Live Stream Processing with Attribute-Based Virtual Scene Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current live streaming technologies fail to meet the diversified requirements of users, resulting in a lack of personalization and attractiveness in live streaming rooms due to the use of a single virtual scene and image set by streamers.
Innovation Solution
A live stream processing method that classifies audiences based on attribute information and generates multiple virtual scenes and images to enrich content, allowing streamers to create personalized live streams tailored to different user groups.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a single virtual scene and image set is used by streamers, then the live streaming setup is simple, but the live streaming content cannot meet diversified requirements of users
Solution Approach 1:
The patent segments the virtual scene system into multiple independent virtual scenes (first virtual scene, second virtual scene, etc.) that can be selectively switched. Each virtual scene represents a distinct content type or theme, allowing the system to serve different user preferences without requiring streamers to manage a monolithic complex scene graph. This segmentation enables diversified content delivery while keeping each individual scene relatively simple.
Solution Approach 2:
The patent introduces dynamic switching capability between different virtual scenes based on user attributes. The system can transition from a static single-scene configuration to a dynamic multi-scene system where the active virtual scene changes according to real-time user attribute information. This dynamic adaptation allows the same live stream to serve multiple user groups with different preferences without increasing the fundamental simplicity of the streaming setup.
2Adaptability or versatility
If multiple virtual scenes are generated to meet diverse user requirements, then user personalization is improved, but the system complexity increases
Solution Approach 1:
The patent implements a feedback mechanism where user attribute information is continuously obtained and used to determine which virtual scene should be displayed. The system receives user attribute data, processes it to identify matching virtual scenes, and switches accordingly. This feedback loop enables personalized content delivery while using simple attribute matching logic rather than complex AI-based recommendation systems.
Solution Approach 2:
The patent changes the parameter of virtual scene selection based on user attribute information. Instead of using a fixed single scene, the system adjusts the scene selection parameter dynamically according to user attributes such as age, gender, or preferences. This parameter-based approach simplifies the processing logic by reducing it to attribute matching and scene selection, rather than requiring complex content generation or transformation.
3Adaptability or versatility
If real-time processing of user attributes is implemented, then live stream personalization is improved, but processing time increases
Solution Approach 1:
The patent applies preliminary action by pre-configuring multiple virtual scenes with their corresponding attribute matching criteria before the live stream occurs. When user attribute information becomes available, the system only needs to match against pre-defined criteria rather than performing complex real-time analysis. This preparation phase shifts the computational workload from real-time processing to offline configuration, significantly reducing processing delay while maintaining real-time adaptability.
Data Source
AI summary
The present application provides techniques for generating live streams. The techniques comprise obtaining motion attribute information and a set of virtual scene types associated with a target live streaming room; generating, based on the motion attribute information and each of the set of virtual scene types, a plurality of live streams to be played that correspond to the set of virtual scene types; receiving a live stream viewing request sent from an audience client device, where the live stream viewing request comprises a user attribute identifier; determining a target live stream among the plurality of live streams based on the user attribute identifier; and returning, to the audience client device, the target live stream for display on the audience client device.