Context-Sensitive Thumbnail Generation for Live TV Streams
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating thumbnail images for television programs often result in unsuitable content, such as pornography, gruesome images, or advertisements, which are not representative of the current program content, making them inappropriate for general audiences and lacking value.
Innovation Solution
A system that captures still images from live video streams, filters them for offensive, defective, or advertising content, and generates suitable thumbnail images for display, using an image filter with modules for detecting pornography, defective images, and advertisements, and stores them in an image cache for association with Electronic Program Guide (EPG) data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If pre-selected publicity photos are used as thumbnails, then the thumbnail generation process is simple, but the thumbnails do not convey what is actually being shown on the program
Solution Approach 1:
The system performs preliminary actions by capturing still frames from the video stream in advance and pre-filtering them to identify suitable thumbnails. This allows the thumbnails to be ready before the user requests them, ensuring both timely delivery and accurate program content representation.
Solution Approach 2:
The system uses the program's own video stream to generate its representative thumbnail, rather than relying on external publicity photos. The video stream itself provides the source material for creating an accurate visual representation of what is being shown.
2Loss of information
If images are captured from live video stream, then the thumbnails represent current program content, but they may contain pornographic, unpleasant, gruesome, or unsuitable content
Solution Approach 1:
The patent introduces an intermediary filtering system that processes captured video frames before they are used as thumbnails. This intermediary layer analyzes each frame for inappropriate content (pornographic, gruesome, unsuitable material) and blocks such images from being displayed, while allowing appropriate frames to pass through as valid thumbnails.
Solution Approach 2:
The filtering system provides feedback by evaluating captured frames against content appropriateness criteria and determining whether each frame is suitable for display. This feedback mechanism enables the system to automatically reject inappropriate frames and select only suitable frames for thumbnail generation.
3Productivity
If arbitrarily captured images are used from video stream, then the capture process is simple, but they include blank screens, commercials, or other inappropriate content with little value to the viewer
Solution Approach 1:
An intermediary analysis process examines captured frames to distinguish between valuable program content and worthless content such as blank screens and commercials. This intermediary layer adds intelligence to the selection process, ensuring that only frames with actual program value are chosen as thumbnails.
Solution Approach 2:
The system changes the selection parameters by evaluating multiple frame attributes beyond simple timing, such as content density, visual interest, and program relevance. By adjusting these parameters, the system identifies frames that best represent the program content while excluding blank screens and commercials.
Data Source
AI summary
A system and methodology provide for automatic generation of thumbnail images of a video stream of images from a television channel. A still frame image of the video stream of the channel is captured. A plurality of content filters is applied to the still frame image to determine whether the image is suitable for display. Images can be filtered for offensive content, defective content, or advertising content. A filtered image that is appropriate for display is selected and stored. If the still frame image is inappropriate for display, a default thumbnail image is selected. The selected image is stored as a thumbnail image in association with EPG data for the television program. The thumbnail image and EPG data are sent for display to a viewer.