Automated Video Segment Boundary Detection via Viewer Measurement Data

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for identifying segment boundaries between video program content and interstitial content in recorded video events rely on human operators, which is expensive and imprecise, and are typically limited to a few channels, lacking an efficient automated solution.

Innovation Solution

A computer-implemented method and system that collects viewer measurement data from presentation devices to analyze user viewing behavior, estimating segment boundaries between video program content and interstitial content, generating cue files that indicate these boundaries, and maintaining them for access by presentation devices to enable automatic skipping of unwanted content during playback.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If human operators manually mark segment boundaries in real time, then measurement precision is improved, but operational cost increases and productivity decreases

Engineering Contradiction:

Improvesegment boundary identification accuracyVSAvoidprocessing speed and cost efficiency

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces the mechanical manual marking process with an automated computer-based system that uses algorithmic analysis of viewer measurement data to identify segment boundaries. The system processes viewing patterns, fast-forwarding behavior, and temporal metadata to automatically determine commercial breaks without human intervention, thereby resolving the contradiction between precision and productivity.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service by allowing the viewing data itself to serve as the basis for boundary identification. Viewer measurement data collected during normal playback is automatically analyzed to extract segment boundary information, eliminating the need for separate manual annotation processes and making the system self-sufficient.

Inventive Principle:

Principle #25Self-service

2Productivity

If automated methods are used to identify segment boundaries, then productivity is improved, but measurement precision deteriorates

Engineering Contradiction:

Improveprocessing speed and operational efficiencyVSAvoidsegment boundary identification accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system incorporates feedback mechanisms where viewer measurement data from actual viewing behavior is continuously analyzed to refine boundary identification algorithms. The system learns from viewing patterns, fast-forwarding sequences, and temporal metadata to progressively improve accuracy, ensuring that automated methods achieve both high productivity and precision through iterative refinement.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent utilizes multiple parameters from viewer measurement data including temporal metadata, fast-forwarding duration, pause patterns, and viewing sequences to dynamically determine segment boundaries. By analyzing changes in these parameters across different viewing sessions, the system adapts its boundary detection to maintain high precision while operating automatically at scale.

Inventive Principle:

Principle #35Parameter changes

3Adaptability or versatility

If commercial skipping is provided for multiple channels, then adaptability is improved, but device complexity increases

Engineering Contradiction:

Improvechannel coverageVSAvoidsystem complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system achieves universality by developing a single unified platform that processes viewer measurement data across multiple channels and content types using the same algorithmic framework. The system handles diverse content including live broadcasts, recorded events, and on-demand streaming through consistent boundary detection methods, enabling multi-channel operation without proportionally increasing complexity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent segments the complex task of multi-channel boundary identification into independent modular components: data collection modules for different channels, processing modules that analyze viewing patterns, and output modules that generate cue files. This segmentation allows the system to scale across channels by instantiating the same modular architecture rather than managing monolithic complexity.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS10812873B2Identifying segment boundaries of recorded video events based on aggregated viewer measurement data

Publication Date: 2020.10.20 DISH NETWORK LLC

AI summary

A computer-implemented and executed method of managing recorded video events is disclosed. The method collects viewer measurement data from presentation devices, the viewer measurement data indicating user viewing behavior associated with playback of recorded video events. The method continues by analyzing the collected viewer measurement data to estimate boundaries between segments of the recorded video events, which results in groups of estimated boundaries corresponding to the recorded video events. Cue files are generated for the recorded video events, the cue files indicating the estimated boundaries corresponding to the recorded video events. The cue files are maintained at a central system for access by the presentation devices.