Method and device for processing a sequence of digital images with spatial or quality scalability

A method of decoding a sequence of digital images coded in a bitstream comprising at least first and second layers linked to each other in a chosen hierarchical relationship of scalability, each image of the second layer comprising a plurality of blocks of pixels, comprising a temporal prediction using motion compensation for at least one current group, wherein the decoding of at least one current block of said group of an image of the second layer comprises the following steps:

• decoding at least one index of a prediction image of the second layer and at least one motion vector associated with said block;
• verifying whether the motion vector so decoded goes out of the current group in the prediction image;
• in case of positive verification, extracting data on the basis of information from the prediction image, or if signaled, data from the image of the first layer temporally corresponding to the prediction image, that spatially corresponds to a position indicated by the motion vector and that are spatially complementary with respect to the current group, wherein pixels in an image of the second layer that are used for temporal prediction are restricted to a group of pixels of said prediction image corresponding to a group of pixels to be predicted;
• and
• using the data so obtained for the motion compensation;
• and
• wherein the bitstream further comprises a third layer and the step of extracting in case of positive verification comprises extracting from an image of the third layer, temporally corresponding to the prediction image, of data spatially corresponding to the position indicated by the motion vector.

A device for temporal prediction using motion estimation in a sequence of digital images in the context of coding in a bitstream comprising at least first and second layers linked to each other in a chosen hierarchical relationship of scalability, each image of the second layer comprising a plurality of groups of blocks of pixels, said coding enabling access to spatial parts of the images, the temporal prediction being applied to at least one current group of an image of a current second layer of the sequence, wherein the temporal prediction device comprises:

• means for constructing at least one set of reference pixels for the temporal prediction on the basis of information from a prediction image of the second layer, or, if signaled, spatially complementary information from an image of the first layer corresponding temporally to said prediction image, wherein pixels in an image of the second layer that are used for temporal prediction are restricted to a group of blocks of said prediction image corresponding to a group of pixels to be predicted;
• and
• processing means constructed to adapted, for each block of the current group of the image of the current second layer, search for at least one block of the reference set of pixels so constructed that is suitable for the temporal prediction, and to determine at least one corresponding motion vector;
• and
• wherein the bitstream further comprises a third layer and the step of extracting in case of positive verification comprises extracting from an image of the third layer, temporally corresponding to the prediction image, of data spatially corresponding to the position indicated by the motion vector.

A device for decoding a sequence of digital images coded in a bitstream comprising at least first and second layers linked to each other in a chosen hierarchical relationship of scalability, each image of the second layer comprising a plurality of groups of blocks of pixels, comprising a temporal compensation using motion compensation for at least one current group, wherein the device comprises, for the decoding of at least one current block of said group of an image of the second layer:

• means for decoding the at least one index of the prediction image and at least one motion vector associated with said block;
• verifying means for verifying whether the motion vector so decoded goes out of the current group in the prediction image;
• processing means that are constructed to, in case of positive verification, extract data on the basis of information from the prediction image, or, if signaled, data from the image of the first layer temporally corresponding to the prediction image that spatially correspond to the position indicated by the motion vector and that are spatially complementary with respect to the current group, wherein pixels in an image of the second layer that are used for temporal prediction are restricted to a group of blocks of said prediction image corresponding to a group of pixels to be predicted, and constructed to use data so obtained for the motion compensation;
• wherein the bitstream further comprises a third layer and the step of extracting in case of positive verification comprises extracting from an image of the third layer, temporally corresponding to the prediction image, of data spatially corresponding to the position indicated by the motion vector.