Nokia Technologies Oy

Nokia Corporation

Nokia Mobile Phones Limited

Core Wireless Licensing S.a.r.l.

Conversant Wireless Licensing S.a R.l.

Jani Lainema

Kemal Ugur

Antti Hallapuro

Miska Matias Hannuksela

Marta Karczewicz

Francesco Cricri

Mehmet Oguz Bici

Emre Baris Aksu

Bogdan-Paul Dobrin

Vinod Kumar Malamal Vadakital

Alireza Aminlou

Alireza Zare

Ramin Ghaznavi Youvalari

Mikko Johannes Honkala

Hamed Rezazadegan Tavakoli

Dmytro Rusanovskyy

Honglei Zhang

Nannan Zou

Kemai Ugur

Francesco Cricrì

Fehmi Chebil

Caglar Aytekin

Ali Karaoglu

Kai Willner

Levent Oktem

Yat Hong Lam

Martz Karczewicz

Pekka Astola

Antti Olli Hallapuro

The embodiments relate to method for encoding and decoding, wherein the method for encoding comprises receiving an input block of a video frame for encoding; applying at least a learning-based model () for said input block as a processing step for encoding the block; combining () an output of a learning-based model with one or more data sources () by a combination process; encoding block to a bitstream (); using a result of the combination process as additional input for the learning-based model for encoding a subsequent block; and encoding to a bitstream combination information () used in the combination process, said combination information comprising at least one or more combination parameters. The embodiments also relate to technical equipment for implementing the methods.

Method, an apparatus and a computer program product for video encoding and video decoding

There is disclosed an apparatus, a method and a computer program for video coding. The apparatus comprises a selector configured for selecting a pixel for prediction; a projection definer configured for determining a projection of said pixel to a set of reference pixels; and a prediction definer configured for selecting one or more reference pixels from said set of reference pixels on the basis of said projection, and using said selected one or more reference pixels to obtain a prediction value for said pixel to be predicted.

Apparatus, a method and a computer program for video coding

A method comprising: obtaining an image block unit comprising samples in color channels of one or two chrominance channels and one luminance channel (); reconstructing samples of said luminance channel of the image block unit (); predicting chrominance samples of at least one color channel of image block unit using at least two cross-component prediction models, each comprising their corresponding linear model parameters, and a luminance threshold value dividing the samples of said luminance channel between said at least two cross-component prediction models (); defining a weighting function for said at least two cross-component prediction models based on location of said two or more chrominance samples and the luminance threshold value (); and applying a smoothing function for samples residing within an offset value from said luminance threshold value, wherein said smoothing function adjusts chrominance values of said samples residing within said offset value by applying said weighting function on the corresponding at least two linear model parameters of the at least two cross-component prediction models ().

Apparatus, a method and a computer program for video coding and decoding

The embodiments relate to a method and technical equipment for implementing the method. The method comprises receiving a block of input samples; performing at least a first inverse transform on a first set of input samples, resulting in a first set of inverse transformed samples; performing at least a second inverse transform on a second set of input samples, resulting in a second set of inverse transformed samples; determining statistical information related to the first set of inverse transformed samples and for determining statistical information related to the second set of inverse transformed samples; updating the first inverse transform based on the determined statistical information related to the first set of inverse transformed samples; and updating the second inverse transform based on the determined statistical information related to the second set of inverse transformed samples.

A method comprising: determining a first sample in an input color space (); determining a component of the input color space that the first sample represents (); determining a pre-quantizer based on the determined component (); quantizing the first sample using the determined pre-quantizer (); and encoding the first sample value with video or image coding means ().

Various embodiments provide an apparatus, a method, and a computer program product. 1. An apparatus incudes at least one processor; and at least one non-transitory memory includes computer program code; wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus at least to: perform an overfitting operation, at an encoder side, to obtain an overfitted probability model, wherein overfitting comprises one or more training operations applied to a probability model, wherein one or more parameters of the probability model are trained; use the overfitted probability model to provide probability estimates to a lossless codec or a substantially lossless codec for encoding data or a portion of the data; and signal information to a decoder on whether to perform the overfitting operation at the decoder side.

Apparatus method and computer program product for probability model overfitting

There is disclosed a method comprising obtaining a block of transform coefficients; searching a position of a last non-zero coefficient in a scan order in the block; selecting a secondary transform mode based on the position and encoding first syntax elements indicative of the selected secondary transform mode. A primary transform mode is selected based on the secondary transform mode. Second syntax elements indicative of the selected primary transform mode and third syntax elements defining the transform coefficients within the block are encoded. There is also disclosed a method comprising receiving encoded information related to a block of transform coefficients; decoding an indication of a position of a last non-zero coefficient in a scan order in the block and using the position to determine if an indication of a secondary transform mode is present in the bit stream. If so, first syntax elements indicative of the secondary transform mode are decoded and used to determine if an indication of a primary transform mode is present in the bit stream. If so, second syntax elements indicative of the primary transform mode are decoded. The transform coefficients are decoded and the determined transform modes are applied to select transforms and the transforms, if any, are applied to the transform coefficients.

A system and method for improving the coding efficiency of motion vector information in video coding. According to various embodiments, a list of motion vector predictor candidates is arranged according to predefined rules. Each motion vector also has a reference index associated with it. One of the motion vector candidates is then selected as a predictor based on predefined rules, or the selection is explicitly signaled in the bitstream. The reference index associated with the selected motion vector is used as a reference index for the current block. The reference index is predicted along with the motion vector. Such embodiments can improve the compression efficiency of modern video codecs.

Combined motion vector and reference index prediction for video coding

A method, apparatus, and computer program product provide for improving parameter selection decisions in hybrid video and image codecs. In the context of a method, the method generates a block of difference samples based on a difference between a first block of samples and a second block of samples. The method also generates a block of transform coefficients by applying a transformation process to the block of difference samples. The method also determines an absolute value for each transform coefficient in the block of transform coefficients and determines a weighted sum value for all absolute values of the transform coefficients within the block of transform coefficients. The method also applies the weighted sum value as a criterion for a selection decision.

DC down-scaled weighted cost function for image/video coding

The embodiments relate to a method for encoding and a decoding, and apparatuses for the same. The method for encoding comprises receiving a block of a video frame for encoding (); making a decision on whether or not a learning-based model is to be applied as a processing step for encoding the block (); applying the learning-based model for said input block according to the decision, where the learning-based model has been selectively fine-tuned according to information relating to activation of the learning-based model of previously-decoded blocks (); encoding a signal corresponding to the decision on usage of the learning-based model into a bitstream (); and encoding the block into a bitstream with an information whether the block is to be used for finetuning ().

Apparatuses, methods and computer programs are provided for utilizing motion prediction in video coding. A block of pixels of a video representation encoded in a bitstream is read, and a type of the block is determined. If the determining indicates that the block is a block predicted by using two or more reference blocks, a first reference pixel location in a first reference block is determined and a second reference pixel location in a second reference block is determined. The first reference pixel location is used to obtain a first prediction. The first prediction has a second precision, which is higher than the first precision. The second reference pixel location is used to obtain a second prediction, which also has the second precision. The first prediction and the second prediction are combined to obtain a combined prediction; and the precision of the combined prediction is reduced to the first precision.

Motion prediction in video coding

A method includes maintaining a set of parameters or weights derived through online learning for a neural net; transmitting an update of the parameters or weights to a decoder; deriving a first prediction block based on an output of the neural net using the parameters or weights; deriving a first encoded prediction error block through encoding a difference of the first prediction block and a first input block; encoding the first encoded prediction error block into a bitstream; deriving a reconstructed prediction error block based on the first encoded prediction error block; deriving a second prediction block based on an output of the neural net using the parameters or weights and the reconstructed prediction error block; deriving a second encoded prediction error block through encoding a difference of the second prediction block and a second input block; and encoding the second encoded prediction error block into a bitstream.

The invention relates to a method for encoding, a method for decoding, an apparatus, computer program products, an encoder and a decoder for video information. The motion vector for a block in a video image is predicted from a set of motion vector prediction candidates determined based on previously-coded motion vectors. A motion vector prediction candidate is included in the set based on the location of the block associated with the first spatial motion vector prediction candidate and in comparison with motion vector prediction candidates already in the set.

Method for coding and an apparatus

An apparatus includes circuitry configured to: partition an input tensor into one or more block tensors; partition at least one of the block tensors into one or more continuation bands, the one or more continuation bands being associated with a caching counter having a value; store the one or more continuation bands in a cache managed using a cache manager; retrieve, prior to a convolution or pooling operation on a current block tensor, the one or more continuation bands of a previous block tensor from the cache that are adjacent to a current block tensor; concatenate the retrieved continuation bands with the current block tensor; apply the convolution or pooling operation on the current block tensor after the concatenation; decrease the respective caching counter value of the retrieved continuation bands; and clear the continuation bands from the cache when its respective caching counter reaches a value of zero.

Caching and clearing mechanism for deep convolutional neural networks

A method comprising: obtaining a block of a picture or a picture in an encoder; determining if the block/picture is used for on-line learning; if affirmative, encoding the block/picture; reconstructing a coarse version of the block/picture or the respective prediction error block/picture; enhancing the coarse version using a neural net; fine-tuning the neural net with a training signal based on the coarse version; determining if the block/picture is enhanced using the neural net; and if affirmative, encoding the block/picture with enhancing using the neural net.

Growing community of inventors

Tampere, Finland

Jani Lainema