Personalized Relevancy Model for Aggregated Camera Feeds

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Users of security camera systems face overwhelming amounts of motion-related video clips, requiring significant time to find relevant content across multiple cameras.

Innovation Solution

A personalized relevancy model is created based on user interactions, assigning relevancy scores to video clips from multiple cameras, and prioritizing the most relevant content for presentation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If video clips from multiple cameras are aggregated and presented in chronological order, then all security events are captured, but users spend excessive time browsing through irrelevant clips

Engineering Contradiction:

Improvenumber of video clipsVSAvoidtime to find relevant clips

Core Design Contradiction:

Quantity of substanceVSLoss of time

Solution Approach 1:

The system changes the parameter of video clip presentation from chronological order to relevancy-scored order. A machine learning model analyzes user interactions (playback history, pause duration, skip patterns) to generate relevancy scores for each video clip, dynamically reordering the feed to prioritize clips most likely to interest the user while maintaining comprehensive coverage of all security events

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The system implements feedback loops where user interactions with video clips (playing, pausing, skipping, watching duration) are continuously monitored and fed back into the machine learning model. This feedback refines the relevancy scoring algorithm over time, adapting to individual user preferences and behaviors to progressively improve clip prioritization and reduce browsing time

Inventive Principle:

Principle #23Feedback

2Reliability

If users review all video clips from individual cameras, then complete coverage is achieved, but the operation becomes complex and time-consuming

Engineering Contradiction:

Improvecoverage of security eventsVSAvoidease of reviewing clips

Core Design Contradiction:

ReliabilityVSEase of operation

Solution Approach 1:

The system extracts and prioritizes the most relevant video clips from the aggregate feed based on relevancy scores, presenting them first in the unified feed. This allows users to quickly access critical security events without needing to manually navigate through all clips from all cameras, while still maintaining the option to review complete coverage if needed

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system merges video clips from multiple individual camera feeds into a single unified feed that is sorted by relevancy score. This consolidation eliminates the need for users to switch between multiple camera interfaces, providing comprehensive security coverage through one streamlined view that adapts to user preferences

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS20250200970A1Aggregating and/or Personalizing Camera Feeds

Publication Date: 2025.06.19 COMCAST CABLE COMM LLC

AI summary

Systems, apparatuses, and methods are described for aggregating and/or personalizing video feeds for users of a system that comprises multiple video capture devices (e.g., home security system cameras). A user's interactions with video clips may be analyzed to build a personalized relevancy model (e.g., a scoring model). Video clips from a plurality of cameras may be aggregated and assigned with relevancy scores based on the relevancy model. Scored video clips may be ranked and presented to the user in an order that presents the most relevant video clips first.