An object of the embodiments is to achieve an improved reference picture handling. That is achieved by taking into account whether the reference pictures in the decoded picture buffer are long-term reference pictures or short-term reference pictures when determining how they should be marked when the information of the reference picture set is received. The reference pictures are marked as “used for short-term reference” or “used for long-term reference” in the Decoded Picture Buffer (DPB) depending on whether they are included as short-term pictures or long-term pictures in the RPS of a current picture.

A method for reconstructing residual blocks of chroma blocks, and a video decoding apparatus are disclosed. According to one embodiment of the present disclosure, a method for reconstructing residual blocks of chroma blocks on the basis of a joint coding of chroma residuals (JCCR) model comprises: determining the mode of the JCCR model based on a Cb chroma component syntax element and a Cr chroma component syntax element; generating, from transform coefficients of a chroma component corresponding to the mode of the JCCR model from among the Cb chroma component and the Cr chroma component, residual samples of the corresponding chroma component; and deriving residual blocks of the chroma blocks by using the generated residual samples.

Systems, devices, and methods related to a Deep Learning Accelerator and memory are described. For example, a removable media (e.g., a memory card, or a USB drive) may be configured to execute instructions with matrix operands and configured with: an interface to receive a video stream; and random access memory to buffer a portion of the video stream as an input to an Artificial Neural Network and to store instructions executable by the Deep Learning Accelerator and matrices of the Artificial Neural Network. Such a removable media can be used to replace an existing removable media used in a surveillance camera to record video or images. The Deep Learning Accelerator can execute the instructions to generate analytics of the buffer portion using the Artificial Neural Network, enabling the surveillance camera that is upgraded via the use of the removable media to provide intelligent services based on the analytics.

Systems, devices, and methods related to a Deep Learning Accelerator and memory are described. For example, a removable media (e.g., a memory card, or a USB drive) may be configured to execute instructions with matrix operands and configured with: an interface to receive a video stream; and random access memory to buffer a portion of the video stream as an input to an Artificial Neural Network and to store instructions executable by the Deep Learning Accelerator and matrices of the Artificial Neural Network. Such a removable media can be used to replace an existing removable media used in a surveillance camera to record video or images. The Deep Learning Accelerator can execute the instructions to generate analytics of the buffer portion using the Artificial Neural Network, enabling the surveillance camera that is upgraded via the use of the removable media to provide intelligent services based on the analytics.

According to one embodiment of the present document, an in-loop filtering procedure in an image/video coding procedure may comprise a cross-component adaptive loop filtering procedure. CCALF, according to the present embodiment, can increase the accuracy of in-loop filtering.

According to one embodiment of the present document, an in-loop filtering procedure in an image/video coding procedure may comprise a cross-component adaptive loop filtering procedure. CCALF, according to the present embodiment, can increase the accuracy of in-loop filtering.

A method for decoding a coded video bitstream is provided. The method includes: obtaining a reference layer syntax element by parsing the coded video bitstream, wherein a value of the reference layer syntax element specifying whether a layer with index k is a direct reference layer for a layer with index i; determining whether the layer with index j is a reference layer for the layer with index i based on the value of the reference layer syntax element; and when a condition is satisfied, predicting a picture of the layer with index i based on the layer with index j, wherein the value of a chroma format related syntax element applied to the layer with index i is the same as the value of a chroma format related syntax element applied to the layer with index j.

A method for decoding a coded video bitstream is provided. The method includes: obtaining a reference layer syntax element by parsing the coded video bitstream, wherein a value of the reference layer syntax element specifying whether a layer with index k is a direct reference layer for a layer with index i; determining whether the layer with index j is a reference layer for the layer with index i based on the value of the reference layer syntax element; and when a condition is satisfied, predicting a picture of the layer with index i based on the layer with index j, wherein the value of a chroma format related syntax element applied to the layer with index i is the same as the value of a chroma format related syntax element applied to the layer with index j.

Disclosed are a partitioning method, an encoder, a decoder and a computer storage medium. The method includes: determining location information of a point of a point cloud to be partitioned; when i is less than or equal to M−1, determining right-shift number N.sub.i of ith LOD layer in the point cloud, M representing a preset maximum quantity of layers for LOD partitioning; for the ith LOD layer, shifting location information of the point rightwards by N.sub.i-digit, performing storing in a preset storage area based on right-shifted location information; determining location information of a parent point corresponding to a current point in the ith LOD layer; according to determined location information of the parent point, searching the preset storage area for a neighbor point of the parent point; partitioning the current point into an (i+1)th LOD layer, or the neighbor point into the ith LOD layer.

A method for processing a video includes performing a determination, by a processor, that a first video block is partitioned to include a first prediction portion that is non-rectangular and non-square; adding a first motion vector (MV) prediction candidate associated with the first prediction portion to a motion candidate list associated with the first video block, wherein the first MV prediction candidate is derived from a sub-block MV prediction candidate; and performing further processing of the first video block using the motion candidate list.