Top Banner
1,001,001 Faces user testing plan for face navigation interface Tzu-Pei Grace Chen Sidney Fels Human Communication Technologies Lab
13

1,001,001 Faces user testing plan for face navigation interface Tzu-Pei Grace Chen Sidney Fels Human Communication Technologies Lab.

Dec 18, 2015

Download

Documents

Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: 1,001,001 Faces user testing plan for face navigation interface Tzu-Pei Grace Chen Sidney Fels Human Communication Technologies Lab.

1,001,001 Faces user testing plan for

face navigation interface

Tzu-Pei Grace ChenSidney Fels

Human Communication Technologies Lab

Page 2: 1,001,001 Faces user testing plan for face navigation interface Tzu-Pei Grace Chen Sidney Fels Human Communication Technologies Lab.

What is this project about?

• Buddhist Deities of Sanjusangen-do Temple raises two interesting questions:– Finite set of faces for

majority to identify with– Effective interface for

face navigation

Page 3: 1,001,001 Faces user testing plan for face navigation interface Tzu-Pei Grace Chen Sidney Fels Human Communication Technologies Lab.

Face Searching Background

• Component vs. Configural navigation• Component and configural approach

complement each other.

Page 4: 1,001,001 Faces user testing plan for face navigation interface Tzu-Pei Grace Chen Sidney Fels Human Communication Technologies Lab.

Face Wheel Design

• Considered color palette, road map, and genetic tree metaphors in our design.

Page 5: 1,001,001 Faces user testing plan for face navigation interface Tzu-Pei Grace Chen Sidney Fels Human Communication Technologies Lab.

Face Wheel Design

• Our interface is configural navigation

• Navigation from average face to distinct

• No verbal description of faces.

• Key feature

Page 6: 1,001,001 Faces user testing plan for face navigation interface Tzu-Pei Grace Chen Sidney Fels Human Communication Technologies Lab.

User Testing Design

• 2 experiments• Within subject test• Factorial design• Task: face matching• Plan to use subjects

with no face recognition impairment.

• Play time after the talk.

Page 7: 1,001,001 Faces user testing plan for face navigation interface Tzu-Pei Grace Chen Sidney Fels Human Communication Technologies Lab.

Experiment 1

• hypothesis– Is it easier to navigate on texture & shape

correlated axes?

interface

axes

wheel

fixed slider

T-S correlated T-S uncorrelated

Page 8: 1,001,001 Faces user testing plan for face navigation interface Tzu-Pei Grace Chen Sidney Fels Human Communication Technologies Lab.

Experiment 2

• Hypotheses– 1. Does the wheel interface work better than

the slider interfaces?– 2. Does the increase of the face space

resolution increase the level of difficulty?

interface

dimension

wheel

fixed slider

dynamic slider

36

fine coarse Resolution

Page 9: 1,001,001 Faces user testing plan for face navigation interface Tzu-Pei Grace Chen Sidney Fels Human Communication Technologies Lab.

Questionnaire

• Experiment 1– What’s your ethnic background?

• Experiment 2– Which interface do you prefer and why?– Is your preferred interface the one you are

most familiar with?– Are there situations you would give different

preferences to each interface?

Page 10: 1,001,001 Faces user testing plan for face navigation interface Tzu-Pei Grace Chen Sidney Fels Human Communication Technologies Lab.

Analysis of Data

• Factorial design calls for ANOVA• For dependent variable, “face closeness”, each face

matching result is given score for ANOVA process• Dependent variable, time, will be analysed

separately.

Page 11: 1,001,001 Faces user testing plan for face navigation interface Tzu-Pei Grace Chen Sidney Fels Human Communication Technologies Lab.

Confounding variables

• Everyone has different experience with faces and hence different face space in their head.

• Other race effect• Subjects improve

from familiarity with faces.

Page 12: 1,001,001 Faces user testing plan for face navigation interface Tzu-Pei Grace Chen Sidney Fels Human Communication Technologies Lab.

The End

Thoughts and Suggestions?

Page 13: 1,001,001 Faces user testing plan for face navigation interface Tzu-Pei Grace Chen Sidney Fels Human Communication Technologies Lab.

Face trivia

• Graeme’s theorem: we’re attracted to faces like ourselves and the opposite.

• Face transplant will no longer be science fiction.