decoder switches between 8×8, 8×4, and 4×8 inverse DCTs during decoding. The video encoder may determine the transform sizes as well as switching levels (e.g., frame, macroblock, or block) in a closed loop evaluation of the different transform sizes and switching levels. When a video encoder or decoder uses spatial extrapolation from pixel values in a causal neighborhood to predict pixel values of a block of pixels, the encoder/decoder can use a re-oriented transform to address non-stationarity of prediction residual values.