Main Subject Selection Using Face and Head Detection Priority
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems struggle to accurately detect and prioritize a main subject, particularly when the subject's face is obscured by goggles or masks, leading to incorrect exclusion from main subject candidates.
Innovation Solution
An image processing apparatus and method that allows users to set a subject as a main subject candidate based on face or head detection, with options to prioritize subjects detected by either face or head, even if the face is undetectable, and employs focus and exposure control based on these settings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If face detection is used to determine main subject, then subject detection accuracy is improved, but face-undetectable individuals (wearing goggles or masks) are incorrectly excluded from main subject candidates
Solution Approach 1:
The patent segments the face detection task into two independent detection units: one for face detection and another for head detection. This segmentation allows the system to independently evaluate both detection results and make informed decisions about main subject candidates, preventing face-undetectable individuals from being incorrectly excluded.
Solution Approach 2:
The patent adds another detection dimension by introducing head detection alongside face detection. Instead of relying solely on face detection, the system now considers both face and head detection results, expanding the detection scope to include individuals wearing goggles or masks.
2Adaptability or versatility
If head detection is added as an alternative to face detection, then coverage of face-undetectable individuals is improved, but system complexity increases
Solution Approach 1:
The determination unit serves multiple functions by integrating both face detection and head detection results. It universally applies a unified decision-making process that considers both detection types, making the system adaptable to various scenarios including face-detectable and face-undetectable individuals without requiring separate processing paths.
Solution Approach 2:
The patent merges face detection and head detection into a single integrated determination process. By combining the results from both detection units and applying a unified selection criteria, the system achieves enhanced detection coverage while managing complexity through integration rather than separate processing streams.
Data Source
AI summary
In a case where a first setting is set, a determination unit determines a main subject region while setting, as a main subject, a subject of which a first subject has been detected by a first detection unit, in priority to a subject of which the first subject has not been detected by the first detection unit, and in a case where a second setting is set, the determination unit determines a main subject region while setting, as a main subject, a subject of which the first subject has not been detected by the first detection unit, and of which the second subject has been detected by a second detection unit, in priority to a subject of which the first and second subjects have not been detected.