Accessible Gesture Typing on Smartphones for People with Low Vision

Area of Science:

• Human-Computer Interaction
• Assistive Technology
• Accessibility Research

Background:

• Gesture typing is common on touchscreens but lacks support for low vision users.
• Existing solutions often hinder usability when screen magnifiers are active.

Purpose of the Study:

• To design and evaluate novel keyboard prototypes enabling gesture typing for individuals with low vision.
• To develop a decoding algorithm accommodating unique typing behaviors of low vision users.

Main Methods:

• Implemented two keyboard prototypes: layout-magnified and key-magnified.
• Developed a kinematics-based decoding algorithm for flexible gesture recognition.
• Conducted a user study comparing prototype performance against conventional gesture typing.

Main Results:

• Both prototypes allowed uninterrupted key access with active screen magnifiers.
• The key-magnified keyboard achieved 5.28 words per minute.
• This represents a 27.5% speed improvement over conventional gesture typing with voice feedback.

Conclusions:

• Novel keyboard designs and a specialized decoding algorithm significantly enhance gesture typing for low vision users.
• The key-magnified keyboard offers a faster and more accessible text input method.