Machine Learning Style Modification for Video Game Accessibility
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio-visual media, such as videogames and movies, lack accessibility features for individuals with disabilities, particularly those with vision impairments, and are time-consuming and labor-intensive to accommodate, with current solutions failing to provide adequate accommodations for colorblind users and dynamic action sequences.
Innovation Solution
An On-Demand Accessibility System that includes modules for Action Description, Scene Annotation, Color Accommodation, Graphical Style Modification, and Acoustic Effect Annotation, utilizing neural networks to enhance media accessibility by adding subtitles, text-to-speech descriptions, and color adjustments, allowing users to selectively activate these features during media consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional manual methods are used to add accessibility features to video games, then accessibility accommodations can be provided, but the process becomes time-consuming and labor-intensive
Solution Approach 1:
The patent replaces manual mechanical processes with automated machine learning systems. Specifically, it uses neural networks to automatically generate color palette adjustments, scene descriptions, and accessibility features, eliminating the need for manual creation of these accommodations.
Solution Approach 2:
The system enables media content to automatically generate its own accessibility accommodations. The machine learning models analyze the game content and autonomously create appropriate accessibility features without requiring external manual intervention for each specific accommodation.
2Adaptability or versatility
If custom color palettes are created for colorblind users, then colorblind accessibility is improved, but the complexity increases due to the sheer number of scenes and colors
Solution Approach 1:
The patent changes the parameters of color palettes dynamically based on user needs. Machine learning models analyze the original game's color scheme and automatically transform it into an accessible color palette that maintains the game's aesthetic while accommodating colorblind users, avoiding manual adjustment of numerous individual color parameters.
Solution Approach 2:
The system creates a universal solution that works across all scenes and color combinations in the game. Rather than manually customizing each scene's color palette, the machine learning model applies a general color transformation approach that adapts to any game content, making the system versatile across different games and scenarios.
3Adaptability or versatility
If video games are designed with accessibility features from the start, then accessibility is improved, but the cost and complexity of game development increases
Solution Approach 1:
The patent applies accessibility features as a preliminary post-processing step rather than requiring them to be built into the core game development process. The machine learning systems generate accessibility accommodations before or during the player's experience, allowing games to be released with full accessibility support without redesigning the entire development pipeline.
Solution Approach 2:
The system introduces an intermediary layer between the original game and the player. This intermediary machine learning system translates the game's visual and audio output into accessible formats in real-time, allowing the original game to remain unchanged while providing accessibility through an intermediate processing layer.
4Measurement precision
If machine learning models are trained on large datasets, then the accuracy of accessibility features is improved, but the training time and computational resources increase
Solution Approach 1:
The patent performs model training as a preliminary action before deployment. By training the machine learning models in advance on comprehensive datasets, the system achieves high accuracy for generating accessibility features. The trained models are then reused during actual game playback, avoiding the need for continuous training and reducing real-time computational burden.
Data Source
AI summary
Graphical style modification may be implemented using machine learning. A color accommodation module receives an image frame from a host system and generates a color-adapted version of the image frame. A Graphical Style Modification module generates a style adapted video stream by applying a style adapted from a target image frame to each image frame in a buffered video stream.