End-Cloud Collaborative Media Data Processing for Special Effects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for special effect rendering in media data processing, such as in short videos and social media applications, face issues with poor rendering quality and long processing times due to limitations in computing resources on terminal devices and increased load on servers during complex media data processing.
Innovation Solution
An end-cloud collaborative media data processing method that splits the processing of complex special effects between local and remote algorithm nodes, utilizing the terminal device for less resource-intensive tasks and the server for more intensive ones, thereby improving rendering quality and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If complex special effect rendering is performed using only local algorithms on terminal devices, then processing speed is maintained, but rendering quality deteriorates due to limited computing resources
Solution Approach 1:
The patent segments the special effect rendering process into multiple algorithm nodes that can be distributed between terminal devices and servers. Complex rendering tasks are divided into sub-tasks that can be executed locally or remotely based on resource availability and task requirements, thereby improving rendering quality without overwhelming local computing resources.
Solution Approach 2:
The patent introduces a processing flow management system that acts as an intermediary between terminal devices and servers. This mediator coordinates the distribution of rendering tasks, manages algorithm node execution, and combines results to achieve high-quality rendering while optimizing resource utilization across the distributed system.
2Manufacturing precision
If complex special effect rendering is performed using only remote algorithms on servers, then rendering quality improves, but processing time increases due to data transmission and server load
Solution Approach 1:
The rendering process is segmented into multiple algorithm nodes that can be executed in parallel or distributed across terminal devices and servers. By dividing complex rendering tasks into smaller sub-tasks, the system reduces data transmission overhead and allows concurrent processing, thereby maintaining high rendering quality while reducing overall processing time.
Solution Approach 2:
The patent allows terminal devices to execute certain algorithm nodes locally when resources permit, performing partial rendering actions without requiring complete server processing. This partial local execution reduces dependency on server availability and network transmission, decreasing processing time while maintaining acceptable rendering quality.
3Productivity
If separate local or remote algorithms are used for special effect rendering, then system complexity is reduced, but rendering effectiveness deteriorates due to inability to leverage both local and remote resources
Solution Approach 1:
The patent creates a universal processing flow framework that can handle both local and remote algorithm execution through a unified interface. The system manages multiple types of algorithm nodes (local processing nodes, remote processing nodes) within a single cohesive architecture, enabling flexible resource utilization without requiring separate systems for different rendering scenarios.
Solution Approach 2:
The processing flow is designed to be dynamic and adaptive, automatically adjusting the distribution of algorithm nodes between terminal devices and servers based on real-time resource availability, task characteristics, and performance requirements. This dynamic allocation optimizes rendering efficiency while managing system complexity through intelligent decision-making rather than rigid predefined configurations.
Data Source
AI summary
Embodiments of the present disclosure provide an end-cloud collaborative media data processing method and apparatus, a device and a storage medium, where the method includes: calling a target processing flow in response to a first operation triggering a target functionality, where the target processing flow includes a local algorithm node and a remote algorithm node, and the target functionality is used for adding a special effect to media data to be processed; executing, based on the target processing flow, a corresponding local algorithm node and remote algorithm node to obtain first processing data output by the local algorithm node and second processing data output by the remote algorithm node; and generating third processing data through the first processing data and/or the second processing data.