Abstract: Video decoding systems and techniques are described. The decoder applies a deblocking (DB) filter to the plurality of sub-blocks of a block of video data to generate a DB-filtered plurality of sub-blocks. The decoder applies the DB filter to one or more lines (e.g., columns) of pixels in an additional sub-block of the block to generate a DB-filtered portion of the additional sub-block. The one or more lines of pixels in the additional sub-block are filtered without filtering an entirety of the additional sub-block using the DB filter. The additional sub-block is adjacent to at least one of the plurality of sub-blocks. The decoder applies a constrained directional enhancement filter (CDEF) to the DB-filtered plurality of sub-blocks and the DB-filtered portion of the additional sub-block to generate a CDEF-filtered plurality of sub-blocks.

Abstract: Systems and techniques are described for data encoding using a machine learning approach to generate a distortion prediction {circumflex over (D)} and a predicted bit rate {circumflex over (R)}, and to use {circumflex over (D)} and {circumflex over (R)} to perform rate-distortion optimization (RDO). For example, a video encoder can generate the distortion prediction {circumflex over (D)} and the bit rate residual prediction based on outputs of the one or more neural networks in response to the one or more neural networks receiving a residual portion of a block of a video frame as input. The video encoder can determine bit rate metadata prediction based on metadata associated with a mode of compression, and determine {circumflex over (R)} to be the sum of and . The video encoder can determine a rate-distortion cost prediction ? as a function of {circumflex over (D)} and {circumflex over (R)}, and can determine a prediction mode for compressing the block based on ?.

Abstract: Systems and techniques are described for data encoding using a machine learning approach to generate a distortion prediction {circumflex over (D)} and a predicted bit rate {circumflex over (R)}, and to use {circumflex over (D)} and {circumflex over (R)} to perform rate-distortion optimization (RDO). For example, a video encoder can generate the distortion prediction {circumflex over (D)} and the bit rate residual prediction based on outputs of the one or more neural networks in response to the one or more neural networks receiving a residual portion of a block of a video frame as input. The video encoder can determine bit rate metadata prediction based on metadata associated with a mode of compression, and determine {circumflex over (R)} to be the sum of and . The video encoder can determine a rate-distortion cost prediction ? as a function of {circumflex over (D)} and {circumflex over (R)}, and can determine a prediction mode for compressing the block based on ?.