Gaze Swipe Selection for Head-Mounted Displays

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing augmented reality systems face challenges in hands-free selection of virtual objects due to the risk of false positive selections when using head tracking, eye tracking, or gaze tracking, particularly when the selection area and identifying criteria overlap, leading to unintentional selection of virtual objects like text while reading.

Innovation Solution

Implementing a gaze swipe interaction method where users perform head movements to track a virtual pointer through two or more edges of a virtual object, with varying gestures and speeds to distinguish between different selection types, and allowing deselection of erroneously selected objects, thereby preventing false selections.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If head tracking or gaze tracking is used for hands-free selection of virtual objects, then ease of operation is improved, but reliability deteriorates due to false positive selections

Engineering Contradiction:
Improvehands-free selectionVSAvoidfalse positive selection
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The selection process is segmented into multiple distinct phases: identification phase (gazing at object) and confirmation phase (swipe gesture). This segmentation separates the initial interest indication from the final selection action, reducing false positives by requiring a deliberate confirmatory gesture beyond mere gaze direction.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system transitions from static gaze-based selection to dynamic swipe gesture-based confirmation. The swipe gesture introduces motion and temporal dynamics, requiring the user to perform a specific trajectory-based action rather than simply maintaining a gaze, thereby reducing accidental selections.

Inventive Principle:
Principle #15Dynamics

2Ease of operation

If the selection area and identifying criteria overlap, then ease of operation is improved, but measurement precision deteriorates leading to unintentional selection

Engineering Contradiction:
Improveselection accessibilityVSAvoidselection accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The system adds a temporal and motion-based dimension to the selection process. Instead of relying solely on spatial overlap between gaze and object (2D), the swipe gesture introduces a trajectory-based verification layer, requiring the pointer to move across the object in a specific pattern, thereby enhancing selection precision without reducing accessibility.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Adaptability or versatility

If gaze swipe gesture speed varies, then adaptability is improved for different selection types, but device complexity increases

Engineering Contradiction:
Improveselection type differentiationVSAvoidgesture recognition system
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system utilizes swipe gesture speed as a variable parameter to differentiate between selection types. By monitoring the velocity of the swipe motion, the system can distinguish between different user intentions (e.g., quick swipe for standard selection, slower swipe for careful selection), providing adaptability without requiring complex additional hardware or multiple gesture types.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3097460B1Gaze swipe selection
Publication Date: 2019.01.09 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP3097460B1 patent drawingFigure 1
  • EP3097460B1 patent drawingFigure 2A
  • EP3097460B1 patent drawingFigure 2B

AI summary

Methods for enabling hands-free selection of virtual objects are described. In some embodiments, a gaze swipe gesture may be used to select a virtual object. The gaze swipe gesture may involve an end user of a head-mounted display device (HMD) performing head movements that are tracked by the HMD to detect whether a virtual pointer controlled by the end user has swiped across two or more edges of the virtual object. In some cases, the gaze swipe gesture may comprise the end user using their head movements to move the virtual pointer through two edges of the virtual object while the end user gazes at the virtual object. In response to detecting the gaze swipe gesture, the HMD may determine a second virtual object to be displayed on the HMD based on a speed of the gaze swipe gesture and a size of the virtual object.