Thumbnail Generation Using Face Extraction for Small Displays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
When applying the motion thumbnail function to a video camera, the small liquid crystal panel screen makes it difficult for users to perceive the contents of moving image thumbnails due to their small size.
Innovation Solution
A thumbnail generating apparatus that extracts images from moving images at shorter intervals, calculates a reduction and enlargement ratio to ensure the face display region has the same aspect ratio and size as a predefined display region, and generates moving image thumbnails by reducing or enlarging the images accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If motion thumbnail function is applied to video camera, then user can perceive contents of moving image, but thumbnail size becomes too small to perceive on liquid crystal panel
Solution Approach 1:
The patent extracts the face region from the entire moving image frame, isolating only the relevant portion (face) for thumbnail generation. This extraction allows the face to be displayed at a larger effective size within the thumbnail region while maintaining the ability to perceive moving image contents.
Solution Approach 2:
The patent applies different processing to different regions of the image - specifically, it identifies and prioritizes the face region while reducing or excluding other regions. This local quality approach ensures that the most important visual information (face) is preserved and enhanced in the thumbnail.
2Device complexity
If still image thumbnail is used, then device complexity is reduced, but user cannot perceive contents of moving image
Solution Approach 1:
The patent creates a simplified copy of the moving image that captures essential visual information (face region) in a still image format. This copying approach allows the thumbnail to represent moving image contents without requiring full motion playback capability.
Solution Approach 2:
The patent changes the parameter of image representation from full-frame still image to face-extracted still image. By transforming the thumbnail generation process to focus on extracting and displaying only the face region, it maintains still image simplicity while improving content perception.
3Loss of information
If full moving image is displayed, then content perception is improved, but device size and power consumption increase
Solution Approach 1:
The patent extracts only the essential face region from the full moving image, reducing the amount of data that needs to be processed and displayed. This extraction significantly lowers power consumption while maintaining the ability to perceive moving image contents.
Solution Approach 2:
The patent applies partial action by processing and displaying only the necessary portion (face) of the moving image rather than the entire frame. This partial processing approach reduces computational load and power consumption while achieving the goal of content perception.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A thumbnail generating apparatus includes a moving image extracting portion, a reduction and enlargement ratio calculating portion and a reducing and enlarging unit. The moving image extracting portion extracts a moving image corresponding to a period shorter than a reproduction time of moving image based on the number and area of face detection regions obtained from face detection information generated by detecting a face of person included in the moving image. The reduction and enlargement ratio calculating portion integrates the face detection regions at intervals of certain time based on the face detection information of the moving image extracted by the moving image extracting portion, and calculates a reduction and enlargement ratio of the moving image based on the integration result such that a face of person included in the moving image extracted by the moving image extracting portion is largely displayed. The reducing and enlarging unit reduces or enlarges the moving image extracted by the moving image extracting portion based on the reduction and enlargement ratio calculated by the reduction and enlargement ratio calculating portion to generate a moving image thumbnail