Back to results
US 10,873,737 B1Granted
VR device and control method for the same
AssigneeLG Electronics Inc.
InventorsKyungyong Kim, Hongsuk Jeong
Life of this patent
Estimating the term…
Abstract
A control method for a VR device is disclosed, which comprises the steps of receiving 360-degree video bitstreams, extracting sub-bitstreams included in the 360-degree video bitstreams, the sub-bitstreams including a Region of Interest (ROI) for the 360-degree video bitstreams, generating restoration data for the ROI from sub-bitstreams, and generating thumbnail data by remapping the restoration data for the ROI based on signaling information.
Claims18
A method of controlling a Virtual Reality (VR) device, the method comprising
- receiving a 360 video bitstream;
- extracting a sub-bitstream included in the 360 video bitstream,
- wherein the sub-bitstream includes Region of Interest (ROI) for the 360 video bitstream;
- decoding the 360 video bitstream in accordance with a scalable coding scheme;
- performing inter-layer prediction (ILP) for the 360 video bitstream;
- performing inter-prediction (IP) for blocks of the 360 video bitstream;
- generating restoration data for the ROI from the sub-bitstream;
- and
- generating thumbnail data by remapping the restoration data for the ROI based on signaling information;
- wherein the ROI is a region for which down-sampling is not performed or a region which is less down-sampled during region-wise packing,
- wherein the signaling information includes a region-wise packing parameter representing information related to the down-sampling,
- wherein the restoration data for the ROI is generated based on the region-wise packing parameter, and
- wherein the region-wise packing parameter packs high-priority part of the ROI into a larger image and rest of the ROI into a smaller image,
- wherein the ILP indicates prediction of higher spatial resolution or quality signal by means of low spatial resolution or quality signal,
- wherein the IP indicates prediction from decoded samples of a current decoded picture and reference pictures,
- wherein the IP includes temporal inter-prediction,
- wherein Motion Constrained Tile Set (MCTS) restricts a range of temporal inter-prediction to a current tile internal range.
14 more elements
A Virtual Reality (VR) device, the device comprising
- a receiver configured to receive a 360 video bitstream;
- a bitstream extractor configured to extract a sub-bitstream included in the 360 video bitstream,
- wherein the sub-bitstream includes Region of Interest (ROI) for the 360 video bitstream;
- a decoder configured to decode the 360 video bitstream in accordance with a scalable coding scheme;
- an inter-layer predictor configured to perform inter-layer prediction (ILP) for the 360 video bitstream;
- an inter-predictor to perform inter-prediction (IP) for blocks of the 360 video bitstream;
- a thumbnail generator configured to generate restoration data for the ROI from the sub-bitstream;
- and
- a remapper configured to generate thumbnail data by remapping the restoration data for the ROI based on signaling information;
- wherein the ROI is a region for which down-sampling is not performed or a region which is less down-sampled during region-wise packing,
- wherein the signaling information includes a region-wise packing parameter representing information related to the down-sampling,
- wherein the restoration data for the ROI is generated based on the region-wise packing parameter, and
- wherein the region-wise packing parameter packs high-priority part of the ROI into a larger image and rest of the ROI into a smaller image,
- wherein the ILP indicates prediction of higher spatial resolution or quality signal by means of low spatial resolution or quality signal,
- wherein the IP indicates prediction from decoded samples of a current decoded picture and reference pictures,
- wherein the IP includes temporal inter-prediction,
- wherein Motion Constrained Tile Set (MCTS) restricts a range of temporal inter-prediction to a current tile internal range.
14 more elements
Dependencies are inferred from the claim text.