Unified Object Processing Model for Diverse Media Types
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies for processing various media types, such as text, audio, and video, require separate models for different types, leading to high deployment costs and inefficiencies due to the need for multiple models and increased complexity.
Innovation Solution
A unified object processing method and apparatus that utilizes a pre-generated target model with a feature extraction module and multiple object and task processing modules, allowing for the processing of different object types and task types through a single model framework, which includes object segmentation and task processing modules tailored to specific types, facilitating efficient model training and deployment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If separate models are used for different media types (text, audio, video), then processing accuracy for each specific media type can be maintained, but deployment cost and system complexity increase significantly
Solution Approach 1:
The patent implements a unified processing model that can handle multiple media types (text, audio, video) through a single system. The model uses modular processing modules that can be dynamically selected based on the input media type, allowing one model to perform multiple functions that previously required separate models for each media type.
Solution Approach 2:
The unified model is divided into independent processing modules, each specialized for specific media types or tasks. This segmentation allows the system to maintain specialized processing capabilities for different media types while consolidating them into a single deployable model, reducing overall system complexity.
2Reliability
If separate models are used for different media types, then each model can be optimized for its specific media type, but deployment cost increases due to multiple models
Solution Approach 1:
The patent merges multiple separate processing models into a single unified model. By combining the processing capabilities for text, audio, and video into one model, the system reduces deployment costs associated with hosting and managing multiple separate models while maintaining the specialized processing accuracy for each media type through modular design.
3Reliability
If multiple separate models are deployed for different tasks and media types, then processing accuracy is maintained, but processing efficiency decreases due to model switching and increased complexity
Solution Approach 1:
The unified model incorporates dynamic module selection capabilities that automatically choose the appropriate processing module based on the input media type and task requirements. This dynamic adaptation eliminates the overhead of switching between multiple separate models while maintaining specialized processing accuracy, thereby improving overall processing efficiency.
Data Source
Figure 1
Figure 2~4
Figure 5~6
AI summary
Embodiments of the present disclosure relate to an object processing method, apparatus, readable medium and electronic device. The method includes: obtaining, (S101), a target object to be processed; determining, (5102), an object type of the target object; determining, (S103), a task type corresponding to a target task for processing the target object; and inputting, (S104), the target object, the object type and the task type into a pre-generated target model to obtain a target result output by the target model. The target model can include a feature extraction module, a plurality of object segmentation modules and a plurality of task processing modules, different object segmentation modules correspond to different object types, and different task processing modules correspond to different task types.