Speaker Tracking False-Positive Filtering with Visible and Infrared Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video conferencing systems often mistakenly identify non-living objects as participants, leading to false-positive head identifications, which can disrupt the framing and accuracy of video processing.
Innovation Solution
The system uses a combination of visible and infrared spectrums to estimate the true size and temperature of detected heads, filtering out misleading images based on threshold size and temperature ranges to improve the accuracy of head detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If algorithms automatically identify people in the room using visible spectrum imaging, then the system can process video streams and detect heads, but the system produces false-positive identifications of non-living objects as participants
Solution Approach 1:
The system changes the detection parameter from visible spectrum imaging to infrared spectrum imaging. By detecting thermal radiation parameters (temperature) rather than visible light parameters, the system can distinguish living beings from non-living objects, thereby reducing false-positive identifications while maintaining automatic detection capability
Solution Approach 2:
The system introduces an intermediary verification step using infrared imaging between the visible spectrum detection and final identification. The infrared temperature detection acts as a mediator that validates whether a detected head belongs to a living participant, filtering out false positives before final processing
2Device complexity
If the system uses only visible spectrum imaging for head detection, then the processing is simple and fast, but the system cannot distinguish real participants from misleading images
Solution Approach 1:
The system makes the imaging system multi-functional by combining visible spectrum imaging (for spatial location and basic detection) with infrared spectrum imaging (for temperature measurement and validity verification). This allows a single system to perform both detection and validation functions, improving measurement precision without proportionally increasing complexity
3Reliability
If the system filters out candidates based on size and temperature thresholds, then false-positive identifications are reduced, but the system may miss real participants with atypical characteristics
Solution Approach 1:
The system implements dynamic threshold adjustment where the size and temperature thresholds are not fixed but can adapt to different contexts. The system continuously learns from the environment and adjusts thresholds to accommodate atypical participants while maintaining accuracy, balancing filtering effectiveness with detection completeness
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach effectively reduces false-positive identifications by ensuring only real participants' images are processed, enhancing the accuracy and reliability of video conferencing systems.
Implementation Method 1
an infrared camera (1021) configured to receive an infrared spectrum (146) of light
Data Source
AI summary
A method including receiving, at a camera, light comprising a visible spectrum and an infrared spectrum (200). The method also includes detecting, using the visible spectrum, a candidate image of a head (202). The method also includes determining an estimated true size of the head (204). The method also includes determining, using the infrared spectrum, a temperature of the head (206). The method also includes filtering the candidate image of the head with respect to an operation of a video controller configured to track one or more images of heads in the visible spectrum. Filtering is based on both the estimated true size of the head and the temperature of the head (208).