Abstract: The invention relates to a method for reconstructing a picture that is part of a sequence of pictures, from coded digital data, representative of said current picture. It comprises the following steps: construct, for each picture block, first and second prediction blocks (from a first picture and a second picture previously reconstructed, and calculate, for each block, a confidence value representative of the proximity between the visual content of the first predictive block and the visual content of the second predictive block, and reconstruct each block from a part of said coded digital data representative of said block and initial auxiliary data calculated from the first and second predictive blocks when the block confidence value is greater than a predetermined threshold.

Abstract: The method comprising a wavelet-based spatial analysis step comprising filtering operations of the picture at the high resolution format and horizontal and/or vertical decimation operations of the filtered picture to supply subband signals, is wherein a decimation operation is carried out according to a factor different from an even value and corresponds to the quotient of the horizontal and vertical dimensions of the first and second format such that the low frequency picture thus obtained relative to the low frequency subband signals, corresponds to the second format.

Abstract: The invention relates to spatially scalable encoding and decoding processes using a method for deriving coding information. More particularly, it relates to a method for deriving coding information for high resolution pictures from the coding information of low resolution pictures. The method mainly comprises the following steps: Computing geometrical parameters characterizing the position of said high layer macroblock relatively to the corresponding base layer macroblocks and deriving from these parameters a macroblock class; Deriving a partition and possibly sub-partitions for each partition of said high layer macroblock from corresponding base layer macroblocks partition and sub-partitions on the basis of the geometrical parameters and HL MB class; and Deriving motion information for said high layer macroblock from motion information of corresponding base layer macroblocks.

Abstract: The invention relates to spatially scalable encoding and decoding processes using a method for deriving coding information. More particularly, it relates to a method for deriving coding information used to encode high resolution images from coding information used to encode low resolution images when the ratio between high resolution and low resolution images dimensions is a multiple of 3/2. The method mainly comprises the following steps: deriving a block coding mode for each 8×8 blocks of a prediction macroblock MBi_pred from the macroblock coding mode of the associated base layer macroblocks on the basis of the macroblock class of MBi and an the basis of the position of the 8×8 block within MBi_pred; deriving a macroblock coding mode for MBi_pred from the coding modes of the associated base layer macroblocks; and deriving motion information for each macroblock MBi_pred from the motion information of the associated base layer macroblocks.

Abstract: Interlaced video can be encoded in two layers, base layer and enhancement layer. A method for optimizing the encoding of motion vectors for enhancement layer is proposed. It comprises defining various different ways to encode the motion vectors, estimate the coding costs for each way, and select the way with the lowest coding cost. The various ways to encode enhancement layer motion vectors can be combined. Either motion vectors from base layer are reused, resulting in a scaling factor and an update vector, or new motion vectors are calculated. Either forward or backward or bi-directional prediction can be used. Either one frame or multiple frames can be taken as reference. The reference frame can be from the enhancement layer, base layer or both. Either the complete motion vectors are encoded, or only the difference between the motion vector and the corresponding BL motion vector is encoded.

Abstract: The method realizes a motion compensated temporal filtering (MCTF), the temporal filtering being replaced by an intra mode coding to obtain at least one low (L) or high (H) frequency picture if the current picture has a level of correlation with a lower previous picture at a threshold, the low frequency pictures obtained (L) being thus scaled to be adapted, at the energy level, to the pictures obtained by motion compensated temporal filtering, and comprises, at the end of analysis: a selection of the pictures obtained by intra coding of a picture of a low decomposition level with the additional condition, for the high frequency pictures, that this picture is derived itself from an intra coding. a calibration of the picture selected by carrying out at least one reverse step of the scaling step. The applications relate to video compression with temporal prediction.

Abstract: The method comprising a temporal analysis implementing a motion compensated temporal filtering, said filtering comprising, to get a high frequency band picture H at a temporal level l, a predict step implementing high pass filtering of pictures L of lower temporal level, and, to get a low frequency band picture L at a temporal level l, an update step implementing low pass filtering of pictures H of same level and/or of pictures L of lower temporal level, also comprises, at least for a temporal level, a predict step which carries out a long filtering and an update step which carries out a short filtering. Applications relate to video compression for transmission or storage of data.

Abstract: The method implements an update step which is a selection among at least the following coding modes: a default coding mode which doesn't use the motion compensation for the calculation of the pixel if the pixel is not connected or which performs a low pass motion compensated temporal filtering of picture H of same level if the pixel is connected, an update_MV coding mode which performs a low pass motion compensated temporal filtering of picture H of same level according to a default motion vector calculated by taking into account motion vectors belonging to regions or pixels in the vicinity of the pixel to be calculated. Applications relate to video compression for transmission or storage of data.

Abstract: In video encoding, the video frames are spatio-temporally filtered for reduction of spatial and temporal redundancy before they are entropy encoded. Known filtering schemes consider temporally successive frames and are static. It is probable but not necessary that successive frames are most efficient to encode. Therefore, a plurality or all possible frame order permutations are considered for a group of frames (GOP) and evaluated based on a global criterion, which is the sum of local criterion values computed over successive subsets of permuted frames. The local criterion value is deduced from motion estimation processed on each considered set of frames. The best ordering is chosen as the one that minimizes the global criterion value.

Abstract: The following steps are carried out in the inventive method: estimation of movement between a current moment corresponding to the current image and a reference moment corresponding to a reference image in order to provide a movement vector, an entropic coding for a high spatial frequency signal relating in the current image, taking into account a temporal context based on the estimation of a movement, one of the parameters used to calculate the temporal context being the parity of a component of the movement vector. Applications are related to video compression with temporal prediction.

Abstract: The method is characterized in that the resolution chosen for the motion information and the complexity of the interpolation filters used during a motion compensated filtering operation depend on a decoding scenario, namely the spatial and temporal resolutions and the bit-rate selected for the decoding or the corresponding temporal decomposition level or a combination of these parameters. The applications relate to the video coders/decoders known as “scalable” for example in the domain of video telephony or video transmission over internet.

Abstract: Fully scalable encoder and decoder for interlaced video. A method for encoding an interlaced sequence of digital video data decomposes the interlaced video sequence into first and second fields, performs digital filtering to get lower frequency and higher frequency component signals of the first fields, and uses spatio-temporal filtering and motion estimation for generating base layer signals being suitable for reconstruction of a progressive mode video sequence in a receiver. Advantageously, both the spatio-temporal filter at the encoder, and the inverse process at the receiver, can perform scaling in spatial and temporal dimension. The second fields are used to generate enhancement signals, which enable a receiver to reproduce an interlaced video sequence of the full, or scaled, spatial and/or temporal resolution.

Abstract: The method comprising a wavelet-based spatial analysis step comprising filtering operations of the picture at the high resolution format and horizontal and/or vertical decimation operations of the filtered picture to supply subband signals, is wherein a decimation operation is carried out according to a factor different from an even value and corresponds to the quotient of the horizontal and vertical dimensions of the first and second format such that the low frequency picture thus obtained relative to the low frequency subband signals, corresponds to the second format.