On the Readability of Boundary Labeling Lukas Barth, Andreas Gemsa, - PowerPoint PPT Presentation

On the Readability of Boundary Labeling Lukas Barth, Andreas Gemsa, Benjamin Niedermann, Martin N¨ ollenburg 1 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Motivation https://de.wikipedia.org/wiki/Auge#/media/File:Eye scheme.svg 2 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Motivation Choroid Sclera Zonular fibres Cornea Iris Pupil Anterior chamber Posterior chamber Ciliary muscle Optic nerve Lens Retina Vitreous humour https://de.wikipedia.org/wiki/Auge#/media/File:Eye scheme.svg 2 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Motivation Choroid Sclera Zonular fibres Cornea Iris Pupil Anterior chamber Posterior chamber Ciliary muscle Optic nerve Lens Retina Outline: Vitreous humour Discussion on mathematical models. First user study on readability of boundary labeling. https://de.wikipedia.org/wiki/Auge#/media/File:Eye scheme.svg 2 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Mathematical Models 3 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Typical Formalization Given: Rectangle R Points P = { p 1 , . . . , p n } in R . For each p i ∈ P : label ℓ i given by bounding box. R p 2 p 4 p 1 p 3 ℓ 2 = ℓ 1 = Snake Elephant ℓ 3 = Cow ℓ 4 = Mouse 4 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Typical Formalization Given: Find: Placement of labels such that Rectangle R labels lie outside of R Points P = { p 1 , . . . , p n } in R . labels touch the border of R For each p i ∈ P : no two labels overlap ∃ simple curve λ i in R label ℓ i given by bounding box. connecting p i with ℓ i . R R Snake p 2 p 2 p 4 p 4 Mouse p 1 p 1 Elephant p 3 p 3 Cow Typicially: ℓ 2 = ℓ 1 = Snake Elephant crossing free ℓ 3 = Cow ℓ 4 = Mouse total length minimization 4 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Typical Formalization Given: Find: Placement of labels such that Rectangle R labels lie outside of R Points P = { p 1 , . . . , p n } in R . labels touch the border of R For each p i ∈ P : no two labels overlap ∃ simple curve λ i in R label ℓ i given by bounding box. connecting p i with ℓ i . Leader R R Snake p 2 p 2 p 4 p 4 Mouse p 1 p 1 Elephant p 3 p 3 Cow Typicially: ℓ 2 = ℓ 1 = Snake Elephant crossing free ℓ 3 = Cow ℓ 4 = Mouse total length minimization 4 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Typical Formalization Given: Find: Placement of labels such that Rectangle R labels lie outside of R Points P = { p 1 , . . . , p n } in R . labels touch the border of R For each p i ∈ P : no two labels overlap ∃ simple curve λ i in R label ℓ i given by bounding box. Many possible parameters: connecting p i with ℓ i . Label size. Allowed positions of labels. R R Snake p 2 p 2 . . . p 4 p 4 Mouse Important parameter: Type of leaders p 1 p 1 Elephant p 3 p 3 Cow Typicially: crossing free ℓ 2 = ℓ 1 = Snake Elephant total length minimization ℓ 3 = Cow ℓ 4 = Mouse 5 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Typical Examples of Leaders straight line L-shaped diagonal S-shaped 6 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Typical Examples of Leaders straight line L-shaped diagonal S-shaped 6 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Typical Examples of Leaders straight line L-shaped diagonal S-shaped 6 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Typical Examples of Leaders straight line L-shaped diagonal S-shaped 6 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Typical Examples of Leaders straight line L-shaped diagonal S-shaped 6 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Algorithmic Results Year Paper Year Paper Leader Type other s po do opo s × × × 2004 Bekos et al. 2011 Bekos et al. Gemsa et al. × × × 2005 Ali et al. × 2012 Fink et al. 2006 Bekos et al. × 2013 Bekos et al. 2008 Lin et al. ⋆ ⋆ Kindermann et al. × × 2009 Benkert et al. 2014 Huang et al. × × Lin et al. × Kindermann et al. 2010 Bekos et al. × × 2015 L¨ offler et al. × Bekos et al. Lin ⋆ L¨ offler and N¨ ollenburg × × N¨ ollenburg et al. 7 On the Readability of Boundary Labeling On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg ollenburg

Algorithmic Results Year Paper Year Paper Leader Type other s po do opo s × × × 2004 Bekos et al. 2011 Bekos et al. Gemsa et al. × × × 2005 Ali et al. Papers on × 2012 Fink et al. 2006 Bekos et al. × 2013 Bekos et al. 2008 Lin et al. ⋆ ⋆ Kindermann et al. × × 2009 Benkert et al. 2014 Huang et al. S-shaped straight line L-shaped diagonal other types × × Lin et al. × Kindermann et al. 3 9 5 9 5 2010 Bekos et al. × × 2015 L¨ offler et al. × Bekos et al. Lin ⋆ L¨ offler and N¨ ollenburg × × N¨ ollenburg et al. 8 On the Readability of Boundary Labeling On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg ollenburg

User Study on Boundary Labeling 9 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

How to Measure Readability? Task A Cat Given: Highlighted label. Dog Find: Corresponding point. Mouse Task B Cat Given: Highlighted point. Dog Find: Corresponding label. Mouse Measure: Response time of user. Success rate of user. 10 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Research Questions For dense sets of points Hypothesis 1a: better ⋆ than straight line L-shaped S-shaped diagonal Hypothesis 1b: better ⋆ than L-shaped S-shaped diagonal ⋆ with respect to response time and success rate 11 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Research Questions For sparse sets of points Hypothesis 2: better ⋆ than L-shaped S-shaped straight line diagonal ⋆ with respect to response time and success rate 12 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Stimuli Density of Points: dense sparse uniform N ( µ, σ large ) N ( µ, σ small ) 13 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Stimuli Density of Points: dense sparse uniform N ( µ, σ large ) N ( µ, σ small ) Size: small: 15 points large: 30 points 13 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Stimuli Density of Points: dense sparse uniform N ( µ, σ large ) N ( µ, σ small ) Size: small: 15 points large: 30 points Type of Leaders: straight line L-shaped S-shaped diagonal 13 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Stimuli Density of Points: dense sparse uniform N ( µ, σ large ) N ( µ, σ small ) Size: small: 15 points large: 30 points Type of Leaders: straight line L-shaped S-shaped diagonal Task A Task B Tasks: Cat Cat Dog Dog Mouse Mouse 13 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Stimuli Density of Points: dense sparse uniform N ( µ, σ large ) N ( µ, σ small ) Size: small: 15 points large: 30 points Type of Leaders: straight line L-shaped S-shaped diagonal Task A Task B Tasks: Cat Cat Dog Dog Mouse Mouse Degree of Difficulty: 3 Levels for choice of leaders. 13 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Stimuli Density of Points: dense sparse uniform N ( µ, σ large ) N ( µ, σ small ) Size: small: 15 points large: 30 points Type of Leaders: 4 × 2 × 4 × 2 × 3 = 144 stimuli per participant! straight line L-shaped S-shaped diagonal Task A Task B Tasks: Cat Cat Dog Dog Mouse Mouse Degree of Difficulty: 3 Levels for choice of leaders. 14 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Examples of Stimuli straight line L-shaped S-shaped diagonal 15 points, sparse distribution 15 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg

Procedure Digital Questionnaire 1. Tutorial 2. 144 stimuli (randomized order) 3. Questionnaire on preferences (identical hardware and conditions) 31 Participants: 6 female 25 male 20-30 years old Two participants stated that they have fundamental knowledge about labeling. 16 On the Readability of Boundary Labeling L. Barth, A. Gemsa, B. Niedermann, N. N¨ ollenburg