Event-Triggered Imaging System for Human-Object Interaction Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Human-object interaction detection in images or videos faces challenges in efficiency and accuracy due to the lack of temporal information in still images and increased resource consumption when using a series of images to capture movement.
Innovation Solution
An imaging system comprising an event sensor and an image sensor, aligned in field of view, captures event data sets and visual images, with a controller detecting humans and triggering image capture only when necessary, combining temporal and visual information for more accurate and efficient interaction detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If a series of still images is used to capture movement information for human-object interaction detection, then the temporal information for recognizing interaction is improved, but the computation and resources required increase significantly
Solution Approach 1:
The system segments the imaging task by using two different sensors with different functions: an event sensor that captures only temporal changes (motion events) and a traditional image sensor that captures visual appearance. This segmentation allows the system to obtain temporal information without processing complete video frames, significantly reducing computational complexity while preserving necessary temporal data for interaction detection.
Solution Approach 2:
The event sensor performs partial action by capturing only the changes in light intensity (temporal information) rather than capturing complete image frames. This partial capture of information is sufficient for detecting motion and temporal patterns, avoiding the excessive computational burden of processing full video sequences while still providing the necessary temporal context for human-object interaction detection.
2Measurement precision
If a traditional image sensor is used to capture visual information continuously, then the visual detail for interaction recognition is improved, but the power consumption increases
Solution Approach 1:
The system implements periodic action by having the traditional image sensor activate only at specific moments (when events are detected by the event sensor) rather than continuously. The event sensor operates continuously to detect temporal changes, and triggers the power-intensive image sensor only when necessary, thereby maintaining high visual information quality while dramatically reducing overall power consumption.
Solution Approach 2:
The event sensor serves itself and the triggering mechanism by continuously monitoring for temporal changes and autonomously determining when the image sensor should be activated. This self-service approach eliminates the need for continuous operation of the power-intensive image sensor, reducing power consumption while ensuring visual information is captured only when temporal changes warrant further analysis.
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
The system achieves higher accuracy and lower power consumption in human-object interaction detection by leveraging temporal and visual information, reducing unnecessary resource usage and improving detection efficiency.
Implementation Method 1
an event sensor...configured to obtain an event data set of the targeted scene according to variations of light intensity sensed by pixels of the event sensor
Implementation Method 2
an image sensor...configured to capture a visual image of the targeted scene
Data Source
AI summary
The present application discloses an imaging system for detecting human-object interaction and a method for detecting human-object interaction thereof. The imaging system includes an event sensor, an image sensor, and a controller. The event sensor is configured obtain an event data set of the targeted scene according to variations of light intensity sensed by pixels of the event sensor when an event occurs in the targeted scene. The image sensor is configured capture a visual image of the targeted scene. The controller is configured to detect human according to the event data set, trigger the image sensor to capture the visual image when the human is detected, and detect the human-object interaction in the targeted scene according to the visual image and a series of event data sets obtained by the event sensor during the event.