Synchronized Video and Sensor Event Association
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for analyzing human activities, particularly in sports, are time-consuming and limited to professional areas, as they require manual video editing and do not effectively quantify short events like soccer kicks or tennis strokes, and existing sensor-based systems fail to analyze motion sequences in detail.
Innovation Solution
A method and system that record videos and sensor data synchronously, allowing for the automatic detection and association of specific events in sports activities by preprocessing, segmenting, and classifying sensor data to identify key frames in the video, using techniques like low-pass filtering, wavelet analysis, and machine learning classifiers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual video editing is used to analyze sports activities, then video summaries can be produced, but the process is time-consuming and limited to professional areas
Solution Approach 1:
The patent replaces manual mechanical video editing with an automated sensor-based detection system. Sensors mounted on athletes' bodies automatically detect motion patterns and trigger video frame capture, eliminating the need for manual review and editing of entire video sequences. This substitution dramatically reduces time loss while maintaining analysis quality.
Solution Approach 2:
The system enables self-service video analysis by allowing athletes to wear sensors that automatically monitor their own performance. The sensors detect specific motion patterns and autonomously trigger video capture and analysis, producing performance feedback without requiring external manual intervention from coaches or analysts.
2Measurement precision
If body-worn sensors are used to capture motion data, then continuous monitoring is possible, but short events like kicks or strokes cannot be effectively quantified
Solution Approach 1:
The patent segments the continuous sensor data stream into discrete event detections by identifying specific motion patterns characteristic of short-duration actions like kicks or strokes. The system divides the monitoring task into detecting preparatory motions, execution phases, and follow-through, allowing precise quantification of individual events within the continuous activity stream.
Solution Approach 2:
The system changes detection parameters dynamically based on the sport and specific events being monitored. Different threshold values, time windows, and motion pattern templates are applied depending on whether detecting a soccer kick, tennis stroke, or golf swing, enabling precise quantification across diverse short-duration events.
3Loss of information
If synchronized video and sensor data are processed, then specific event frames can be identified, but complex preprocessing and classification are required
Solution Approach 1:
The patent performs preliminary synchronization of video and sensor data streams before event detection, establishing time-correlated references in advance. This preliminary action creates a unified temporal framework that simplifies subsequent event detection and frame association, reducing the complexity of real-time processing while preserving complete event information.
Data Source
AI summary
Described are methods and systems for associating frames in a video of an activity of a person with an event. The methods include recording a video of an activity of a person; storing a time-series of a plurality of sensor data (82) obtained from a sensor assembly (12) of at least one sensor (31, 32, 33, 34, 35) coupled to the person while the person is performing the activity; synchronizing the video with the sensor data (82); detecting an event in the time-series; and associating the event with at least one corresponding frame in the video showing the event.