Camera Position Detection for Motion Recognition in Electronic Displays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The relative position between a camera and a display in electronic devices, such as TVs, can vary, leading to inconvenience in video calls and reduced recognition rates in motion-based applications due to inconsistent camera placement, affecting image capture and user experience.
Innovation Solution
An electronic apparatus that includes a processor to acquire images from a camera, determine its position relative to the display, select a corresponding motion identification model, and adjust the image or position to improve recognition rates and user experience by displaying a GUI to guide camera placement and correcting images for better alignment with prepared models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the camera is placed at various positions according to user convenience, then the ease of operation is improved, but the recognition rate of motion identification deteriorates
Solution Approach 1:
Multiple motion identification models are prepared in advance for different camera positions. The system pre-processes training data for each position and creates corresponding models before actual use, so when the camera is placed at any position, the appropriate model is already ready for immediate application, ensuring high recognition rates regardless of camera placement
Solution Approach 2:
The system changes the parameters of motion identification models according to camera position parameters. Different models have different parameter sets optimized for specific positions, allowing the system to adapt to various camera placements while maintaining high recognition accuracy by selecting the model with parameters matching the current camera position
2Device complexity
If a single motion identification model is used for all camera positions, then the device complexity is reduced, but the recognition rate varies depending on camera position
Solution Approach 1:
The single motion identification model is segmented into multiple position-specific models. Each model is specialized for a particular camera position, dividing the general recognition task into position-specific sub-tasks. This segmentation ensures that each model operates within its optimal performance range, maintaining consistent high recognition rates across different camera placements
Solution Approach 2:
The system creates a universal solution by developing multiple models that collectively cover all possible camera positions. While each model is specialized, the set of models together provides universal coverage, allowing the system to handle any camera placement scenario with high recognition accuracy through appropriate model selection
3Ease of operation
If the camera position is not properly aligned with the display, then the ease of operation is improved, but the user experience in video calls deteriorates
Solution Approach 1:
The system uses the camera position information as feedback to automatically select the appropriate motion identification model. This feedback mechanism ensures that even when the camera is placed at various positions for user convenience, the system adapts by choosing the model that corresponds to the detected position, thereby maintaining video call quality and user experience
Data Source
AI summary
An electronic apparatus including: an interface; a display; and a processor configured to: acquire an image captured by a camera while the camera is connected through the interface, obtain information about a position where the camera is placed with respect to the display among a plurality of positions where the camera is placeable with respect to the display, select a motion identification model corresponding to the position where the camera is placed with respect to the display among a plurality of motion identification models prepared based on data corresponding to the plurality of positions, identify a motion of a user from the image captured by the camera based on the selected motion identification model, and perform a function based on the identified motion of the user.