View Angle Recognition via Object Removal in First Person Screens
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic apparatuses face challenges in accurately recognizing the view angle of first person view screens, particularly due to interference from specific objects that hinder the recognition accuracy of neural network models used for this purpose.
Innovation Solution
The electronic apparatus captures the first person view screen, removes specific objects to generate a preprocessed image, and inputs this image into a neural network model to recognize the view angle, thereby improving recognition accuracy and enabling subsequent functions based on the recognized angle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a neural network model is used to recognize view angle of first person view screen, then recognition functionality is provided, but recognition accuracy deteriorates due to interference from specific objects in the screen
Solution Approach 1:
The patent extracts and removes specific objects (such as hands, weapons, or other foreground elements) from the first person view screen before inputting the image to the neural network model. This preprocessing step isolates the relevant visual information for view angle recognition while eliminating distracting elements that would interfere with accurate angle detection.
2Measurement precision
If specific objects are removed from the screen to improve recognition accuracy, then preprocessing steps are added, but system complexity increases
Solution Approach 1:
The patent performs preliminary actions by pre-processing the first person view screen to remove specific objects before the main recognition task. This preliminary removal of interfering elements simplifies the subsequent view angle recognition process and improves overall system efficiency despite the added preprocessing step.
Data Source
AI summary
An electronic apparatus and a method for recognizing view angle of displayed screen thereof. The method is adapted to the electronic apparatus and includes the following steps. A first person view screen displayed by a display is captured. A specific object in the first person view screen is removed to generate a preprocessed image. The preprocessed image is inputted into a neural network model to recognize a view angle of the first person view screen. A function is performed according to the view angle of the first person view screen.