Adaptive Statistical Motion Detection for Video Summaries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Reviewing live or recorded video streams from security cameras is tedious and prone to errors due to the presence of irrelevant image data, making it difficult to efficiently access important footage.
Innovation Solution
A method and system for detecting anomalous motion in video streams by obtaining motion indicators, estimating statistical parameters, and determining anomaly states in time windows to process and summarize relevant video segments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a viewer reviews live or recorded footage from security cameras to identify important events, then the viewer can detect anomalies and important occurrences, but the process is tedious and time-consuming due to the vast majority of footage being irrelevant
Solution Approach 1:
The system performs preliminary analysis of video streams by computing motion indicators and statistical parameters for each time window before human review. This preliminary action identifies anomalous time windows that deviate from normal motion patterns, allowing the system to pre-filter and prioritize footage for viewer review, thus reducing the time required to find important events while maintaining detection accuracy
Solution Approach 2:
The system extracts and isolates only the anomalous time windows from the complete video stream based on statistical deviation analysis. By taking out and presenting only these relevant segments to the viewer, the system eliminates the need to review irrelevant footage while preserving all important occurrences that deviate from normal patterns
2Reliability
If the viewer loses concentration during manual review of video streams, then errors may occur in identifying important footage, but automated systems may fail to distinguish relevant from irrelevant motion patterns
Solution Approach 1:
The system implements feedback through statistical parameter estimation where each time window's anomaly state is determined based on motion indicators and statistical parameters from preceding time windows. This continuous feedback mechanism adapts to changing scene characteristics and provides consistent, reliable anomaly detection without human concentration, achieving high detection reliability through adaptive statistical analysis
Solution Approach 2:
The system performs self-service by automatically analyzing motion indicators, computing statistical parameters, and identifying anomalous time windows without requiring continuous human supervision. The automated statistical analysis service maintains consistent detection reliability while reducing the complexity of human-operated manual review processes
Data Source
AI summary
A computer-implemented method, comprising: obtaining motion indicators for a plurality of samples of a video stream; obtaining an anomaly state for each of a plurality of time windows of the video stream, each of the time windows spanning a subset of the samples, by (i) obtaining estimated statistical parameters for the given time window based on measured statistical parameters characterizing the motion indicators for the samples in at least one time window of the video stream that precedes the given time window and (ii) determining the anomaly state for the given time window based on the plurality of motion indicators obtained for the samples in the given time window and the estimated statistical parameters; and processing the video stream based on the anomaly state for various ones of the time windows.