Abstract:
For frames sequentially functioning as a reference frame in a video sequence frame set, a motion estimate set is accessed. One motion estimate characterizes motion associated with pixels of each region of the reference frame in relation to regions of one frame of the set of frames, which is temporally displaced in time with respect to other frames in the set of frames. An additional motion estimate characterizes motion associated with pixels of each reference frame region in relation to a second frame of the set, which is temporally displaced from the one frame and other frames of the frame set. A temporal image prediction set, corresponding to the first and additional motion estimate, is predicted, based on an alignment of the reference frame regions over the frame set. The temporal image predictions are blended and a temporal predictor is generated over the frame set based on the blending.
Abstract:
According to a general aspect, a decoded picture is removed from memory based on dependency information that describes one or more inter-view dependency relationships for the picture. Implementations may also include accessing a picture from a first view and the dependency information. The dependency information may describe one or more inter-view dependency relationships for the picture from the first view. The picture from the first view may be decoded to produce the decoded picture. The decoded picture may be stored in memory. One application area involves removing pictures from a decoded picture buffer when such pictures are no longer needed as a reference picture.
Abstract:
Some video compression standards use multiple temporal predictions. One or more of the oldest temporal predict ions are overwritten with another prediction. A prediction used in an enhancement encoder is in embodiment overwritten by a prediction produced in a base stream encoder.
Abstract:
There are provided methods and apparatus for adaptive reference filtering. An apparatus includes an encoder (100) for encoding at least one picture. The encoder (100) performs adaptive filtering of at least one reference picture to respectively obtain at least one filtered reference picture, and predictively codes the at least one picture using the at least one filtered reference picture. The at least one reference picture is a picture wherein at least one sample thereof is used for inter-prediction either of subsequent to the at least one sample being applied to an in-loop filter or in an absence of the at least one sample being applied to an in-loop filter.
Abstract:
There are provided a method and apparatus for adaptive weight selection for motion compensated prediction. The apparatus includes an encoder (100) for encoding a picture by deriving a set of weighting parameters, selecting at least one weighting parameter in the set based upon a selection criteria, and applying the selected at least one weighting parameter to a reference picture used to encode the picture.
Abstract:
The present invention provides a video signal decoding method including the steps of checking whether to perform luminance compensation of a current block, performing the luminance compensation on the current block, and performing motion compensation by considering the luminance compensation, wherein a prediction value for the luminance compensation is obtained from a neighbor block neighbor to the current block.
Abstract:
The present invention provides a method of decoding a video signal. The method includes the steps of obtaining identification information indicating an assignment method of a reference index from the video signal, reordering reference pictures for inter-view prediction according to the identification information, and decoding the video signal using the reordered reference pictures, wherein the reference index includes view information for identifying a view of the reference picture.
Abstract:
A video encoder using inter-frame differential coding has a store (3) for containing a plurality of reference pictures. For each incoming picture it chooses one (or more) of the stored reference pictures to use. Control means (10) select which locally decoded pictures are to be entered into, and removed from, so that the store always contains at least one relatively older picture that lies within a defined time window and at least one relatively younger picture that is more recent than the time window. When it receives an error message from a receiver, it switches the coder from its normal a state (in which it is allowed to choose any picture as predictor) into a state in which it is permitted to choose only a said relatively older picture for prediction. The beginning of the time window is preferably set according to the estimated (or observed) round trip time between the encoder and decoder such that at least one of the relatively older pictures always predated the error message.