Coding of the last coded coefficient position is performed by basing the coding of they coordinate of the position of the last coded coefficient on knowledge of the size of the partial transform used to generate a block of coefficients from a block of video pixels. This enables a context adaptive coding of the last coded coefficient parameter to be performed much more efficiently.

Coding of the last coded coefficient position is performed by basing the coding of they coordinate of the position of the last coded coefficient on knowledge of the size of the partial transform used to generate a block of coefficients from a block of video pixels. This enables a context adaptive coding of the last coded coefficient parameter to be performed much more efficiently.

A learning model is trained for rate-distortion behavior prediction against a corpus of a video hosting platform and used to determine optimal bitrate allocations for video data given video content complexity across the corpus of the video hosting platform. Complexity features of the video data are processed using the learning model to determine a rate-distortion cluster prediction for the video data, and transcoding parameters for transcoding the video data are selected based on that prediction. The rate-distortion clusters are modeled during the training of the learning model, such as based on rate-distortion curves of video data of the corpus of the video hosting platform and based on classifications of such video data. This approach minimizes total corpus egress and/or storage while further maintaining uniformity in the delivered quality of videos by the video hosting platform.

A learning model is trained for rate-distortion behavior prediction against a corpus of a video hosting platform and used to determine optimal bitrate allocations for video data given video content complexity across the corpus of the video hosting platform. Complexity features of the video data are processed using the learning model to determine a rate-distortion cluster prediction for the video data, and transcoding parameters for transcoding the video data are selected based on that prediction. The rate-distortion clusters are modeled during the training of the learning model, such as based on rate-distortion curves of video data of the corpus of the video hosting platform and based on classifications of such video data. This approach minimizes total corpus egress and/or storage while further maintaining uniformity in the delivered quality of videos by the video hosting platform.

Data may be encoded to minimize distortion after decoding, but the quality required for presentation of the decoded data to a machine and the quality required for presentation to a human may be different. To accommodate different quality requirements, video data may be encoded to produce a first set of encoded data and a second set of encoded data, where the first set may be decoded for use by one of a machine consumer or a human consumer, and a combination of the first set and the second set may be decoded for use by the other of a machine consumer or a human consumer. The first and second set may be produced with a neural encoder and a neural decoder, and/or may be produced with the use of prediction and transform neural network modules. A human-targeted structure and a machine-targeted structure may produce the sets of encoded data.

Data may be encoded to minimize distortion after decoding, but the quality required for presentation of the decoded data to a machine and the quality required for presentation to a human may be different. To accommodate different quality requirements, video data may be encoded to produce a first set of encoded data and a second set of encoded data, where the first set may be decoded for use by one of a machine consumer or a human consumer, and a combination of the first set and the second set may be decoded for use by the other of a machine consumer or a human consumer. The first and second set may be produced with a neural encoder and a neural decoder, and/or may be produced with the use of prediction and transform neural network modules. A human-targeted structure and a machine-targeted structure may produce the sets of encoded data.

Techniques related to video encoding include inline downscaling hardware in multi-pass encoding.

An example method of encoding or decoding video data includes determining a motion vector for a block of video data using a decoder side motion vector derivation process that includes performing an iterative search process, wherein performing the iterative search process includes: determining a minimum cost through a previous search iteration; updating an overall minimum cost through a current search iteration; and terminating the iterative search process early based on a comparison of the minimum cost through the previous search iteration and the overall minimum cost through the current search iteration; and encoding or decoding the block of video data using the motion vector.

An example method of encoding or decoding video data includes determining a motion vector for a block of video data using a decoder side motion vector derivation process that includes performing an iterative search process, wherein performing the iterative search process includes: determining a minimum cost through a previous search iteration; updating an overall minimum cost through a current search iteration; and terminating the iterative search process early based on a comparison of the minimum cost through the previous search iteration and the overall minimum cost through the current search iteration; and encoding or decoding the block of video data using the motion vector.

A method for decoding a video bitstream is disclosed. The method comprises: entropy decoding a first portion of a video bitstream, wherein first portion of video bitstream is associated with a video frame, thereby producing a first portion of decoded data; entropy decoding a second portion of video bitstream, wherein second portion of video bitstream is associated with video frame, thereby producing a second portion of decoded data, wherein entropy decoding second portion of video bitstream is independent of entropy decoding first portion of video bitstream; and reconstructing a first portion of video frame associated with video bitstream using first portion of decoded data and second portion of decoded data.