Neural Network Object Processing Order Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current neural network systems face inefficiencies in data processing due to the bottleneck phenomenon caused by consecutive object recognition operations, particularly in scenarios like autonomous driving, where precise and timely object recognition is critical.
Innovation Solution
A neural network system that detects and prioritizes object candidates based on metadata, setting processing orders to perform object recognition and corresponding tasks in units of objects, thereby preventing bottlenecks and ensuring efficient data processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If object recognition operations are performed consecutively for all object candidates, then complete object recognition is achieved, but processing time increases and bottleneck phenomenon occurs
Solution Approach 1:
The patent segments the object recognition process into individual object-level tasks that can be executed independently. Each object candidate is processed separately with its own recognition task, allowing parallel execution and eliminating the bottleneck of sequential processing. This segmentation enables the system to maintain complete recognition while reducing overall processing time through concurrent operations.
Solution Approach 2:
The patent implements preliminary action by pre-processing object candidates and preparing their recognition tasks before actual execution. The system prepares object candidate data, metadata, and recognition parameters in advance, then executes recognition operations in a streamlined manner. This preliminary preparation reduces the time required during actual recognition execution and prevents bottlenecks.
2Ease of operation
If all object candidates are processed with equal priority, then processing simplicity is maintained, but high-importance objects are not handled preferentially
Solution Approach 1:
The patent applies local quality by assigning different processing priorities to different object candidates based on their importance. Instead of uniform treatment, the system evaluates metadata of each object candidate (such as spatial position, object type, or contextual relevance) and assigns appropriate priority levels. High-importance objects receive preferential processing, while less critical objects are processed accordingly. This localized differentiation maintains operational simplicity while significantly improving productivity for critical objects.
Data Source
AI summary
A neural network system, includes: a processor configured to detect a plurality of object candidates included in a first image, generate metadata corresponding to the plurality of object candidates based on the first image, and set data processing orders of the plurality of object candidates based on the metadata; and at least one resource configured to perform data processing with respect to the plurality of object candidates. The processor is configured to sequentially provide pieces of information related to data processing of the plurality of object candidates to the at least one resource according to the set data processing orders, and the at least one resource is configured to sequentially perform data processing with respect to the plurality of object candidates according to an order in which a piece of information related to data processing of each of the plurality of object candidates is received.