Gaze-Based Main Object Selection for Multi-Object Image Pickup
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image pickup apparatuses struggle to select a main object when multiple objects are near the gaze point, as they lack a systematic approach to determine the intended object based on the user's visual line.
Innovation Solution
A control apparatus and method that detects a user's gaze point, acquires feature areas, sets selection determining areas, and selects a main object area based on the gaze point's inclusion in these areas, using information on the feature areas to enhance object selection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the gaze point is used to select the main object area, then the object selection follows the user's visual intention, but multiple objects near the gaze point cause selection ambiguity
Solution Approach 1:
The patent segments the selection determination into two distinct areas: the feature area (detection target) and the selection determining area (selection criterion). By dividing these functional zones, the system can accurately distinguish between multiple objects near the gaze point, selecting based on which object's selection determining area contains the gaze point while using the feature area for precise object identification.
Solution Approach 2:
The patent applies different quality criteria to different areas: the feature area uses detection accuracy and object characteristics for identification, while the selection determining area uses gaze point inclusion for selection. This local differentiation allows the system to handle multiple objects near the gaze point by evaluating each object's specific area properties rather than using a uniform selection criterion.
2Ease of operation
If the object area is enlarged to assist selection of small objects, then small objects become easier to select, but the selection determining area may include unintended objects
Solution Approach 1:
The patent separates the feature area (which may be enlarged for small objects to ensure detection) from the selection determining area (which maintains precise boundaries for accurate selection). This segmentation allows small objects to be detected and selected by having their selection determining area contain the gaze point, while the enlarged feature area simply aids in object identification without causing selection of unintended objects.
Data Source
AI summary
A control apparatus includes a processor configured to detect a gaze point of a user on an image based on a visual line, acquire at least one feature area detected using the image, set at least one selection determining area each corresponding to one of the one feature area, select as a main object area a first feature area among the at least one feature area when the gaze point is included in the first feature area, select as the main object area a second feature area corresponding to one selection determining area including the gaze point in a case where the gaze point is not included in the at least one feature area and the gaze point is included in the at least one selection determining area, and set a condition in selecting the main object area using information on the at least one feature area.