Adaptive Real-Time TV Ad Detection Using Binary Descriptors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting televised video ads in broadcasted channels are inefficient due to high error rates, long detection times, and lack of language and platform agnosticism, failing to accurately differentiate between programs and advertisements.
Innovation Solution
A computer-implemented method that selects frames from a live stream, extracts keypoints, derives binary descriptors, and uses a special pyramid structure with an n-ary tree data structure to compare and score frames for ad detection, reducing detection time to 0.6-1 second with low error rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If audio fingerprinting is used for ad detection, then ad detection capability is provided, but error rate increases and detection time increases
Solution Approach 1:
The video stream is segmented into individual frames, and a subset of frames is selected for analysis rather than processing the entire continuous stream. This segmentation allows the system to process discrete units (frames) independently, reducing overall detection time while maintaining accuracy through strategic frame selection.
Solution Approach 2:
The system performs preliminary actions by pre-defining the set of frames to be analyzed and pre-processing them before actual ad detection. Frames are selected and prepared in advance according to predetermined criteria, which streamlines the subsequent detection process and reduces real-time processing requirements.
2Reliability
If supervised detection of key features is performed, then ad detection is enabled, but detection time increases due to computational inefficiency
Solution Approach 1:
The system extracts only the essential key features from video frames that are relevant for ad detection, rather than analyzing all visual data. By taking out and focusing on specific key features (such as prominent visual elements or patterns), the system achieves accurate detection with reduced computational overhead, thereby improving processing speed.
3Reliability
If audio fingerprints are matched probabilistically, then ad validation is performed, but false positives increase
Solution Approach 1:
The system replaces the acoustic-based detection mechanism (audio fingerprinting) with a visual-based mechanism (image processing and feature extraction). This substitution eliminates the fundamental limitation of audio-based methods that cannot distinguish between ads and non-ad content with similar audio characteristics, thereby improving detection precision and reducing false positives.
4Reliability
If manual detection is assigned to each user per channel, then ad monitoring is performed, but efficiency decreases
Solution Approach 1:
The system creates a universal automated detection mechanism that can monitor multiple channels simultaneously through a single processing framework. The same image processing and feature extraction algorithms are applied across all channels, eliminating the need for separate manual monitoring assignments and dramatically improving overall monitoring efficiency while maintaining comprehensive coverage.
Data Source
AI summary
The present disclosure provides a computer-implemented method and system for adaptively reducing detection time in real time supervised detection of televised ads in media content of a channel. The method includes reception of the media content and selection of a set of frames per second from the media content. The method includes extraction of keypoints from each selected frame and derivation of binary descriptors from extracted keypoints. The method includes assignment of weight value to each binary descriptor and creation of a special pyramid of the binary descriptors. The method includes obtaining a first vocabulary of binary descriptors for each selected frame and accessing a second vocabulary of binary descriptors. The method includes comparison of each binary descriptor in the first vocabulary with binary descriptors in second vocabulary. The method includes progressively scoring each selected frame of the media content and detecting a first ad in the media content.