Split XR Rendering with Per-Object 5G Streaming QoS
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in efficiently processing extended reality (XR) data, particularly in managing dynamic virtual objects with varying quality of service (QoS) and charging requirements in streaming sessions.
Innovation Solution
A method and device for processing XR data involves initializing streaming sessions for each dynamic virtual object, configuring QoS and charging information, and retrieving media data to render XR scenes with dynamic virtual objects, utilizing split rendering across multiple devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple streaming sessions are initialized for each dynamic virtual object, then individual QoS and charging information can be configured for each object, but device complexity and network resource consumption increase
Solution Approach 1:
The patent segments the XR data stream into multiple independent streams, with each stream corresponding to a specific dynamic virtual object. This segmentation enables independent QoS configuration for each object while maintaining manageable complexity through structured organization. The client device creates separate streaming sessions for different object types (e.g., character models, textures, animations), allowing granular control over bandwidth allocation and quality parameters for each segment.
Solution Approach 2:
The patent introduces a new dimension of organization by mapping virtual objects to specific streaming sessions based on object characteristics and QoS requirements. This dimensional mapping allows the system to manage complexity through structured categorization rather than raw numerical increase in sessions. Each dimension (object type, priority level, bandwidth requirement) provides a framework for managing the multiplicity of streams.
2Reliability
If separate streaming sessions are used for each dynamic virtual object, then quality of service can be optimized for each object, but network bandwidth consumption increases
Solution Approach 1:
The patent applies local quality by assigning different QoS parameters to different virtual objects based on their specific requirements. Critical objects like character models receive higher priority and bandwidth allocation, while less critical objects like background elements receive lower allocation. This localized quality adjustment optimizes overall system performance without uniformly increasing bandwidth consumption across all streams.
Solution Approach 2:
The patent dynamically adjusts streaming parameters (bitrate, resolution, frame rate) for each virtual object based on network conditions and object priority. When network bandwidth is constrained, the system modifies parameters for lower-priority objects while maintaining high-quality streams for critical objects. This parameter adaptation allows QoS guarantees for important objects without proportionally increasing total bandwidth consumption.
3Manufacturing precision
If media data is retrieved for each dynamic virtual object through separate sessions, then rendering quality improves, but data retrieval complexity and processing time increase
Solution Approach 1:
The patent implements preliminary action by pre-fetching and buffering media data for virtual objects before they are needed in the XR scene. The system anticipates which objects will be required based on scene configuration and user position, initiating data retrieval in advance. This preliminary data preparation reduces actual retrieval time during rendering while maintaining high quality through complete data availability.
Solution Approach 2:
The patent maintains continuous data retrieval operations across multiple streaming sessions, ensuring that data flow for each virtual object is uninterrupted. By establishing persistent connections and continuous streaming for each object type, the system avoids stop-start retrieval patterns that would increase total processing time. The parallel continuous operations optimize throughput while maintaining rendering quality.
Data Source
AI summary
An example device for processing extended reality (XR) data includes a processors configured to: parse entry point data of an XR scene to extract information about one or more required virtual objects for the XR scene, the required virtual objects including a number of dynamic virtual objects equal to or greater than one, each of the dynamic virtual objects including at least one dynamic media component for which media data is to be retrieved; initialize a number of streaming sessions equal to or greater than the number of dynamic virtual objects using the entry point data; configure quality of service (QoS) and charging information for the streaming sessions; retrieve media data for the dynamic virtual objects via the streaming sessions; and send the retrieved media data to a rendering unit to render the XR scene to include the retrieved media data at corresponding locations within the XR scene.