Interpretation of Geometric Shapes
Miquel PratsSteve GarnerThe Design GroupThe Open UniversityUK
ETRA – Eye Tracking Research & Applications22-24 March 2010Austin, Texas
Project funded by:
Project partners:
Iestyn JowersAlison McKaySchool of Mechanical EngineeringUniversity of LeedsUK
Nieves PedreiraInformation and CommunicationsTechnologies University of A Coruña Spain
The Designing with Vision project
investigates the possibility of using eye-tracking technology as an interface for systems intended to support the creation, exploration, and development of shape alternatives in conceptual design.
Project funded by:
Project partners:
Interpretation of Geometric Shapes
The Designing with Vision project
is interested in exploring the possibilities of using eye-tracking technology as an interface for systems intended to support the creation, exploration, and development of design shape alternatives in conceptual design.
Project funded by:
Project partners:
Interpretation of Geometric Shapes
Project funded by:
Project partners:
Interpretation of Geometric Shapes
Project funded by:
Project partners:
Interpretation of Geometric Shapes
The Designing with Vision project seeks to:
– Understand how eye-tracking technology can be used to recognise different interpretations of shapes.
– Build a computational design system that supports fluid design reinterpretation and exploration.
Project funded by:
Project partners:
Interpretation of Geometric Shapes
Project funded by:
Project partners:
Interpretation of Geometric Shapes
Computationalshape interpretation
The Designing with Vision project seeks to:
– Understand how eye-tracking technology can be used to recognise different interpretations of shapes.
– Build a computational design system that supports fluid design reinterpretation and exploration.
Interpretation of Geometric Shapes
Computationalshape interpretation: recognition
Gross (2001)Back of an Envelope
The program has been trained to recognise ‘interesting’ shapes
The program has not been trained to recognise these shapes
Saund and Moran (1994)What You Perceive Is What You Get
Interpretation of Geometric Shapes
Computationalshape interpretation: selection
Gesture-based shape selection
Interpretation of Geometric Shapes
Computationalshape interpretation: generation
Jowers et al. (2008)Design Synthesis and Shape Generation
Based on the shape grammar formalism (Stiny and Gips 1972)
Interpretation of Geometric Shapes
Eye tracking andshape perception
Yarbus (1967)Eye movements and fixations are influenced by the viewer’s intent
Interpretation of Geometric Shapes
Our study
– Analyse eye movement data with the objective to determine when a shape is recognised.
– Analyse eye movement data with the objective to determine which shape is selected.
Main Goals
Interpretation of Geometric Shapes
Our study
z– Analyse eye movement data with the obj the moment when a shape is recognised..
Main Goals
We call it ‘Now moment’
Interpretation of Geometric Shapes
Our study
The experiment7 participants3 tasks
Task 1
Find the square Find the arrow
Interpretation of Geometric Shapes
Our study
The experiment7 participants3 tasks
Task 1
Find the square Find the arrow
Interpretation of Geometric Shapes
Our study
The experiment7 participants3 tasks
Task 1
Interpretation of Geometric Shapes
Our study
The experiment7 participants3 tasks
Task 2
Say which shapes each image is composed of
Interpretation of Geometric Shapes
Our study
The experiment7 participants3 tasks
Task 2
Extract of participants’ transcript
Interpretation of Geometric Shapes
Our study
The experiment7 participants3 tasks
Task 3
Find this shape in the following figures
Interpretation of Geometric Shapes
Our study
The experiment7 participants3 tasks
Task 3
Interpretation of Geometric Shapes
Main conclusions
Our results…
… support previous findings that indicate the relationship between saccade amplitude and fixation duration in visual fixations (Unema et al. 2005, Velichkovsky et al. 2005).
… show that fixation duration and saccade amplitude can be used as indicators of shape interpretation.
… support the hypothesis that eye movements could be used to select shapes in a natural and non intrusive way.
Interpretation of Geometric Shapes
Future work
– Evaluate the data by means of statistical analysis.
– Examine whether fixation duration and saccade amplitude are also indicators of shape recognition in ‘designerly’ tasks.
– Explore ways to determine which shapes are selected.
– Build a computational drawing system that uses eye movement data to recognise any shape interpreted in a design.
Interpretation of Geometric Shapes
Future work
Experiments to determine which shape is interpreted
Development of Designing with Vision software
Project funded by:
Project partners:
Interpretation of Geometric Shapes
Project funded by:
Project partners:
Interpretation of Geometric Shapes
Project funded by:
Project partners:
Interpretation of Geometric Shapes
Project funded by:
Project partners:
Interpretation of Geometric Shapes
Project funded by:
Project partners:
Interpretation of Geometric Shapes
http://design.open.ac.uk/DV/