Wide-Angle Image Thumbnail Generation via Object Orientation Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wide-angle images captured by imaging devices often suffer from distortion, making it difficult for viewers to understand the photographer's intended subject or location, as existing techniques do not effectively convey this information in thumbnail images.
Innovation Solution
An electronic device and method that acquires a wide-angle image and controls the display to show a partial range of the image in the direction a specific object is facing, allowing viewers to understand the photographer's intended capture by determining the object's frontal direction and averaging or weighting these directions for thumbnail image generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a partial region of a wide-angle image is cut out and displayed as a thumbnail image, then the thumbnail image becomes easier to view and less distorted, but the viewer cannot understand what the photographer intended to capture
Solution Approach 1:
The patent segments the wide-angle image by identifying multiple candidate regions (first candidate region containing the photographer, second candidate region containing the intended subject) and selects the most appropriate region for the thumbnail based on segmentation criteria. This allows the thumbnail to display a focused, undistorted view while preserving information about the photographer's intent through intelligent region selection.
Solution Approach 2:
The patent introduces an intermediary processing step that analyzes the wide-angle image to determine the photographer's intended subject. By using image processing techniques to identify candidate regions and select the most representative one, the system acts as an intermediary between the raw wide-angle image and the final thumbnail, ensuring the thumbnail conveys the photographer's intent while maintaining visual clarity.
2Quantity of substance
If the thumbnail image shows the photographer as the subject, then the photographer appears in the thumbnail, but the viewer fails to grasp the shooting location and intended subject
Solution Approach 1:
The patent divides the image into distinct candidate regions: one containing the photographer and another containing the intended subject. By segmenting these regions and evaluating them against selection criteria, the system can choose to display the region containing the intended subject rather than the photographer, thereby preserving information about what was captured while maintaining thumbnail readability.
Solution Approach 2:
The patent applies different quality criteria to different regions of the image. Instead of uniformly selecting based on photographer presence, the system evaluates regions based on their ability to represent the photographer's intent. The selection process considers local qualities of each candidate region (such as subject importance, composition, and relevance to the captured scene) to determine the optimal thumbnail content.
Data Source
AI summary
An electronic device acquires a wide-angle image. In a case when the wide-angle image includes a specific object, control is performed so as to display, on a screen, a partial range of the wide-angle image, being a range of a direction in which the specific object is facing, in the wide-angle image.