Image Playback System Using Index-Based Relevance Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic apparatuses struggle to create engaging moving pictures by limiting displayed images to those within a designated directory, missing relevant but unexpected still images, and lacking in audio-visual appeal.
Innovation Solution
An electronic apparatus with an indexing module to generate index information for still images, a first playback module to create a moving picture, a first image select module to select relevant images, and a second playback module to play back a second moving picture using the selected images, automatically extracting and incorporating images based on attributes like date, time, and face recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If still images are classified into directories based on date/time and only images in the designated directory are displayed, then the image selection process is simple and fast, but the moving picture becomes less interesting and misses relevant unexpected images
Solution Approach 1:
The patent segments the image selection process into two stages: first selecting a key image from the designated directory, then separately selecting additional relevant images from the entire storage based on similarity to the key image. This segmentation allows the system to maintain simplicity in the primary selection while expanding versatility in the secondary selection.
Solution Approach 2:
The patent extends the image selection from a single directory dimension to multiple dimensions by searching across the entire image storage based on content similarity. This dimensional expansion allows the system to find relevant images beyond the designated directory, incorporating unexpected but related images into the moving picture.
2Device complexity
If images are merely selected successively from still images, then the processing is simple and fast, but the audio-visual effect is insufficient and user engagement is low
Solution Approach 1:
The patent introduces an image similarity calculation module as an intermediary between simple image selection and complex video editing. This mediator calculates similarity between the key image and other images to intelligently select complementary images, enhancing audio-visual quality without requiring full video editing complexity.
Solution Approach 2:
The system performs self-service by automatically calculating image similarities and selecting relevant images without requiring manual user intervention. This automation enhances the audio-visual effect while keeping the user interface simple, resolving the contradiction between processing complexity and output quality.
3Ease of operation
If the user designates the directory as the processing target, then the operation is straightforward and fast, but the system cannot present unexpected relevant images from other directories
Solution Approach 1:
The patent performs preliminary action by automatically selecting additional relevant images based on similarity to the key image before the user completes their selection. This preliminary expansion of the image set ensures that unexpected but relevant images are included without requiring the user to manually search multiple directories.
Solution Approach 2:
The system uses feedback from image similarity calculations to automatically supplement the user's selected images with relevant images from other directories. This feedback mechanism allows the system to learn from the key image and autonomously discover and include related images, preventing loss of relevant information while maintaining operational simplicity.
Data Source
AI summary
According to one embodiment, an electronic apparatus includes an indexing module, a first playback module, a first image select module, an image extraction module, and a second playback module. The indexing module is configured to generate index information indicative of attributes of a plurality of still images. The first playback module is configured to play back a first moving picture by using the plurality of still images. The first image select module is configured to select a first still image from the plurality of still images. The image extraction module is configured to extract a still image group including still images relevant to the first still image, based on the index information. The second playback module is configured to play back, in place of the first moving picture, a second moving picture by using the still image group.