Abstract: A method performed by an apparatus to determine compression parameters to choose when compressing images or videos for use in a machine vision task is provided. The method includes compressing an uncompressed original image or video at a plurality of different quality levels and/or bit rates to create a plurality of compressed images or videos. The method further includes, for each compressed image or video compressed at the different quality levels and/or bit rates: decompressing the compressed image or video to create a decompressed image or video, executing a machine vision algorithm on the decompressed image or video to generate machine vision results for the decompressed image or video and deriving a performance value indicating a performance of the decompressed image or video based on comparing the machine vision results to an assumed truth.

Abstract: There is provided a method for encoding a picture. The method comprises, for each picture on either side of the current picture with a distance of one or two away from the current picture, computing a first error value, E, wherein the first error value, E, is computed for each block of a set of blocks of a first block size of the current picture and is based on the respective block of the current picture and a corresponding block in the picture. The method comprises, for each picture on either side of the current picture with a distance of one or two away from the current picture, aggregating the first error value, E, over a block of a second block size for each block of the set of blocks of the first block size contained in the block of the second block size to generate a second error value, E?, corresponding to the picture.

Abstract: A prediction unit for an encoder or decoder implements matrix based intra prediction. Input boundary samples for a current block are downsampled to obtain reduced boundary samples for matrix multiplication and/or linear interpolation, or both. In one embodiment, downsampling is performed in a manner that aligns the reduced boundary samples with an output of a matrix multiplication unit of the prediction unit. In other embodiments, downsampling is performed without averaging. The embodiments reduce the complexity of the prediction unit and the latency of the encoder or decoder.

Abstract: There is provided a method of training a machine learning, ML, model used for generating encoded video data or decoded video data. The method comprises obtaining original video data. The method further comprises converting the original video data into ML input video data. The method further comprises providing the ML input video data into the ML model, thereby generating first ML output video data, and training the ML model based on a difference between the original video data and the ML input video data and a difference between the original video data and the first ML output video data.

Abstract: There is provided a method of selecting from a picture a patch for training a machine learning, ML, model used for encoding or decoding video data. The method comprises randomly selecting one or more coordinates of the patch, and converting said one or more coordinates of the patch into converted one or more coordinates of the patch. The method further comprises training the ML model based on the converted one or more coordinates of the patch. Each of said one or more converted coordinates of the patch is an integer multiple of 2p, where p is an integer.

Abstract: A method performed by a decoder for decoding a bitstream comprising a picture parameter set, PPS, and a first set of slices. The method includes obtaining the picture parameter set. The method also includes decoding a syntax element included in the picture parameter set to obtain an indicator value. The decoder is configured such that if the indicator value is set to a first value then the decoder determines that a picture header included in the bitstream comprises a parameter value corresponding to a particular parameter, otherwise the decoder determines that each slice included in the first set of slices comprises a parameter value corresponding to the particular parameter. If the picture header comprises the parameter value corresponding to the particular parameter, then this parameter value is used to decode slice data of each slice included in the first set of slices.

Abstract: A method for encoding or decoding a region of a picture included in a video stream is provided. The method comprises obtaining a first value of a luma quantization parameter (QP) for luma component, and based on the first value of the luma QP, determining a first value of a chroma QP for chroma component. The method further comprises obtaining a second value of the luma QP, and based on the second value of the luma QP, determining a second value of the chroma QP. The method further comprises encoding or decoding the region of the picture in a second resolution using the second value of the luma QP and the second value of the chroma QP.

Abstract: A method for generating an encoded video or a decoded video is provided. The method comprises obtaining values of reconstructed samples, and obtaining input information comprising any one or a combination of: i) information about filtered samples, ii) information about predicted samples, or iii) information about skipped samples. The method further comprises providing the values of reconstructed samples and the input information to a machine learning, ML, model, thereby generating at least one ML output data, and based at least on said at least one ML output data, generating the encoded video or the decoded video.

Abstract: A prediction unit for an encoder or decoder implements matrix based intra prediction. Input boundary samples for a current block are downsampled to obtain reduced boundary samples for matrix multiplication and/or linear interpolation, or both. In one embodiment, downsampling is performed in a manner that aligns the reduced boundary samples with an output of a matrix multiplication unit of the prediction unit. In other embodiments, downsampling is performed without averaging. The embodiments reduce the complexity of the prediction unit and the latency of the encoder or decoder.

Abstract: There is provided a method, performed by a temporal filter. The method comprises obtaining a current picture and a set of one or more neighbor pictures comprising a first neighbor picture, wherein each of the one or more neighbor pictures either precedes or follows the current picture in a picture count order. The method comprises, based on a first block of pixel values of the current picture and a first block of pixel values in the first neighbor picture, determining a first filter weight, W1, for at least a first pixel value in the first block of pixel values of the current picture, wherein the first block of pixel values in the first neighbor picture correspond to the first block of pixel values in the current picture.

Abstract: There is provided a method of generating encoded video data or decoded video data. The method comprises providing input data to a first convolution layer, CL, thereby generating first convoluted data, generating residual data based on the first convoluted data, and generating the encoded video data or the decoded video data based on a combination of the input data and the residual data.

Abstract: There is provided a method for encoding one or more images. The method comprises filtering, with a first filter, input samples I(x,y) to generate a first filtered output Ifirst(x,y)=I(x,y)+?Ifirst(x,y). The method comprises estimating parameters for a second filter based at least in part on the first filtered output Ifirst(x,y), the input samples I(x,y), and original samples Iorig(x,y). The method comprises filtering, with the second filter, input samples I(x,y) to generate a second filtered output Isecond(x,y)=I(x,y)+?Isecond(x,y), wherein filtering, with the second filter, is based at least in part on the parameters estimated for the second filter. The method comprises generating a combined output ICOMB(x,y)=I(x,y)+?Ifirst(x,y)+?Isecond(x,y).

Abstract: Methods, decoders, and programs are provided. All blocks of an image that has been encoded are decoded. At least one sample of at least one block of the image is filtered by, for each block being filtered: for each sample being filtered: selecting a plurality of neighboring samples. For at least one neighboring sample, a difference is calculated between the neighboring sample and the sample. A modifier value is obtained. A modifier sum is derived based on a sum of the modifier values for the neighboring samples. A correction value is determined based on a conditional addition of at least one shifted version of the modifier sum, each shift having a same number of shift steps in each correction value determination. A filtered sample is derived based on the sample and correction value. The filtered sample is stored. The block and filtered sample is output towards a decoder output.

Abstract: Methods, decoders, and programs are provided. All blocks of an image that has been encoded are decoded. At least one sample of at least one block of the image is filtered by, for each block being filtered: for each sample being filtered: selecting a plurality of neighboring samples. For at least one neighboring sample, a difference is calculated between the neighboring sample and the sample. A modifier value is obtained. A modifier sum is derived based on a sum of the modifier values for the neighboring samples. A correction value is determined based on a conditional addition of at least one shifted version of the modifier sum, each shift having a same number of shift steps in each correction value determination. A filtered sample is derived based on the sample and correction value. The filtered sample is stored. The block and filtered sample is output towards a decoder output.

Abstract: A method performed by a decoder for decoding a bitstream comprising a picture parameter set, PPS, and a first set of slices. The method includes obtaining the picture parameter set. The method also includes decoding a syntax element included in the picture parameter set to obtain an indicator value. The decoder is configured such that if the indicator value is set to a first value then the decoder determines that a picture header included in the bitstream comprises a parameter value corresponding to a particular parameter, otherwise the decoder determines that each slice included in the first set of slices comprises a parameter value corresponding to the particular parameter. If the picture header comprises the parameter value corresponding to the particular parameter, then this parameter value is used to decode slice data of each slice included in the first set of slices.

Abstract: A method (400) for thinning a video comprising a sequence of pictures. The method includes the deciding whether or not to perform a video thinning process on a picture of the video. The method also includes performing a video thinning process on the picture of the video as a result of deciding to perform a video thinning process. The method also includes deciding whether or not to perform a video thinning process on another picture of the video. The method also includes, after deciding not to perform a video thinning process on the another picture, encoding the another picture to produce an encoded picture. The method further includes adding the encoded picture to a bitstream.

Abstract: The solution presented herein defines from which previously decoded picture to retrieve the values for calculating POC of the current picture regardless of how many temporal layers have been decoded. That is achieved by determining the POC of the current picture, to be used by the decoder, as a sum of a syntax element pic_order_cnt_Isb and most significant bits of the POC, PicOrderCntMsb, of the current picture, wherein the PicOrderCntMsb of the current picture is derived using at least a prevPicOrderCntMsb and a prevPicOrderCntLsb and the prevPicOrderCntMsb is set equal to the most significant bits of the POC of a previous reference picture in decoding order that has a layer identity equal to zero and prevPicOrderCntLsb is set equal to the value of the least significant bits of the POC of a previous reference picture in decoding order that has a layer identity equal to zero.

Abstract: A method, decoder apparatus, and computer programs are provided. A bit stream having a video block encoded with a multiple transform selection (MTS) index is received. The MTS index is parsed to determine a value of it. Responsive to the MTS index value being a predetermined value (or one of a predetermined set of indices), a LFNST index is parsed from the bit stream to determine a value of it, the value providing an indication whether a LFNST transform set is to be used in decoding the video block. Responsive to the MTS index value not being the predetermined value (or one of the set of indices), the value of the LFNST index is set to a default value indicating no LFNST transform is to be used in the decoding. The video block is decoded based on the values of the MTS index and the LFNST index.

Abstract: A method for encoding a video signal includes receiving an input including a sequence of pictures. A subset of the pictures from the sequence is selected. The selected subset of the pictures from the sequence is then filtered. The filtered selected subset of the pictures, and unfiltered pictures not in the selected subset, are then applied to an encoder to produce an encoded signal.

Abstract: There is provided a method, performed by a temporal filter. The method comprises obtaining a current picture and a set of one or more neighbor pictures comprising a first neighbor picture, wherein each of the one or more neighbor pictures either precedes or follows the current picture in a picture count order. The method comprises, based on a first block of pixel values of the current picture and a first block of pixel values in the first neighbor picture, determining a first filter weight, W1, for at least a first pixel value in the first block of pixel values of the current picture, wherein the first block of pixel values in the first neighbor picture correspond to the first block of pixel values in the current picture.