Main Object Determination via Feature Point Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for determining a main object in images with multiple objects impose a high processing load on image capturing apparatuses, making it difficult to accurately identify the object that matches the user's intention.
Innovation Solution
A main object determination apparatus that acquires images at different timings, selects candidates based on feature points, and determines if they are the same using feature amount information, reducing processing load while ensuring accurate identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If simultaneous tracking of multiple objects using recurrent neural network is implemented, then object detection capability is improved, but processing load increases significantly
Solution Approach 1:
The patent segments the object detection task by first identifying candidate objects using simplified criteria (object recognition unit), then selectively applying complex tracking algorithms only to these candidates rather than all detected objects. This segmentation reduces the processing load while maintaining detection accuracy for main objects.
Solution Approach 2:
The patent applies partial action by implementing tracking only for candidate objects that meet specific criteria rather than all detected objects. The system performs sufficient tracking action on selected candidates to determine main objects without the excessive processing required for comprehensive tracking of all objects in the scene.
2Measurement precision
If comprehensive object tracking is performed to accurately determine main object, then detection accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary action by pre-selecting candidate objects based on simple recognition criteria before applying complex tracking algorithms. This preliminary filtering step reduces the number of objects requiring intensive processing, thereby reducing overall processing time while maintaining accuracy for the final main object determination.
Solution Approach 2:
The patent applies different processing qualities to different objects: simple recognition for all detected objects, and more rigorous tracking only for candidate objects. This local differentiation of processing quality reduces overall processing time while ensuring high accuracy for the main object determination.
Data Source
AI summary
A main object determination apparatus includes an acquisition unit configured to acquire images captured at different timings, a selection unit configured to select a candidate of a main object from among objects using information about a feature point of each object in the images, and a determination unit configured to determine whether candidates of the main object selected at the different timings are the same using the information, wherein the main object is determined in a case where the determination unit determines that the candidates of the main object selected by the selection unit in an image of interest and in at least one image captured within a predetermined period of time before the image of interest is captured are the same.