Camera Gesture Recognition for Hands-Free Photo Mode Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The inconvenience of switching photographing modes on electronic devices, such as smartphones, by tapping the screen during video recording or photo taking, hinders a seamless user experience.
Innovation Solution
Implementing a method where the electronic device recognizes user gestures, specifically hand flip and swipe gestures, to switch between different photographing modes, including dual-scene or multi-scene photographing, by determining gesture vectors based on hand image analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the user taps a control on the screen to switch photographing modes, then the mode switching can be completed, but the operation becomes inconvenient and hinders seamless user experience
Solution Approach 1:
The patent replaces the mechanical touch screen interaction (tapping controls) with a gesture recognition system that uses camera imaging and bone point detection to identify hand gestures. This substitution eliminates the need for screen tapping and enables more natural, seamless mode switching through intuitive hand movements.
Solution Approach 2:
The system enables self-service by automatically detecting and responding to user gestures without requiring manual screen interaction. The camera continuously captures hand images, the bone point detection algorithm automatically processes these images to identify gestures, and the system autonomously switches photographing modes based on detected gestures, making the entire process hands-free and automatic.
2Measurement precision
If deep learning algorithms are used for gesture recognition, then recognition accuracy can be improved, but computational complexity and processing time increase
Solution Approach 1:
The patent extracts only the essential features needed for gesture recognition by detecting bone points in hand images rather than processing entire images through complex deep learning models. This extraction approach focuses computational resources on identifying key anatomical landmarks (bone points) that define gesture characteristics, significantly reducing processing time while maintaining recognition accuracy.
Solution Approach 2:
The gesture recognition process is segmented into distinct stages: hand detection, bone point identification, gesture vector calculation, and gesture classification. This segmentation allows each stage to be optimized independently, with bone point detection serving as an efficient intermediate step that bridges simple image capture and complex gesture interpretation, reducing overall processing time.
Data Source
AI summary
A photographing method is disclosed. By implementing the method, an electronic device 100, for example, a mobile phone, may determine bone points of a hand of a user in an image by using a bone point recognition algorithm, and determine two gesture feature points from the bone points. Then the electronic device 100 may construct, by using positions of the two gesture feature points on a horizontal coordinate axis, a gesture vector representing a state of the hand (a palm or the back of the hand). Further, the electronic device 100 may determine a gesture vector of a subsequent image frame by using the same method. When recognizing that a direction of the gesture vector in the subsequent image frame is opposite to a direction of a gesture vector in an initial image frame, the electronic device 100 may determine that the user has made a hand flip gesture.