Apple Inc.

Ati Technologies Ulc

Advanced Micro Devices Corporation

Fittingbox

Khaled Mammou

Alexandros Michael Tourapis

Fabrice A Robinet

Jungsun Kim

Yeping Su

Valery G Valentin

David William Singer

Gabor Sines

Lei Zhang

Hsi-Jung Wu

David Flynn

Ihab M A Amer

Ihab Amer

Layla A Mah

David I Glen

Edward A Harold

Bruce Montag

Haibo Liu

Rajabali M Koduri

Christophe Dehais

Fabio Gulino

Andrea Cremaschi

Michael L Schmit

Maneli Noorkami

Ariel Choukroun

Daniel Wong

Afshin Taghavi Nasrabadi

Ahmed M Abdelkhalek

Sylvain Le Gallou

Benedict Chien

Ho Hin Lau

Ehsan Mirhadi

Zhenzhen Gao

Marius Preda

Samuel D Naffziger

Guy Cote

Jim Chen Chou

Syed Athar Hussain

Mehdi Saeedi

Guennadi Riguer

Arun Sundaresan Iyer

Ying Luo

Andy Sung

Dominik Mehlem

Ping Chen

Arash Hariri

Zhiqi Hao

Deepak S Tolani

Alexandras Tourapis

Vladyslav S Zakharchenko

Dmytro U Elperin

Stephen A J Bagshaw

Lawrence A Bair

Naveed A Gazi

Steven Lok-Man Doo

Jonathon Walter Riley

John-Paul A Compagnone

Li Baochun

Gerald S C Chan

Chao Cao

Liviu Traian Christian Tulvan

Disclosed herein is a region-based reference management system using in video frame encoding. Source content, such as video game streaming or remote desktop sharing, that includes scene changes or significant instantaneous changes in a region from one frame to the next can present encoding challenges. Techniques disclosed herein use hints about changes in regional frame content, dissect frame content into regions, and associate the dissected regions with stored reference frame data using the hints and information about the regions to more efficiently encode frames.

Region-based reference frame management

An example method includes receiving a plurality of points that represent a point cloud; representing a position of the point in each dimension of a three-dimensional space as a sequence of bits, where the position of the point is encoded according to a tree data structure; partitioning at least one of the sequences of bits into a first portion of bits and a second portion of bits; quantizing each of the second portions of bits according to a quantization step size, where the quantization step size is determined according to an exponential function having a quantization parameter value as an input and the quantization step size as an output; and generating a data structure representing the point cloud and including the quantized second portions of bits.

In-tree geometry quantization of point clouds

A decoding computing device receives a bit stream for compressed 3D volumetric content. The bit stream includes video encoded image frames comprising packed attribute patch images and depth maps for the 3D volumetric content. Instead of generating a mesh having a vertex for each depth value signaled in the depth map, the decoder performs a real-time mesh simplification process to reduce a resolution of the mesh, such that the mesh resolution is reduced without exceeding an error threshold, which may be dynamically determined. Additionally, the decoder may perform a re-meshing of particular regions of the mesh for the 3D volumetric content to avoid cracks or gaps.

Real time simplification of meshes

A system receives encoded data regarding a points in a point cloud. The data includes a prediction tree having a nodes generated based on spatial information regarding the points and properties of a sensor system that obtained the spatial information. A value of each node represents first spatial coordinates of a respective one of the points according to a first coordinate system, and the value of at least a first node in the prediction tree is determined based on ancestor nodes of the first node and the properties of the sensor system. The system decodes the data to determine first data, including the first spatial coordinates of at least some of the points, and quantization parameters associated with the first spatial coordinates. The system determines second data based on the first data, including second spatial coordinates of at least some of the points according to a second coordinate system.

Predictive coding for point cloud compression

A device and method for processing Virtual Reality (VR) data is disclosed. The method comprises transmitting feedback information from the device to a server, wherein the feedback information is captured in the device, receiving data from the server to be presented on the device based on the feedback information, wherein the data includes video data and audio data where the video data is a frame of video data in a sequence of frames and the audio data is the corresponding audio data of the frame, decoding the video data and corresponding audio data of the frame, and controlling the presentation of the video data and corresponding audio data on the device such that the video data is synchronized with the corresponding audio data.

Low latency wireless virtual reality systems and methods

An encoder encodes three-dimensional (3D) volumetric content, such as immersive media, using video encoded attribute patch images packed into a 2D atlas to communicate the attribute values for the 3D volumetric content. The encoder also uses mesh-encoded sub-meshes to communicate geometry information for portions of the 3D object or scene corresponding to the attribute patch images packed into the 2D atlas. The encoder applies decimation operations to the sub-meshes to simplify the sub-meshes before mesh encoding the sub-meshes. A distortion analysis is performed to bound the level to which the sub-meshes are simplified at the encoder. Mesh simplification at the encoder reduces the number of vertices and edges included in the sub-meshes which simplifies rendering at a decoder receiving the encoded 3D volumetric content.

3D volumetric content encoding using 2D videos and simplified 3D meshes

A system comprises a prediction module configured to predict an attribute value, such as a texture coordinate, for a vertex of a triangle in a two-dimensional (2D) attribute representation based on known positions of vertices of a corresponding triangle in a three-dimensional (3D) geometric representation. In some embodiments, the prediction module adaptively selects a prediction technique between multiple available prediction techniques based on availability of vertices information in the 3D geometric representation and in the 2D attribute representation and further based on compression efficiency and distortion minimization. The prediction module enables compression of attribute information being signaled for volumetric visual content, such as a mesh with texture.

Attribute value compression for a three-dimensional mesh using geometry information to guide prediction

An encoder is configured to compress point cloud information using a blocks of nodes determined from a prediction tree. A prediction tree is generated for a point cloud. Segments of the prediction tree are identified. The segments are divided into blocks that are predicted by predecessor blocks within the segments. The blocks of the prediction tree may then be encoded and may be provided for transmission to a decoder that can regenerate the point cloud from the blocks of the prediction tree.

Block-based predictive coding for point cloud compression

A system compresses and decompresses attribute information for visual volumetric content, such as a mesh representation. Attribute values are included in the visual volumetric representation, wherein at least some of the attribute values include unitary vectors, such as surface normal vectors or surface tangent vectors having a magnitude of one unit. In order to compress the attribute information the three-dimensional unit vectors are mapped into two dimensional parametric coordinates for a planar representation of a unit sphere. To reduce negative effects on compression due to distortion or discontinuities in the planar representation, mappings for compressing respective unit vectors are adaptively selected.

Compression of attribute values comprising unit vectors

An encoder is configured to compress point cloud geometry information using an octree geometric compression technique that utilizes a binary arithmetic encoder, a look-ahead table, a cache, and a context selection process, wherein encoding contexts are selected based, at least in part, on neighborhood configurations. In a similar manner, a decoder is configured to decode compressed point cloud geometry information utilizing a binary arithmetic encoder, a look-ahead table, a cache, and a context selection process.

Point cloud geometry compression using octrees and binary arithmetic encoding with adaptive look-up tables

A system comprises an encoder configured to compress attribute information and/or spatial information for three-dimensional (3D) visual volumetric content and/or a decoder configured to decompress compressed attribute and/or spatial information for the 3D visual volumetric content. The encoder is configured to convert 3D visual volumetric content, such as a point cloud or mesh, into image based patch representations. The encoder is further configured to select one or more reference patches for copying or prediction, such that metadata for copying or predicting a patch based on the reference patch is signaled without explicitly signaling a full set of information for the copied or predicted patch. Likewise, a decoder is configured to receive such information and reconstruct a 3D version of the 3D visual volumetric content using both signaled and predicted or copied patches.

Video-based point cloud compression with predicted patches

A system comprises an encoder configured to compress and encode data for a three-dimensional mesh using a video encoding technique. To compress the three-dimensional mesh, the encoder determines sub-meshes and for each sub-mesh: texture patches and geometry patches. Also the encoder determines patch connectivity information and patch texture coordinates for the texture patches and geometry patches. The texture patches and geometry patches are packed into video image frames and encoded using a video codec. Additionally, the encoder determines boundary stitching information for the sub-meshes. A decoder receives a bit stream as generated by the encoder and reconstructs the three-dimensional mesh.

Three-dimensional mesh compression using a video encoder

A system comprises an encoder configured to compress attribute information for a point cloud and/or a decoder configured to decompress compressed attribute information. Attribute values for at least one starting point are included in a compressed attribute information file and attribute correction values are included in the compressed attribute information file. An order for the points is determined based on a space filling curve, wherein an encoder and a decoder determine a same order for the points based on the space filling curve. Levels of detail are determined by sampling the ordered points according to different sampling parameters, and attribute values are predicted for the points in the levels of detail using the determined order. The encoder determines attribute correction values based on a comparison of the predicted values to an original value prior to compression. The decoder corrects the predicted attribute values based on received attribute correction values.

Point cloud compression using a space filling curve for level of detail generation

A system comprises an encoder configured to compress attribute information and/or spatial information for a point cloud and/or a decoder configured to decompress compressed attribute and/or spatial information for the point cloud. The encoder is configured to convert a point cloud into an image based representation. The encoder packs patch images into an image frame and fills empty spaces in the image frame with a padding. Various compression strategies may be used to encode an occupancy map and related block-to-patch information indicating which portions of the image frame correspond to which packed patches. Packed image frames comprising patches and padding, along with an encoded occupancy map and related block-to-patch information are sent to a decoder. The decoder is configured to generate a decompressed point cloud based on the packed image frames, compressed occupancy map, and related block-to-patch information.

Occupancy map block-to-patch information compression

Growing community of inventors

Vancouver, Canada

Khaled Mammou