公开(公告)号：US11558628B2

公开(公告)日：2023-01-17

申请号：US17549039

申请日：2021-12-13

Applicant: Nokia Technologies Oy

Inventor： Honglei Zhang ,  Francesco Cricri ,  Hamed Rezazadegan Tavakoli ,  Jani Lainema ,  Emre Aksu ,  Nannan Zou

IPC: H04N19/423 ,  H04N19/107 ,  H04N19/176 ,  H04N19/119

Abstract: 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.

公开(公告)号：US11375204B2

公开(公告)日：2022-06-28

申请号：US17218967

申请日：2021-03-31

Applicant: Nokia Technologies Oy

Inventor： Honglei Zhang ,  Hamed Rezazadegan Tavakoli ,  Francesco Cricri ,  Miska Matias Hannuksela ,  Emre Aksu ,  Nam Le

IPC: H04N19/146 ,  H04N19/103 ,  H04N19/436 ,  H04N19/85

Abstract: An apparatus includes at least one processor; and at least one non-transitory memory including 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: decode encoded data to generate decoded data, the encoded data having a bitrate lower than that of original data, and extract features from the decoded data; decode encoded residual features to generate decoded residual features; and generate enhanced decoded features as a result of combining the decoded residual features with the features extracted from the decoded data.

公开(公告)号：US11363287B2

公开(公告)日：2022-06-14

申请号：US16973620

申请日：2019-07-08

Applicant: Nokia Technologies Oy

Inventor： Francesco Cricri ,  Antti Hallapuro ,  Miska Hannuksela ,  Jani Lainema ,  Emre Aksu ,  Caglar Aytekin ,  Ramin Ghaznavi Youvalari

IPC: H04N7/18 ,  H04N19/50 ,  H04N19/105 ,  H04N19/14 ,  H04N19/154 ,  H04N19/172 ,  H04N19/196 ,  G06N3/04 ,  G06N3/08

Abstract: Video data is obtained or received. At least a current frame or previous frame(s) of the obtained or received video data are provided to an input of a neural network. A predicted output is generated at an output of the neural network. The predicted output includes at least one of predicted future frame(s) of the video data and predicted properties of future frame(s) of the video data. Processing decision(s) are determined based, at least in part, on the predicted output. The current frame of the video data is processed at least partially according to the processing decision(s).

公开(公告)号：US20220141471A1

公开(公告)日：2022-05-05

申请号：US17575946

申请日：2022-01-14

Applicant: Nokia Technologies Oy

Inventor： Miska HANNUKSELA ,  Mikko Honkala ,  Jani Lainema ,  Francesco Cricri ,  Emre Aksu

IPC: H04N19/149 ,  H04N19/176 ,  H04N19/436 ,  H04N19/65 ,  G06N3/08

Abstract: 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.

公开(公告)号：US11109066B2

公开(公告)日：2021-08-31

申请号：US16634952

申请日：2018-08-13

Applicant: Nokia Technologies Oy

Inventor： Sebastian Schwarz ,  Miska Hannuksela ,  Alireza Aminlou ,  Emre Aksu

IPC: H04N19/597 ,  G06T7/593 ,  H04N13/376 ,  H04N13/383 ,  H04N13/117 ,  H04N13/341 ,  H04N13/194 ,  H04N13/161 ,  H04N13/243 ,  G06T15/20

Abstract: There are provided methods, apparatuses, systems and computer program products for coding volumetric video, where a first texture picture is coded, the first texture picture comprising a first projection of texture data of a first source volume of a digital scene model, the scene model comprising a number of further source volumes, the first projection being from the first source volume to a first projection surface, a first geometry picture is coded, the first geometry picture representing a mapping of the first projection surface to the first source volume, and first projection geometry information of the first projection is coded, the first projection geometry information comprising information of position of the first projection surface in the scene model.

公开(公告)号：US11094130B2

公开(公告)日：2021-08-17

申请号：US16745770

申请日：2020-01-17

Applicant: Nokia Technologies Oy

Inventor： Igor Curcio ,  Sujeet Mate ,  Kashyap Kammachi Sreedhar ,  Emre Aksu ,  Miska Hannuksela ,  Ari Hourunranta

IPC: G06T19/00 ,  G06F3/01 ,  H04N13/161 ,  H04N13/111

Abstract: The embodiments relate to a method, and a technical equipment for implementing the method. The method comprises generating a bitstream defining a presentation, the presentation comprising an omnidirectional visual media content; and indicating in the bitstream a definition for an external media to be overlaid on the omnidirectional visual media content during rendering; wherein the definition comprises at least an overlay placement information for the external media on the omnidirectional visual media content. The embodiments also relate to a method and technical equipment for decoding the bitstream.

公开(公告)号：US20210209829A1

公开(公告)日：2021-07-08

申请号：US17134711

申请日：2020-12-28

Applicant: Nokia Technologies Oy

Inventor： Lauri Ilola ,  Lukasz Kondrad ,  Emre Aksu ,  Miska Matias Hannuksela ,  Sebastian Schwarz

IPC: G06T15/00 ,  G06T1/20 ,  G06T1/60

Abstract: An apparatus includes at least one processor; and at least one non-transitory memory including 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: receive a scene description comprising data associated with a scene; place the data associated with the scene into data buffers and create command buffers; adapt the data placed within the data buffers and synchronize the data within the data buffers with information provided from local media or network media; signal information about the adaptation to update the command buffers that command a renderer; and render the scene using the data within the data buffers and the command buffers.

公开(公告)号：US20210195206A1

公开(公告)日：2021-06-24

申请号：US16771115

申请日：2018-12-03

Applicant: Nokia Technologies Oy

Inventor： Miska Hannuksela ,  Mikko Honkala ,  Jani Lainema ,  Francesco Cricri ,  Emre Aksu

IPC: H04N19/149 ,  H04N19/65 ,  H04N19/436 ,  H04N19/176 ,  G06N3/08

Abstract: A method comprising: deriving a first prediction block (608) at least partly based on an output of a neural net (602) using a first set of parameters; deriving a first encoded prediction error block (614-620) through encoding a difference of the first prediction block and a first input block; encoding (620) the first encoded prediction error block into a bitstream; deriving a first reconstructed prediction error block (624) from the first encoded prediction error block; deriving a training signal (628) from one or both of the first encoded prediction error block and/or the first reconstructed prediction error block (624); retraining (630) the neural net (602) with the training signal (628) to obtain a second set of parameters for the neural net (602); deriving a second prediction block (608) at least partly based on an output of the neural net using the second set of parameters; deriving a second encoded prediction error block (614-620) through encoding a difference of the second prediction block and a second input block; and encoding (620) the second encoded prediction error block into a bitstream. The invention relates to image or video encoding or decoding, especially by online training a neural network (602) that is in the prediction loop.

公开(公告)号：US20200036955A1

公开(公告)日：2020-01-30

申请号：US16495288

申请日：2018-03-20

Applicant: Nokia Technologies Oy

Inventor： Mika Pesonen ,  Kimmo Roimela ,  Johannes Pystynen ,  Ville Timonen ,  Johannes Rajala ,  Emre Aksu

IPC: H04N13/161 ,  H04L29/06 ,  H04N13/15 ,  H04N13/111 ,  H04N13/282

Abstract: The invention relates to a method and technical equipment for implementing the method. The method comprises generating a three-dimensional segment of a scene of a content; generating more than one two-dimensional views of the three-dimensional segment, each two-dimensional view representing a virtual camera view; generating multi-view streams by encoding each of the two-dimensional views; encoding parameters of a virtual camera to the respective stream of the multi-view stream; receiving a selection of one or more streams of the multi-view stream; and streaming only the selected one or more streams.

公开(公告)号：US12242969B2

公开(公告)日：2025-03-04

申请号：US17354398

申请日：2021-06-22

Applicant: Nokia Technologies Oy

Inventor： Honglei Zhang ,  Francesco Cricri ,  Hamed Rezazadegan Tavakoli ,  Joachim Wabnig ,  Iraj Saniee ,  Miska Matias Hannuksela ,  Emre Aksu

IPC: G06N3/08 ,  G06N3/04 ,  G06N3/082

Abstract: An apparatus includes at least one processor; and at least one non-transitory memory including 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: estimate an importance of parameters of a neural network based on a graph diffusion process over at least one layer of the neural network; determine the parameters of the neural network that are suitable for pruning or sparsification; remove neurons of the neural network to prune or sparsify the neural network; and provide at least one syntax element for signaling the pruned or sparsified neural network over a communication channel, wherein the at least one syntax element comprises at least one neural network representation syntax element.