Abstract: The disclosure relates to a method comprising quantizing parameters of an input tensor, said quantizing using a codebook whose size is obtained according to a distortion value determined between the at least one tensor and a quantized version of said at least one tensor. The disclosure also relates to a method for quantizing parameters of the input tensor using a pdf-based initialization bounded according to at least one first pdf factor, said first pdf factor being selected among several candidate bounding pdf factors according to resulting entropy. The disclosure also relates to corresponding signal; bitstream, storage media and encoder and/or decoder devices.

Abstract: The disclosure relates to a method comprising, responsive to a determination that at least one first tensor of at least one layer of at least one Deep Neural Network is decomposed into a second tensor and a third tensor whose parameters are encoded in a bitstream, decoding from the bitstream a size of at least one of the second tensor and the third tensor, and decoding the at least one of the second tensor and the third tensor from the bitstream based on the decoded size. Corresponding apparatus, encoding method, signal; bitstream, storage media and encoder and/or decoder devices are also provided.

Abstract: The disclosure relates to a method for compression including codebook-based quantization of a data set and corresponding decompression method, signal; bitstream, and encoder and/or decoder device.

Abstract: The disclosure relates, according to a first aspect, to a method for compressing data including encoding at least one information representative of a use, during the compression, of a compressed sparse format. The disclosure relates, according to a second aspect, to a method for decompressing input data comprising obtaining information representative of zero or non-zero values in at least a part of the input data, and using only the non-zero values of the zero or non-zero values for a further processing of the part of the input data, based on the representative information. Corresponding devices, system, non-transitory program product, computer storage medium and signal are also disclosed.

Abstract: The present disclosure relates to a method including encoding a data set in a signal, the encoding comprising quantizing the data set by using a codebook obtained by clustering the data set, the clustering taking account of a probability of appearance of data in the dataset; the probability being bounded to a bounding value. The present disclosure also relates to a method including encoding in a signal a first weight of a layer of a Deep Neural Network, the encoding taking into account an impact of a modification of a second weight on an accuracy of the Deep Neural Network.

Abstract: The present disclosure relates to a method including reshaping a first tensor of weights, by using one or more second tensor having a lower dimension than the first tensor dimension and encoding the second tensor in a signal The present disclosure relates to a method including obtaining a first tensor of weights by reshaping one or more second tensor hav ing a lower dimension than the first tensor dimension, the one or more second tensor being decoded from a signal. The present disclosure further relates to the corresponding dev ices, signal, and computer readable storage media.

Abstract: Different implementations are described, particularly implementations for video encoding and decoding based on linear weighted intra prediction, also called matrix based intra prediction, are presented. Accordingly, for a block being encoded or decoded in linear weighted intra prediction, obtaining intra predicted samples from at least two matrix-vector products between at least two selected weight matrices of reduced size and a set of neighboring reference samples. Advantageously, such arrangement allows to reduce the amount of memory for storing data and to reduce the complexity of the intra prediction samples computation.

Abstract: A method and apparatus for conveying information in a bitstream for deep neural network compression, such as in matrices representing weights, biases and non-linearities, to iteratively compress a pre-trained deep neural network by low displacement rank based approximation of the network layer weight matrices. The low displacement rank approximation allows for replacement of an original layer weight matrices of the pre-trained deep neural network as the sum of small number of structured matrices, allowing compression and low inference complexity. A decoder stage parses a bitstream for inference.

Abstract: A method and an apparatus for performing deep neural network compression use an approximation training set along with information, such as in matrices representing weights, biases and non-linearities, to iteratively compress a pre-trained deep neural network by low displacement rank based approximation of the network layer weight matrices. The low displacement rank approximation allows for replacement of an original layer weight matrices of the pre-trained deep neural network as the sum of a small number of structured matrices, allowing compression and low inference complexity.

Abstract: A method of providing immersive reality content includes determining a field of view intersection between a first field of view from a first device and a second field of view from a second device, generating a first immersive reality content based on the field of view intersection and providing the first immersive reality content. An apparatus for providing immersive reality content includes a processor and at least one memory coupled to the processor, the processor being configured to determine a field of view intersection between a first field of view from a first device and a second field of view from a second device, generate a first immersive reality content based on the field of view intersection and provide the first immersive reality content. A computer-readable storage medium, computer program and a non-transitory article of manufacture are also described.

Abstract: A method of presenting personalized additional information along with displayed content includes accessing a content viewing history and a user profile. Detecting an event by extracting metadata related to content provided for display, the content provided for display related to other content not contained within the content viewing history, and accessing additional information related to the detected event. Additional features include overlaying an object onto the content being displayed on the content display device, and displaying background image information the surrounds the content display device in the view of the AR user. Any or all of the features may be present simultaneously.

Abstract: Apparatus and method for indoor localization involving sampling illumination in a location such as a room of a building, producing a frequency domain analysis of the illumination, comparing the frequency domain analysis to a reference frequency domain analysis associated with a reference location, and providing a notification indicating a result of the comparison such as whether the location of the sampling is the reference location.

Abstract: A method of providing immersive reality content includes determining a field of view intersection between a first field of view from a first device and a second field of view from a second device, generating a first immersive reality content based on the field of view intersection and providing the first immersive reality content. An apparatus for providing immersive reality content includes a processor and at least one memory coupled to the processor, the processor being configured to determine a field of view intersection between a first field of view from a first device and a second field of view from a second device, generate a first immersive reality content based on the field of view intersection and provide the first immersive reality content. A computer-readable storage medium, computer program and a non-transitory article of manufacture are also described.

Abstract: A system for converting image data from standard dynamic range (SDR) format to a high dynamic range (HDR) format involves partitioning input data representing a SDR image into a plurality of portions of data, wherein each portion represents a respective one of a plurality of SDR patches of the SDR image and each of the plurality of SDR patches covers a portion of the SDR image and the set of the plurality of SDR patches fully covers the SDR image; processing each of the plurality of SDR patches through a deep learning autoencoder responsive to a plurality of model weights representing a model of SDR to HDR conversion to produce a respective plurality of estimated HDR patches; and stitching the estimated HDR patches together to form a HDR image version of the SDR image.

Abstract: The present principles generally relate to augmented reality (AR) apparatuses and methods, and in particular, to an exemplary augmented reality system (100) in which content characteristics are used to affect the individual viewing experience of the content. One exemplary embodiment involves a user specified modification of the content by using an augmented reality device to provide a preview (600) for a parent or a guardian of a viewer, or a third-party curator of contents. A modified content (705, 05) may be created by replacing or obscuring the objectionable content or scenes of the one or more of the original contents. The apparatus and method is employed in a system having one or more augmented reality devices (125-1-125-n) such as e.g., one or more pairs of AR glasses. The system may also include a non-AR display screen (191, 192) to display the content to one or more viewers. Accordingly, different forms of the same content may be presented on the different AR glasses and also on the shared screen.

Abstract: The present principles generally relate to augmented reality (AR) apparatuses and methods, and in particular, to an exemplary augmented reality system (100) in which content characteristics are used to affect the individual viewing experience of the content. One exemplary embodiment involves a user specified modification of the content by using an augmented reality device (125-1) to provide a preview for a parent or a guardian of a viewer, or a third-party curator of contents a time period before a potentially objectionable scene is to be shown to other viewers. A modified content (705, 1005) may be created by replacing or obscuring the objectionable content or scenes of the one or more of the original contents. The apparatus and method is employed in a system having one or more augmented reality devices (125-1-125-n) such as e.g., one or more pairs of AR glasses. The system may also include a non-AR display screen (191, 192) to display the content to one or more viewers.

Abstract: A method of presenting personalized additional information along with displayed content includes accessing a content viewing history and a user profile. Detecting an event by extracting metadata related to content provided for display, the content provided for display related to other content not contained within the content viewing history, and accessing additional information related to the detected event. Additional features include overlaying an object onto the content being displayed on the content display device, and displaying background image information the surrounds the content display device in the view of the AR user. Any or all of the features may be present simultaneously.

Abstract: A method to determine a body pose of a user in a virtual reality or augmented reality system includes acquiring sensor data from a plurality of sensors in a garment worn by a user. The sensor data is processed to generate a processed sensor data set, wherein the processed sensor data set is scaled for the size of the user. The processed sensor data set is converted to a pose data set. The pose vector data set is then used by a viewer device to render the body pose of the user.

Abstract: Approaches for clock synchronization in digital video environments. In an embodiment, an encoder/transcoder calculates a ratio between a system clock and a source clock. The source clock is used by a source device to encode or transcode digital video. The system clock is used by the encoder/transcoder. After the encoder/transcoder receives the digital video from the source device, the encoder/transcoder uses the calculated ratio to create a recovered clock. The recovered clock is locked to a frequency of the source clock but not to the phase of the source clock. The encoder/transcoder uses the recovered clock to encode or transfer the digital video received from the source device. The encoder/transcoder ensures that the frequency of the recovered clock does not change faster than a certain rate, e.g., 0.075 Hz/second.

Abstract: Approaches for clock synchronization in digital video environments. In an embodiment, an encoder/transcoder calculates a ratio between a system clock and a source clock. The source clock is used by a source device to encode or transcode digital video. The system clock is used by the encoder/transcoder. After the encoder/transcoder receives the digital video from the source device, the encoder/transcoder uses the calculated ratio to create a recovered clock. The recovered clock is locked to a frequency of the source clock but not to the phase of the source clock. The encoder/transcoder uses the recovered clock to encode or transfer the digital video received from the source device. The encoder/transcoder ensures that the frequency of the recovered clock does not change faster than a certain rate, e.g., 0.075 Hz/second.