Methods, devices and systems for video coding and encoding, which include conforming to constraints on collocated pictures, are described. One example method of video processing includes performing a conversion between a video including a current picture and a bitstream of the video, wherein the bitstream conforms to a format rule, and wherein the format rule specifies that the bitstream includes a flag for disabling a use of a temporal motion vector prediction tool for the current picture based on whether a reference picture in a decoded picture buffer has a spatial resolution and a scaling window offset that are identical to that of the current picture.

Methods, devices and systems for video coding and encoding, which include conforming to constraints on collocated pictures, are described. One example method of video processing includes performing a conversion between a video including a current picture and a bitstream of the video, wherein the bitstream conforms to a format rule, and wherein the format rule specifies that the bitstream includes a flag for disabling a use of a temporal motion vector prediction tool for the current picture based on whether a reference picture in a decoded picture buffer has a spatial resolution and a scaling window offset that are identical to that of the current picture.

A method of multi-scale neural image compression with intra-prediction residuals is performed by at least one processor and includes downsampling an input image, generating a current predicted image, based on a previously-recovered predicted image, and generating a prediction residual based on a difference between the downsampled input image and the generated current predicted image. The method further includes encoding the generated prediction residual, decoding the encoded prediction residual, and generating a currently-recovered predicted image based on an addition of the current predicted image and the decoded prediction residual. The method further includes upsampling the currently-recovered predicted image, generating a scale residual based on a difference between the input image and the upsampled currently-recovered predicted image, and encoding the scale residual.

A method, computer program, and computer system is provided for aligning across layers in a coded video stream. A video bitstream having multiple layers is decoded. One or more subpicture regions are identified from among the multiple layers of the decoded video bitstream, the subpicture regions including a background region and one or more foreground subpicture regions. An enhanced subpicture is decoded and displayed based on a determination that a foreground subpicture region is selected. The background region is decoded and displayed based on a determination that a foreground subpicture region was not selected.

An image processing apparatus, including processing circuitry configured to decode a bit stream to generate quantized data. The bit stream includes a flag, for each block, that specifies whether or not a difference quantization parameter is present in the bit stream. The flag is included in a first layer that is higher than a second layer in which the difference quantization parameter is set. The processing circuitry is configured to set, according to the flag, a current quantization parameter for a current sub block formed by block partitioning, which splits a block into smaller sub blocks. The processing circuitry is configured to inversely quantize the generated quantized data using the set current quantization parameter.

An image processing apparatus, including processing circuitry configured to decode a bit stream to generate quantized data. The bit stream includes a flag, for each block, that specifies whether or not a difference quantization parameter is present in the bit stream. The flag is included in a first layer that is higher than a second layer in which the difference quantization parameter is set. The processing circuitry is configured to set, according to the flag, a current quantization parameter for a current sub block formed by block partitioning, which splits a block into smaller sub blocks. The processing circuitry is configured to inversely quantize the generated quantized data using the set current quantization parameter.

Downsampled video content is generated in a subsampling color space from linearized video content in the subsampling color space. The linearized video content represents a first spatial dimension, whereas the downsampled video content represents a second spatial dimension lower than the first spatial dimension. Opponent channel data is derived in a transmission color space from the downsampled video content. Output video content is generated from luminance data in the linearized video content and the opponent channel data in the transmission color space. The output video content may be decoded by a downstream recipient device to generate video content in an output color space.

Downsampled video content is generated in a subsampling color space from linearized video content in the subsampling color space. The linearized video content represents a first spatial dimension, whereas the downsampled video content represents a second spatial dimension lower than the first spatial dimension. Opponent channel data is derived in a transmission color space from the downsampled video content. Output video content is generated from luminance data in the linearized video content and the opponent channel data in the transmission color space. The output video content may be decoded by a downstream recipient device to generate video content in an output color space.

To increase efficiency of a bit-depth scalable data-stream an inter-layer prediction is obtained by mapping samples of the representation of the picture or video source data with a first picture sample bit-depth from a first dynamic range corresponding to the first picture sample bit-depth to a second dynamic range greater than the first dynamic range and corresponding to a second picture sample bit-depth being higher than the first picture sample bit-depth by use of one or more global mapping functions being constant within the picture or video source data or varying at a first granularity, and a local mapping function locally modifying the one or more global mapping functions and varying at a second granularity smaller than the first granularity, with forming the quality-scalable data-stream based on the local mapping function such that the local mapping function is derivable from the quality-scalable data-stream.

To increase efficiency of a bit-depth scalable data-stream an inter-layer prediction is obtained by mapping samples of the representation of the picture or video source data with a first picture sample bit-depth from a first dynamic range corresponding to the first picture sample bit-depth to a second dynamic range greater than the first dynamic range and corresponding to a second picture sample bit-depth being higher than the first picture sample bit-depth by use of one or more global mapping functions being constant within the picture or video source data or varying at a first granularity, and a local mapping function locally modifying the one or more global mapping functions and varying at a second granularity smaller than the first granularity, with forming the quality-scalable data-stream based on the local mapping function such that the local mapping function is derivable from the quality-scalable data-stream.