Video Endpoint Overcrowding Detection via Computer Vision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing videoconferencing systems are inadequate in accurately determining and managing the number of persons at an endpoint to prevent overcrowding, which poses safety risks and inefficiencies during meetings.
Innovation Solution
Implementing a videoconferencing system that uses image data capture, computer vision, and machine learning to detect the number of individuals at a meeting endpoint, issuing alerts and taking remedial actions such as halting meetings, suggesting alternative locations, or restricting access when capacity is exceeded, and tracking overcapacity trends to inform facility management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If people detection methods are used to minimize the risk of overcrowding at videoconferencing endpoints, then safety is improved, but the existing methods are not wholly successful in accurately determining the number of persons
Solution Approach 1:
The patent introduces an intermediary system comprising image capture devices, processors, and notification devices that mediate between the physical meeting space and the videoconferencing system. This intermediary automatically detects persons using image data, determines quantities, compares against thresholds, and triggers notifications or access control actions, thereby achieving accurate and reliable person detection that directly addresses the shortcomings of existing methods
Solution Approach 2:
The patent replaces manual or simple detection methods with an automated computer vision system using image capture devices and processors. This substitution of mechanical/manual monitoring with automated optical detection and image processing enables precise, real-time person detection and counting, directly improving both the reliability and measurement precision of the detection system
2Productivity
If the number of persons at an endpoint is not accurately monitored, then meeting efficiency is maintained, but safety risks arise from overcrowding
Solution Approach 1:
The patent implements preliminary action by establishing predetermined thresholds for safe occupancy and using automated detection to identify when these thresholds are approached or exceeded. The system proactively triggers notifications to attendees and can prevent access before overcrowding occurs, thereby eliminating safety risks while maintaining meeting efficiency through automated early warning rather than reactive measures
Solution Approach 2:
The patent implements feedback mechanisms where the system continuously monitors person detection data, compares it against safety thresholds, and provides real-time notifications to attendees when capacity is exceeded. This closed-loop feedback system enables dynamic adjustment of meeting attendance based on actual conditions, maintaining both safety and efficiency through automated information flow between the detection system and meeting participants
Data Source
AI summary
Systems and methods of managing videoconferencing endpoint capacity based on safety concerns (e.g., COVID-19), comprising: capturing a video feed corresponding to a geographic location; detecting a presence of one or more persons corresponding to the geographic location, based on data of the video feed; determining a quantity of persons corresponding to the geographic location, based the data of the video feed; determining that the quantity of persons corresponding to the geographic location exceeds a first predetermined threshold; and—in response to the determination that the quantity of persons corresponding to the geographic location exceeds the first predetermined threshold—issuing one or more alert messages and/or taking other appropriate action(s).