Gesture Detection Region Adjustment via Motion Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current gesture recognition technologies based on machine vision require full-screen detection, leading to slow recognition speeds and difficulties in smooth operation on terminal equipment, while also increasing computational load.
Innovation Solution
A gesture detection method that performs a shrinking process and background modeling on images to generate a background image, determining motion regions, and adjusting gesture detection regions based on the motion range, thereby reducing the scanning range and computational load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full-screen detection is performed for gesture recognition, then gesture recognition accuracy is maintained, but computation speed decreases and computational load increases
Solution Approach 1:
The patent divides the full-screen detection area into multiple sub-regions based on motion detection results. Instead of detecting gestures across the entire screen, the system first identifies motion regions and then performs gesture detection only within those specific sub-regions. This segmentation approach maintains gesture recognition accuracy while significantly reducing the computational load and improving processing speed.
Solution Approach 2:
The patent applies different processing qualities to different regions of the screen. Motion regions identified through background subtraction receive full gesture detection processing, while static background regions are excluded from gesture detection. This local quality approach ensures accurate gesture recognition in relevant areas while avoiding unnecessary computation in irrelevant areas.
2Measurement precision
If full-screen detection is performed for gesture recognition, then gesture recognition accuracy is maintained, but operation smoothness deteriorates
Solution Approach 1:
The patent segments the detection process into two stages: first identifying motion regions through background subtraction, then performing gesture detection only within those regions. This segmentation reduces the amount of data processed in real-time, enabling smoother operations on terminal equipment with limited computing resources while maintaining accurate gesture recognition.
3Productivity
If detection region is reduced to enhance computation speed, then computation load decreases, but gesture detection coverage may be insufficient
Solution Approach 1:
The patent performs preliminary motion detection through background subtraction before conducting gesture detection. By first identifying regions with motion activity, the system prepares the detection process in advance, ensuring that gesture detection is applied to the correct regions. This preliminary action maintains comprehensive gesture detection coverage while enabling reduced computation load in the subsequent gesture recognition stage.
Data Source
AI summary
A gesture detection method, a gesture processing device, and a computer readable storage medium. The gesture detection method includes: performing a shrinking process and a background modeling on a first frame of original image, to generate a first frame of background image; determining a motion region of the first frame of background image based on the first frame of background image; determining a first gesture detection area in the first frame of original image according to the motion region of the first frame of background image; and performing a gesture detection within the first gesture detection area.