Scribd
Upload
25 views

02 Introduction To Data Visualisation (Part 2)

The document discusses the visualisation pipeline and characteristics of data. It describes the standard stages in visualization including data acquisition, enhancement, mapping, and rendering. It also covers data attributes like dimensionality, source, and characteristics that influence visualization method selection.

Uploaded by

Misya Syafiqah Mohd Safian
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
25 views

02 Introduction To Data Visualisation (Part 2)

The document discusses the visualisation pipeline and characteristics of data. It describes the standard stages in visualization including data acquisition, enhancement, mapping, and rendering. It also covers data attributes like dimensionality, source, and characteristics that influence visualization method selection.

Uploaded by

Misya Syafiqah Mohd Safian
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 25

Introduction to Data Visualisation

(Part 2)

Visualisation Pipeline and About data

Farhan bin Mohamed

[email protected]
Previous Topics: Lecture #1
Content of Lecture 1
• Visualisation – Definition
• Application Examples
• Goals: for presentation, analysis, exploration (new
Insights)
• SciVis vs. InfoVis vs. Infographics
Overview of Lecture #2
Content of Lecture 2:
• Visualisation Pipeline
• About Data
• Visualisation Examples
The Visualisation
Pipeline
Standard processing stages in
Visualisation

4
Vis. Pipeline – Overview
Data Acquisition

Data is produced or acquired

Data Enhancement

Data is prepared (or “pre-processed”)

Visualisation Mapping

Data is mapped to geometric primitives

Rendering (3D ↔ 2D / nD)


Data is transformed to images
VISUALISATION

5
Vis. Pipeline – Stage 1
Data Acquisition

Data is produced or acquired

Data acquisition = Data produced


• Measurement (e.g., CT, MRI).
• Simulation (e.g., Computational Fluid Dynamics
Simulation (CFD)).
• Modelling (e.g., Computational Aided Design (CAD),
Dynamic systems).
6
Vis. Pipeline – Stage 2
Data is produced or acquired

Data Enhancement

Data is prepared (or “pre-processed”)

Data enhancement = Data preparation/pre


processing
• Filtering (e.g., smoothing (noise filtering))
• Resampling = modify grid representation
• Derive new data (e.g., gradients)
• Data interpolation
7
Vis. Pipeline – Stage 3
Data is prepared (or “pre-processed”)

Visualisation Mapping

Data is mapped to geometric primitives

Visualisation mapping = Data is now visible


• Compute isosurface
• Compute graph layout
• Compute glyphs and or icons
• Compute voxel (primitive) attributes: colour, transparency, ...
• Geometric primitives: points, lines, triangles, cubes,
tetrahedra, etc.

8
Vis. Pipeline – Stage 4
Data is mapped to geometric primitives

Rendering (3D ↔ 2D / nD)


Data is transformed to images

VISUALISATION

Rendering = Presentation with Computer Graphics (CG)


• Visibility calculation
• Shading
• Compositing (accumulate transparency and colour values)
• Animation

9
Please read the following definition
in the link

https://link.springer.com/referenceworkentry/10.1007%2F978-0-387-39940-9_1133
About Data
Characteristics, attributes, and
dimensionality
Data – General
• Data
• The focal point of visualisation: visualisation starts
with data.
• Data: a “driving factor” (along with the user) with
respect to the choice and attributes of the
visualisation method.
• Important questions:
§ Where is the dimensionality of the data?
§ Where does the data come from?
§ Which visualisation method(s) makes sense?
Data Dimensionality
• What dimensionality is the data?
§ Inherent spatial domain (SciVis):
§ 1D, 2D, 2.5D or 3D dimensionality given
§ Instantaneous or time-dependent (+1D)
§ Examples: medical data, data from flow simulation,
GIS-data, etc.
• No inherent spatial reference (InfoVis):
§ Abstract data,
§ Spatial arrangement through visualisation
§ Example(s): database(s)
Data Characteristics
• What characteristics does the data have?
§ Data Types:
• Scalar = numeric (natural, integer, rational, real number,
complex numbers)
• Non-numeric (nominal, ordinal values)
• Multi-dimensional values (n-dim, vectors, tensors)
§ Characteristics: dimensionality, domain (upper
and lower bounds)
Data Presentation
• How can data be presented?
– Inherent spatial domain?
• Yes → Use inherent domain?
• No → Which spatial domain?
– Which dimensions are used for what?
• Relationship between dimensionality ↔ data
characteristics
• Available presentation space (2D/3D/4D)
• Where is the focus?
• What can be left out?
Data vs. Visualisation Dimensionality
Vis
Data 1D 2D 3D

1D y = f(x) space curve x(t)

contour lines
2D

3D slices
Visualisation
Visualisation Examples
Data Description Visualisation Examples

N1 → R1 Series of values bar chart, pie chart, etc.

R1 → R1 function graph

R2 → R1 function over R2 2-D heightfields in 3D, contour lines


in 2D
R2 → R2 2D-vector field Hedgehog plot, LIC, streamlets, etc.

R3 → R1 3D-Density values iso-surface in 3D, volume rendering

(N1 → )Rn tuple quantities parallel coordinates, glyphs, icons,


(multi-attribute data) etc.
Visualisation Examples (1)
Data Description Visualisation Examples

N1 → R1 Series of values bar chart, pie chart, etc.


Visualisation Examples (2)
Data Description Visualisation Examples

R1 → R1 function Graph (line, curve)


Visualisation Examples (3)
Data Description Visualisation Examples

R2 → R1 function over R2 2-D heightfields in 3D, contour lines


in 2D
Visualisation Examples (4)
Data Description Visualisation Examples

R2 → R2 2D-vector field Hedgehog plot, LIC, streamlets, etc.


Visualisation Examples (5)
Data Description Visualisation Examples

R3 → R3 3-D Flow Streamlines, Streamsurfaces


Visualisation Examples (6)
Data Description Visualisation Examples

R3 → R1 3D-Density Values Iso-surfaces in 3D, volume rendering

Center for Medical Image Science and Visualization


(Anders Ynnerman)
Visualisation Examples (7)
Data Description Visualisation Examples

(Nn → )Rn tuple quantities Parallel coordinates, glyphs, icons,


(multi-attributes data) metaphor, etc.

Twitter users as organisms →


(Kunal Anand)
These Twitter creatures float around, each one representing a person (user). Size and behavior are represent rank, which
comes from Infochimps. What each action, movement, and size actually represents in the data, I'm not sure. I'm guessing
higher rank means bigger creatures and activity is mostly retweets and mentions.
Acknowledgements
• We thank the following people for the composition/materials
in this lecture are based on lectures/materials/text of:
§ Robert S. Laramee
§ Inge Tastl
§ M. Eduard Gröller
§ Helwig Hauser

You might also like