Game Image Generation for Sign Language Visibility
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video games face challenges in implementing signing characters due to issues such as characters' hands not being visible during cut scenes or players inadvertently looking away from signing characters during gameplay, which hinders understanding of sign language communication.
Innovation Solution
The system adjusts image generation to ensure signing characters' hands are visible by panning, zooming, or adding alternative visual cues like subtitles or an interpreting character, and uses machine learning to detect and translate sign language.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the camera focuses on a character's face during cut scenes, then the character's facial expressions are visible, but the character's hands are not visible making sign language communication difficult to understand
Solution Approach 1:
The patent introduces a second camera angle that captures both the character's face and hands simultaneously. This multi-dimensional camera approach allows the system to capture information in multiple spatial dimensions, resolving the contradiction between showing facial expressions and showing hands for sign language communication.
Solution Approach 2:
The patent uses an intermediary processing system that receives footage from multiple camera angles and synthesizes a composite view. This intermediary process combines the facial expression information from one angle with the hand visibility from another angle, providing both types of information simultaneously to the viewer.
2Ease of operation
If the player controls the camera freely during gameplay, then the player can explore the scene, but the player may inadvertently look away from the signing character's hands
Solution Approach 1:
The patent implements a feedback mechanism that monitors the player's camera orientation and provides cues when the signing character's hands are out of view. This feedback loop allows the player to maintain camera control freedom while being informed when important visual information is lost, enabling them to adjust their viewing accordingly.
Solution Approach 2:
The patent provides preliminary visual cues or notifications to the player before the hands go out of view, or automatically adjusts the camera to a predetermined angle that ensures hand visibility when sign language communication is detected, preventing information loss before it occurs.
3Measurement precision
If the camera zooms in on a specific object or character, then detailed view is achieved, but the signing character's hands may be excluded from the frame
Solution Approach 1:
The patent makes the camera system dynamic by allowing automatic adjustment of zoom and framing based on detected sign language communication. When sign language is detected, the system dynamically modifies camera parameters to ensure hands remain visible, balancing detailed view requirements with accessibility needs.
Data Source
AI summary
An entertainment device, method and computer program, the entertainment device includes processing circuitry to process a game state of a game and to generate, based on a plurality of game assets, an image of a scene, and output circuitry to output the image for display. The processing circuitry is configured to determine, based on the plurality of game assets, whether the scene includes at least one character communicating using sign language, to adjust the generation of the image in dependence on whether the scene comprises the at least one character communicating using sign language.