Video Moving Thumbnail Generation via Frame Quality Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video thumbnail generation methods fail to provide users with sufficient information for informed selection, as static thumbnails are limited and moving thumbnails may spoil the video content or provide too little information.
Innovation Solution
A method for creating moving thumbnails by sampling video frames, determining frame-level quality scores, applying a sliding window to group frames, and selecting a subset based on quality scores to generate a representative portion of the video item, while ensuring the subset includes timestamps near shot boundaries and adjusting playback speed based on motion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If a moving thumbnail includes more video frames to provide more information, then the information content increases, but the risk of spoiling video content increases
Solution Approach 1:
The patent applies local quality by selecting specific video frames based on their positional characteristics relative to shot boundaries. Instead of uniformly selecting frames throughout the video, the method prioritizes frames located near shot boundaries (within a predetermined time range), thereby providing informative content while avoiding frames that would reveal later plot developments. This localized selection strategy resolves the contradiction between providing sufficient information and preventing spoilers.
Solution Approach 2:
The patent employs preliminary action by pre-determining the temporal boundaries of shots in the video before selecting thumbnail frames. The system identifies shot boundaries and establishes safe time ranges in advance, then selects frames from these pre-defined regions. This preliminary structuring allows the moving thumbnail to provide informative content while systematically avoiding spoilers, as the selection criteria are established before the actual frame selection process.
2Loss of information
If a moving thumbnail uses a longer duration to show more content, then the information provided increases, but the user attention span and interface cleanliness decrease
Solution Approach 1:
The patent applies partial action by selecting only the essential portion of video content for the moving thumbnail. Instead of using the entire video or excessive frames, the method selectively extracts frames from specific temporal regions (near shot boundaries) within a controlled duration. This partial selection provides sufficient information for user decision-making while maintaining a compact size that preserves interface cleanliness and user attention.
3Productivity
If the moving thumbnail playback speed is increased to show more content in less time, then the information delivery efficiency increases, but the user comprehension and motion quality decrease
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the playback speed of the moving thumbnail based on the motion characteristics of the selected video frames. The system calculates motion metrics for candidate frames and modifies the playback speed parameter accordingly, ensuring that frames with significant motion are displayed at appropriate speeds for comprehension. This parameter adjustment resolves the contradiction between efficient information delivery and maintainable comprehension quality.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method of generating a moving thumbnail is disclosed. The method includes sampling video frames of a video item. The method further includes determining frame-level quality scores for the sampled video frames. The method also includes determining multiple group-level quality scores for multiple groups of the sampled video frames using the frame-level quality scores of the sampled video frames. The method further includes selecting one of the groups of the sampled video frames based on the multiple group-level quality scores. The method includes creating a moving thumbnail using a subset of the video frames that have timestamps within a range from the start timestamp to the end timestamp.