Abstract: Methods, apparatus, systems and articles of manufacture to generate packed video frames are disclosed. A video encoding system disclosed herein includes a configuration determiner to create a packed video frame layout that includes regions into which video components are to be placed. The system also includes a frame generator to form packed video frames that include the video components placed into different regions. The encoding system further includes a frame information generator that generates packed video frame information that identifies characteristics of the packed video frame including (i) the identities of regions included in the packed video frame layout, (ii) types of video components included in the regions, or iii) information identifying the locations and dimensions of the regions. A video encoder of the encoding system encodes the frames and includes the packed video frame information to signal the inclusion of the packed video frames in the encoded bitstream.

Abstract: One embodiment provides for a graphics processor comprising a block of graphics compute units, a graphics processor pipeline coupled to the block of graphics compute units, and a programmable neural network unit including one or more neural network hardware blocks. The programmable neural network unit is coupled with the block of graphics compute units and the graphics processor pipeline. The one or more neural network hardware blocks include hardware to perform neural network operations and activation operations for a layer of a neural network. The programmable neural network unit can configure settings of one or more hardware blocks within the graphics processor pipeline based on a machine learning model trained to optimize performance of a set of workloads.

Abstract: Embodiments described herein provided for an instruction and associated logic to enable a processing resource including a tensor accelerator to perform optimized computation of sparse submatrix operations. One embodiment provides hardware logic to apply a numerical transform to matrix data to increase the sparsity of the data. Increasing the sparsity may result in a higher compression ratio when the matrix data is compressed.

Abstract: Methods, apparatus, systems and articles of manufacture to sample enabled views per atlas in immersive video are disclosed. An example apparatus includes an interface to obtain a bitstream corresponding to an immersive video; a bitstream analyzer to determine that metadata corresponding to an atlas of the bitstream identifies that at least part of a view of interest corresponds to the atlas, the metadata included in the bitstream; and a filter to generate a filtered bitstream by removing atlases that do not identify at least part of a view of interest from the bitstream, the interface to provide the filtered bitstream to a decoder.

Abstract: Embodiments are generally directed to transport controlled video coding. An embodiment of an apparatus includes one or more processors to process data; a memory to store data, including data for video streaming; and a video processing mechanism including an encoder and a transport mechanism, wherein the video processing mechanism is to generate a prediction of channel throughput for a network channel, encode one or more bitstreams based on the prediction, including encoding a plurality of bitstreams including a first bitstream and a second bitstream if the prediction indicates an increase or decrease in channel throughput and encoding a single bitstream if the prediction indicates a stable channel throughput; and select a bitstream of the one or more bitstreams for a current frame.

Abstract: Devices and techniques related to implementing patch based video coding for machines are discussed. Such patch based video coding includes detecting regions of interest in a frame of video, extracting the detected regions of interest to one or more atlases absent the frame at a resolution not less than the resolution of the regions of interest, and encoding the one or more atlases to a bitstream.

Abstract: Embodiments are directed to neural network processing for multi-object three-dimensional (3D) modeling. An embodiment of a computer-readable storage medium includes executable computer program instructions for obtaining data from multiple cameras, the data including multiple images, and generating a 3D model for 3D imaging based at least in part on the data from the cameras, wherein generating the 3D model includes one or more of performing processing with a first neural network to determine temporal direction based at least in part on motion of one or more objects identified in an image of the multiple images or performing processing with a second neural network to determine semantic content information for an image of the multiple images.

Abstract: Methods, apparatus, systems and articles of manufacture for video coding using object metadata are disclosed. An example apparatus includes an object separator to separate input views into layers associated with respective objects to generate object layers for geometry data and texture data of the input views, a pruner to project the first object layer of a first basic view of the at least one basic views against the first object layer of a first additional view of the at least one additional views to generate a first pruned view and a first pruning mask, a patch packer to tag a patch with an object identifier of the first object, the patch corresponding to the first pruning mask, and an atlas generator to generate at least one atlas to include in encoded video data, the atlas including the patch.

Abstract: An example apparatus for adaptively encoding video frames includes a network analyzer to predict an instant bitrate based on channel throughput feedback received from a network. The apparatus also includes a content analyzer to generate ladder info based on a received frame. The apparatus further includes an adaptive decision executer to determine a frame rate, a video resolution, and a target frame size based on the predicted instant bitrate and the ladder outputs. The apparatus further includes an encoder to encode the frame based on the frame rate, the video resolution, and the target frame size.

Abstract: Methods, apparatus, systems and articles of manufacture for object-based volumetric video coding are disclosed. An example apparatus disclosed herein includes a point annotator to receive point cloud data associated with an object and annotate points of the point cloud data with an object identifier of the object. The disclosed example apparatus also includes a projector to project the point cloud data onto projection planes to produce texture images and geometry images. The disclosed example apparatus further includes a patch generator to generate a patch based on the object identifier, the patch including the texture images and the geometry images of the object, the patch associated with the object identifier of the object. The disclosed example apparatus also includes an atlas generator to generate an atlas to include in encoded video data, the atlas including the patch.

Abstract: A mechanism is described for facilitating monochrome bit depth coding. The apparatus comprises one or more processors to receive depth data and convert the depth data into color space data, including converting a first segment of the depth data into a monochrome component of the color space data and converting a second segment of the depth data into one or more color components of the color space data.

Abstract: Embodiments described herein provide for techniques to reduce the complexity of rendering immersive 3D video content. One embodiment provides for an apparatus comprising one or more processors to receive a data set that represents a two-dimensional encoding of planar projections of a frame of a three-dimensional video, decode the two-dimensional encoding into texture data, geometry data, and metadata, determine, based on the metadata, a visibility status and an occupancy status for a sample position in the three-dimensional video, and render video data for the sample position when the sample position is visible and occupied.

Abstract: A mechanism is described for facilitating adaptive resolution and viewpoint-prediction for immersive media in computing environments. An apparatus of embodiments, as described herein, includes one or more processors to receive viewing positions associated with a user with respect to a display, and analyze relevance of media contents based on the viewing positions, where the media content includes immersive videos of scenes captured by one or more cameras. The one or more processors are further to predict portions of the media contents as relevant portions based on the viewing positions and transmit the relevant portions to be rendered and displayed.

Abstract: Embodiments of a video codec may include technology to derive a conformance point for a sub-region of coded pictures in a coded video sequence in the video data, group any combination of sub-pictures that form a rectangular region into a sub-picture set, and/or derive a level indicator corresponding to a sub-picture based on a level of the coded video sequence and a relative size of the coded picture and the sub-picture set. Other embodiments are disclosed and claimed.

Abstract: Techniques related to immersive video coding are discussed and include immersive video sequences and output units for random access to the immersive video, coding improvements for camera parameters coding, and coding efficiency improvements for atlas parameters.

Abstract: An embodiment of a video codec may include technology to store video data, and utilize a generalized bypass bin to one or more of encode and decode the video data. Other embodiments are disclosed and claimed.

Abstract: An apparatus to facilitate encoding of point cloud data is disclosed. The apparatus includes one or more processors to receive point cloud data including a plurality of images and camera parameters, generate encoded point cloud data including a color texture image and a depth image having cropped regions for each of the plurality of images, and metadata to describe the camera parameters and a mapping of the cropped regions to a real coordinate space.

Abstract: Embodiments are generally directed to affine motion compensation for current picture referencing. An embodiment of an apparatus includes one or more processors for processing of data; a memory for storage of data including video data; and an encoder for encoding of video data to generate encoded video data, wherein the encoder includes a component to provide affine motion compensation for current picture references in the video data.

Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed to encode and decode video using quantization matrices. An example apparatus includes interface circuitry to access an input frame of video, quantization matrix syntax encoder circuitry to encode a set of user-defined quantization matrices into a sequence header associated with a sequence of video frames including the input frame, adaptive quantization matrix selector circuitry to select a subset of quantization matrices from a combination of a set of default quantization matrices and the set of user-defined quantization matrices, adaptive segment selector circuitry to select a first one of the subset of quantization matrices for a first segment of the input frame, the input frame to be divided into a plurality of segments including the first segment, and encoder circuitry to quantize transform coefficients of the first segment of the input frame based on the first one of the subset of quantization matrices.

Abstract: Disclosed examples include video frame segmenter circuitry to generate segmentation data of first video frame pixel data, the segmentation data including metadata corresponding to a foreground region and a background region, the foreground region corresponding to the first video frame pixel data. The disclosed examples also include video encoder circuitry to generate a first foreground bounding region and a first background bounding region based on the segmentation data, determine a first virtual tile of the first video frame pixel data, the first virtual tile located in the first foreground bounding region, encode the first virtual tile into a video data bitstream without encoding the first background bounding region, and transmit the video data bitstream via a network.