Content uploaded by Payam Tabrizian
Author content
All content in this area was uploaded by Payam Tabrizian on Aug 02, 2017
Content may be subject to copyright.
2
Tangible Immersion
for Ecological Design
1 Collaboravely designing and
immersively visualizing a park with
Tangible Landscape
Vaclav Petras
North Carolina State University
Helena Mitasova
North Carolina State University
Ross Meentenmeyer
North Carolina State University
Payam Tabrizian
North Carolina State University
Brendan Harmon
Louisiana State University
Anna Petrasova
North Carolina State University
1
ABSTRACT
We introduce tangible immersion – virtual reality coupled with tangible interacon – to foster inter-
disciplinary collaboraon in a crical, yet creave design process. Integrang tangible, embodied
interacon with geospaal modeling and immersive virtual environments (IVE) can make 3D
modeling fast and natural, while enhancing it with realisc graphics and quantave analycs. We
have developed Tangible Landscape, a technology that links a physical model with a geographic
informaon system and 3D modeling plaorm through a real-me cycle of interacon, 3D scan-
ning, geospaal computaon, and 3D rendering. With this technology landscape architects, other
professionals, and the public can collaboravely explore design alternaves through an iterave
process of intuive ideaon, geocomputaonal analysis, realisc rendering, and crical analysis. This
is demonstrated with a test case for interdisciplinary problem solving in which a landscape architect
and geoscienst use Tangible Landscape to collaboravely design landforms, hydrologic systems,
planng, and a trail network for a browneld site. Using this tangible immersive environment they
rapidly explored alternave scenarios. We discuss how the parcipants used real-me analycs
to collaboravely assess trade-os between environmental and experienal factors, balancing
landscape complexity, biodiversity, remediaon capacity, and aesthecs. Together they explored
how the relaonship between landforms and natural processes aected the performance of the
designed landscape. Technologies that couple tangible geospaal modeling with IVEs have the
potenal to transform the design process by breaking down disciplinary boundaries, but may also
oer new ways to imagine space and democraze design.
3
ACADIA 2017
| DISCIPLINES + DISRUPTION
TOPIC (ACADIA team will ll in)
INTRODUCTION
Recent advances in compung, sensors, and human-computer
interacon are transforming the pracce of design. With compu-
taonal modeling, analysis and simulaon designers can generate
novel forms, explore parametric variaons, and quantavely test
the performance of their designs. Digital fabricaon technologies
allow designers to precisely and rapidly build prototypes and
then complex, high performance structures, while augmented
and virtual reality enable designs to be immersively experienced.
With tangibles – technologies that couple physical objects with
digital data – designers can naturally interact with and trans-
form digital models. Synthesizing these emerging technologies
will enable designers to rapidly generate designs, quantavely
test them, and immersively experience them in a uid, seamless
creave process.
Design problems oen require collaboraon with other experts
like engineers and sciensts – to address the structural tectonics
and energy uxes of a building, the ow of trac and people
through a city, or ow of water and sediment through a land-
scape. Furthermore, design problems also involve non-experts
such as clients, stakeholders, and the general public. Designers’
creave processes, sciensts’ research methods, and public
parcipaon, however, are rarely integrated in an iterave
process. Collaboraon between scienc disciplines – even
mutually dependent disciples like ecology and geomorphology –
can be challenging due to technical, methodological, and lexical
dierences, focuses on dierent spaal and temporal scales, and
gaps in knowledge (Renscheler et al. 2007). It is especially chal-
lenging for ongoing research to connually inform a design – and
vice versa – when the soware, tools, and methods are so dispa-
rate. While landscape architects may use drawing, 3D modeling
and visualizaon, or digital fabricaon to imagine new landscapes,
sciensts such as ecologists and geomorphologists typically
use geospaal modeling and stascal inference to study the
paerns and processes that shape landscapes. Sciensts cannot
apply numerical models to hand drawn sketches or clay models
of topography; they may need to convert their geospaal and
stascal data before designers can work with it in computer
aided design plaorms.
These analycal disconnects have led to slow, complex work-
ows and stymied interdisciplinary collaboraon between
designers and sciensts. When design and research workows
are so complicated non-experts are unlikely to have technical
skills and understanding to meaningfully parcipate in the
process. In this paper we propose methods to bridge disci-
plinary divides and engage the public by synthesizing emerging
technologies such as tangibles, digital fabricaon, and virtual
environments with disciplinary tools and methods to create
integrated workows for creave exploraon and quantave
analysis. Seamless collaboraons that couple creave design and
scienc analysis have great potenal to radically transform the
pracce of ecological design (Figure 1).
BACKGROUND
Tangibles
Tangibles – tangible user interfaces – are systems that couple
physical objects with digital data for more natural, embodied
interacon (Dourish 2001). Tangibles give digital data an interac-
ve, physical form and presence that users can kinaesthecally
sense and manipulate. Advances in sensors, machine vision, and
robocs have radically accelerated the development of tangibles
– including tangibles designed for designers. Recent prototypes
for tangible design include the Collaborave Design Plaorm
(Schubert 2012) and CityScope (MIT Media Lab 2017) for urban
design and Tangible Landscape (Petrasova et al. 2015), the Rapid
Landscape Prototyping Machine (Robinson 2014), and Cyborg
Ecologies (Responsive Arfacts and Environments Lab 2017) for
landscape architecture.
Tangible interfaces for landscape architecture have mainly been
designed for terrain modeling. While terrain modeling is an
important aspect of landscape architecture, tangible interfaces
have not been developed for planng design. While landscape
architects typically use relavely simple techniques to digitally
illustrate planng through photomontage or 2D billboards
in 3D dioramas (Ervin 2001), there are more sophiscated
methods in computer graphics for modeling and rendering
realisc 3D plants including trees and grasses used by the
computer gaming and lm industries. The recent emergence
of aordable head-mounted displays has further accelerated
the development of realisc 3D graphics. Vegetaon modeling
has also been advanced by the development of sophiscated
methods in Geographic Informaon Systems (GIS) for producing
landscape-scale models of vegetaon that quanfy ecological
diversity and fragmentaon (Petras 2017). Seamlessly inte-
grang 3D planng with realisc rendering into the creave
design process could revoluonize the pracce of landscape
architecture.
Immersion
Immersive virtual environments (IVEs) combine immersion and
interacon with virtual environment to enhance presence — a
feeling of being physically present in a non-physical world. By
surrounding users with a connuous stream of visual smuli,
linked to their head and body movements, IVEs enable users to
acvely engage with and beer understand a virtual environ-
ment. In design IVEs have been used as a cost-eecve tool for
enhancing presentaons and improving communicaon between
4
architects, collaborators, and end-users. Furthermore, realisc
representaons can aid designers’ spaal cognive abilies by
reducing the mental eort needed to generate internal repre-
sentaons of designs. Recent advances in CPUS, GPUs, and 3D
modeling and rendering soware have enabled designers to
generate photorealisc representaons that mimic the condi-
ons through which human perceive the designed landscape.
However, generang highly immersive and realisc visualizaons
requires considerable me and eort, especially in the context of
an iterave design process. While photorealisc renderings are
the best way to represent how people will perceive a landscape,
abstract representaons are also essenal tools for communi-
cang conceptual and analycal data.
METHODS
Tangible Landscape
Tangible Landscape – a tangible interface for geospaal modeling
with realisc, real-me 3D rendering (Petrasova et al. 2015,
Tabrizian et al. 2016) – was designed to bridge disciplinary
divides by coupling physical and digital landscape models so
that users can intuively transform and immersively experi-
ence 3D data. It enables 3D sketching informed by real-me
analycs. Conceptually it couples a physical and digital model of
a landscape through a real-me cycle of 3D scanning, geospaal
computaon, 3D modeling, 3D rendering, and projecon (Figure
2). It is powered by GRASS GIS, an open source scienc plat-
form for geoprocessing (GRASS Development Team) – drawing
on its extensive libraries of peer-reviewed scienc models, anal-
yses, and simulaons – and Blender, an open-source plaorm
for 3D modeling, rendering, animaon, and game development
(Blender Development Team) (Figure 3). Tangible Landscape is
an open source project with repositories at hps://github.com/
tangible-landscape/.
Tangible Landscape evolved from Illuminang Clay (Ra et al.
2004) and the Tangible Geospaal Modeling System (Mitasova
et al. 2006, Tateosian et al. 2010). Illuminang Clay, an early
tangible interface developed by the MIT Media Lab, augmented
a laser scanned clay model of a landscape with projected spaal
analycs such as elevaon, slope, aspect, and drain direcon. It
was designed to “streamline the landscape design process and
result in a more eecve use of GIS, especially when distributed
decision-making and discussion with non-experts are involved”
(Ra 2004). The analyses were custom implemented in C++
for the project, but many were adapted from GRASS GIS’ open
source libraries (Piper 2002a). While highly innovave this
system supported a relavely small selecon of basic terrain
analyses, the interacon was limited to surface modicaons,
Tangible Immersion for Ecological Design Tabrizian et al.
2 Tangible Landscape couples a physical and digital model through a cycle of 3D scanning, geospaal computaon, 3D modeling and rendering, and projecon.
5
ACADIA 2017
| DISCIPLINES + DISRUPTION
TOPIC (ACADIA team will ll in)
and the simple, abstract visualizaons limited by the hardware
available at the me had a low degree of realism.
A call for collaboraon with the GRASS GIS community (Piper
2002b) led to a new prototype – the Tangible Geospaal
Modeling System – that was powered by GRASS GIS (Mitasova
2006, Tateosian 2010) with a library of over 350 modules
including comprehensive topographic analyses, solar irradiaon
models, robust overland water ow simulaon, and many others.
While more exible with a large open source library and higher
resoluon visualizaons, this system did not support real-me
feedback, interacon was sll limited to surface modicaons,
and it did not yet support the realisc rendering of vegetaon or
built form.
Building on the Tangible Geospaal Modeling System and
inspired by the Augmented Reality Sandbox (Kreylos 2012),
Tangible Landscape took advantage of new developments in 3D
sensors and malleable materials to replace the expensive laser
scanner with a low-cost Kinect scanner and the clay model with
polymer-enriched sand (Petrasova 2015). Rapid scanning and
imaging by Kinect supported development of a wide range of
modes of interacon such as surface sculpng through depth
sensing and digizing points, areas, and volumes through object
and color detecon. Users can sculpt landforms, place building
blocks, plant trees, draw paths, and more.
3D modeling, 3D rendering, and support for virtual reality
powered by Blender were added to the system to make
3 Soware architecture for coupling GRASS GIS and Blender.
visualizaons more immersive and give users a ground view
experience (Tabrizian 2016). Blender was chosen to power
Tangible Landscape’s 3D modeling and rendering because it has
a GIS add-on, realisc viewport rendering, and a virtual reality
add-on. The BlenderGIS add-on supports the import, export, and
processing of georeferenced data in Blender (domlysz). In Blender
the viewport supports ambient occlusion, ambient lighng,
an-aliasing, and raytracing powered by OpenGL for realisc
on-the-y rendering. The Virtual Reality Viewport add-on
supports head-mounted displays (HMD) with built-in head
tracking including the Oculus Ri DK2 and CV1 (Felinto).
3D planting and realistic rendering
To address planng design, a key aspect of landscape archi-
tecture, Tangible Landscape has now been extended with 3D
planng including trees, shrubs, and groundcover and realisc
real-me rendering. It has new modes of interacon and oers
an enhanced immersive experience. Designers can place patches
of colored felt or draw with a laser pointer to create planng
areas. They can also place color-coded wooden or 3D printed
models of trees to digize individual specimens. Plants are
immediately modeled and rendered, either on a screen or a HMD.
Tangible Landscape uses simple tangible interacons drawing
on childhood play such as sculpng sand, arranging blocks and
markers, and placing felt patches (Figure 4) because people
already know how to do these things – these are intuive acons
using exisng sensorimotor schemas. These physical media –
sand, wood blocks, and felt – have a familiar feel and aesthec,
while the computer graphics supplement them with realism.
6
4 Tangible Landscape supports mulple modes of interacon including a) sculpng,
b) carving, c) placing objects as points, d) drawing lines with laser pointer
e) designang views using a color-coded marker, and f) placing patches of
colored felt as areas.
Dierent species of plants are detected by color. Patches of
colored felt represenng dierent types of vegetaon are
captured as colors by Kinect, grouped using superpixel segmen-
taon (Achanta et al. 2012), and then automacally classied by
vegetaon type in GRASS GIS. The classied planng areas are
then imported into Blender and populated with 3D models of
the designated plants using a parcle system. The 3D plants can
either be procedurally generated based on custom parameters,
procedurally generated with parameters, twigs, and textures from
a plant library such as the Grove (Keulen), or imported from a
library of 3D plant models with texturing mapping such as Xfrog
(Oliver and Lintermann 2010, Xfrog). The scene is interacvely
rendered in real-me.
Designers can immersively explore the landscape that they have
modeled by tangibly idenfying views. They can place a wooden
marker on the physical model to designate both a viewpoint
and a view direcon. The marker is color coded with one end
represenng the viewpoint and the other represenng the view
direcon (Figure 4e). The marker is detected based on eleva-
on change and color. It is imported into Blender as a vector to
align an eye level camera. By simply moving the marker around
designers can quickly explore dierent views, seeing what they
have built and planted as if on the ground.
Aer individual trees or patches of trees have been planted by
the designer and automacally detected and classied, Tangible
Landscape analyses the structure of the landscape, compung
metrics that quantavely describe the paern, distribuon,
spaal organizaon, variaon, and shape of elements in the land-
scape. These metrics are projected as a dashboard of charts or
graphs beside the model as addional feedback for the designer
(Figure 6). The landscape metrics include Shannon’s diversity
index, mean shape index, edge density, and patch number (Baker
et al. 1992).
CASE STUDY
Through a case study we explored how tangible, immersive
interacon can support a collaborave, sciencally informed,
yet creave design process that bridges disciplinary boundaries.
To test a diverse set of interacons we created a hypothecal 1
square kilometer browneld study site with several mounds of
capped, contaminated soil. We tasked a landscape architect and
a geoscienst to design this site as a park. Their objecves were
to retain and phytoremediate as much surface water as possible
in order to minimize the migraon of contaminants, while
enhancing biodiversity, improving aesthecs, and planning a trail
network between two entrances.
The landscape architect and geoscienst used Tangible
Landscape with simple tools and media to develop their
designs. They worked with a polymer-enriched sand model of
the landscape cast with a CNC milled mold, wooden sculpng
tools, colored wooden markers, and colored felt (Figure 1). As
they created their design they received real-me feedback from
Tangible Landscape including maps of geospaal analycs, graphs
with design metrics, and 3D renderings.
To regrade the topography and manage the hydrology of the
site they reshaped the polymeric sand model of the landscape,
sculpng changes by hand. As they sculpted the sand the model
was scanned, interpolated as a digital elevaon model (DEM) and
a contour map was derived and projected over the model. This
new DEM was also used to simulate overland water ow and
ponding, with the results projected over the model along with
the contours, all in real-me, as the changes were made. Shallow
overland water ow was simulated as a diusive wave approxi-
maon of the bivariate form of the St. Venant equaons solved
using a Monte Carlo path sampling method (Mitasova et al.
2004), while ponding was computed using the standard depres-
sion lling technique. Elevaon, contour and water depth maps
were projected onto the model and the total water depth and
Tangible Immersion for Ecological Design Tabrizian et al.
4
a b
c d
e f
7
ACADIA 2017
| DISCIPLINES + DISRUPTION
TOPIC (ACADIA team will ll in)
5
A hydrologic map with topographic
contours (top) and charts for the
surface area and depth of the pond
(boom).
6
A classied landcover map (top)
and landscape structure metrics
(boom) showing the percentage of
remediated soil, number of patches,
species richness, mean patch size,
shannon diversity, and edge index.
7 A map of slope of the boardwalk
(top) and a chart with the prole
of the boardwalk (boom). Red,
yellow, and green signify high,
medium, and low slopes.
area were projected as bar charts beside the model to provide
analycal feedback about the performance of the proposed
changes (Figure 8a).
To design the planng they placed colored coded pieces of felt
represenng patches of trees and small wooden models repre-
senng individual trees. The felt and models were sensed and
classied as vegetaon types by color. The vegetaon – decid-
uous trees, evergreen trees, wetland trees, and shrubs – were
projected on the model as a map with colored polygons and
modeled and rendered in 3D on the display. Metrics for land-
scape structure and diversity – including patch number, richness,
Shannon index, and shape index – and the percentage of water
that could be phytoremediated by wetland planng were
projected as bar charts beside the model. Remediaon potenal
was computed as the intersecon of water and the wetland
planng root zone (Figure 8b).
To plan the trail system they placed wooden markers represenng
waypoints between the entrances. The opmal route between
the entrances and waypoints was computed as the least cost
path across an anisotropic cumulave cost surface of walking
eort (Fontanari 2002). As they placed waypoints, the trail was
dynamically adjusted and projected on the model, the prole of
the trail was projected as a graph beside the model (Figure 8c).
The trail route was extruded to create a 3D boardwalk.
The geoscienst and the landscape architect collaboravely
developed another design (Figure 9). Together they evaluated
the performance of their design using the dashboard with its
landscape structure and remediaon metrics and explored 3D
rendered views of the park from a variety of vantage points.
Aer idenfying many promising viewpoints they planned a trail
network with walkable slopes that would take advantage of the
lakeside experience, while oering panoramic views of the park.
Many design iteraons later, they had developed a looping trail
system around the lake, over one of the islands, and along the
ridges with their commanding views (Figure 9c). Finally each of
them donned an Oculus Ri to explore the nal design with a
walkthrough along the trail (Figure 9d3). Throughout the design
process the geoscienst and the landscape architect had a rich
dialogue, informed by geospaal analycs. Tangible Landscape
helped them to rapidly implement, visualize, evaluate, and
crique their ideas on the y. They were able to collaboravely
develop a high performance design with realiscally rendered
views in just an hour aer having explored many alternave
design ideas including a dozen dierent opons for the trail
network. See a video of their design process at: hps://youtu.be/
akCTeknStmQ.
RESULTS AND REFLECTIONS
The landscape architect and the geoscienst were able to
collaborate eecvely because Tangible Landscape has a large,
easily extendible library of analyses, adequate resoluon for
the design problem, and fast enough feedback. The analyses
used – water ow, ponding, ecological structure, and phytore-
mediaon potenal in this case study – can easily be changed
and customized with GRASS GIS’ API and extensive library of
funcons. The analyses, projected as maps and charts, informed
the design team’s decisions. As they worked they received both
spaal and quantave feedback about the ecological impact of
their intervenons.
With the Kinect sensor mounted 60 cm above the model
Tangible Landscape has 2 mm scanning resoluon with a mean
error of 0.02 mm. Markers and colored patches of felt need to
be at least 0.5 cm2 to be detected. At this scanning resoluon
the modeling media – polymer enriched sand, wooden markers,
and felt – allow for ne-grained tangible interacons.
The rate of geospaal feedback for Tangible Landscape depends
upon the size of the physical model and the analyses used. For
765
8
8 The landscape architect’s design process
9
ACADIA 2017
| DISCIPLINES + DISRUPTION
TOPIC (ACADIA team will ll in)
9 The geoscienst and landscape architect’s design process
10
this case study we used a small 34 cm2 scale model of the 1 km2
study landscape. For this size model the average rate of feedback
for tangible interacon with the water ow simulaon is 2.02 ±
0.05 seconds. It takes 0.97 seconds to scan the 34 cm2 physical
model and interpolate the point cloud as a DEM. Then it takes
another 1.05 ± 0.05 seconds to compute water ow. Tangible
Landscape’s rate of feedback is fast enough for users to make a
change, remove their hands, and then see the result in a rapid,
iterave process of observaon, hypothesis generaon and
tesng, and inference.
While the Blender viewport can update in real-me for simple
scenes, it can take a couple of seconds for more complex scenes
depending upon the number of trees and their polygon count.
Blender’s viewport shading supports raytraced shadows, textures,
environment lighng, and ambient occlusion. It has the level of
realism needed to quickly explore design scenarios. However, for
more sophiscated, higher resoluon rendering users can switch
to Cycles rendering engine, which uses path tracing to simu-
late so shadows, depth of eld, moon blur, causcs, ambient
occlusion, and indirect lighng. Depending on the complexity of
the scene, render sengs, and hardware, a Cycles rendering can
vary from seconds to minutes.
Future work
Planng design with Tangible Landscape could be even richer,
more immersive, and more accessible. In the near future we
intend to improve planng design by creang an open source
library of both procedural and 3D plants, preparing a variety of
plant communies for use, developing tools for plant selecon
using Blender’s GUI, and developing an ecological succession and
phytoremediaon models in GRASS GIS. We intend to create an
open source library of other 3D assets, implement architectural
and urban modeling with Tangible Landscape, develop a more
sophiscated dashboard, and support mulple render engines
for viewport shading, viewport rendering, and rendering. We
plan to use Blender’s game engine for richer interacon with the
virtual environment. We also plan to integrate digital fabricaon
into the system for physical feedback from computaon so that
a landscape evoluon simulaon could change the shape of the
physical model or an epidemiological simulaon could remove
plant markers from the model.
CONCLUSIONS
Tangible Landscape – a tangible interface for 3D modeling
landscapes – enables landscape architects to rapidly, iteravely
develop designs, realiscally visualize landscapes, and collaborate
with others. Coupling physical and virtual models of landscapes
can synthesize analog and digital design methods, while coupling
scienc and architectural models can bridge disciplinary
boundaries. Such tangible interfaces for design can make the
design process so natural and intuive that experts from other
disciplines or even the public can easily parcipate, collabo-
ravely understanding and reimagining space in novel ways.
Technologies like Tangible Landscape that invite public parcipa-
on and engagement in the design process could democraze
design, giving people a means of expression, giving them agency
in their built environment, and encouraging collecve creavity.
REFERENCES
Achanta, Radhakrishna, Appu Shaji, Kevin Smith, Aurelien Lucchi,
Pascal Fua, and Sabine Susstrunk. 2012. “SLIC Superpixels
Compared to State-of-the-Art Superpixel Methods.” IEEE Trans.
Paern Anal. Mach. Intell. 34 (11). Washington, DC, USA
Baker, William L, and Yunming Cai. 1992. “The r.le Programs for
Mulscale Analysis of Landscape Structure Using the GRASS
Geographical Informaon System.” Landscape Ecology 7 (4):
291–302.
Blender Online Community. “Blender – a 3D Modelling and
Rendering Package.” Accessed May 8, 2017. hp://www.blender.
org.
Deussen, Oliver, and Bernd Lintermann. 2010. Digital Design of
Nature: Computer Generated Plants and Organics. Springer.
domlysz. “BlenderGIS.” Accessed May 11, 2017. hps://github.
com/domlysz/BlenderGIS.
Dourish, Paul. 2001. Where the Acon Is: The Foundaons of
Embodied Interacon. Cambridge, MA: MIT Press.
Ervin, Stephen, and Hope Hasbrouck. 2001. Landscape
Modeling: Digital Techniques for Landscape Visualizaon.
McGraw-Hill Professional.
Felinto, Dalai. “Virtual Reality Viewport.” Accessed May 11, 2017.
hps://github.com/dfelinto/virtual_reality_viewport.
GRASS Development Team. “GRASS GIS.” Accessed May 8, 2017.
hps://grass.osgeo.org.
Keulen, Wybren van. “The Grove.” Accessed May 8, 2017. hps://
www.thegrove3d.com/.
Mitasova, Helena, Chris Thaxton, Jaroslav Hoerka, Richard
McLaughlin, Amber Moore, and Lubos Mitas. 2004. “Path
Sampling Method for Modeling Overland Water Flow, Sediment
Transport, and Short Term Terrain Evoluon in Open Source GIS.”
11
ACADIA 2017
| DISCIPLINES + DISRUPTION
TOPIC (ACADIA team will ll in)
Developments in Water Science 55: 1479–90.
Mitasova, Helena, Lubos Mitas, Carlo Ra, Hiroshi Ishii, Jason
Alonso, and Russell S. Harmon. 2006. “Real-Time Landscape
Model Interacon Using a Tangible Geospaal Modeling
Environment.” IEEE Computer Graphics and Applicaons 26 (4):
55–63.
MIT Media Lab. 2017. “CityScope.” hp://cp.media.mit.edu/
cityscope/.
Petras, Vaclav, Douglas J. Newcomb, and Helena Mitasova. 2017.
“Generalized 3D Fragmentaon Index Derived from Lidar Point
Clouds.” Open Geospaal Data, Soware and Standards 2 (1).
Springer Open.
Petrasova, Anna, Brendan Harmon, Vaclav Petras, and Helena
Mitasova. 2015. Tangible Modeling with Open Source GIS.
Springer Internaonal Publishing.
Piper, Ben, Carlo Ra, and Hiroshi Ishii. 2002. “Illuminang
Clay: A Tangible Interface with Potenal GRASS Applicaons.” In
Proceedings of the Open Source GIS - GRASS Users Conference
2002. Trento, Italy.
Piper, Ben, Carlo Ra, and Hiroshi Ishii. 2002. “Illuminang Clay:
A 3-D Tangible Interface for Landscape Analysis.” In Proceedings
of the SIGCHI Conference on Human Factors in Compung
Systems - CHI ’02, 355. Minneapolis: ACM Press.
Ra, Carlo, Yao Wang, Hiroshi Ishii, Ben Piper, Dennis
Frenchman, John P Wilson, A Stewart Fotheringham, and Gary J
Hunter. 2004. “Tangible User Interfaces (TUIs): A Novel Paradigm
for GIS.” Transacons in GIS 8 (4): 407–21.
Renschler, Chris S., Marn W. Doyle, and Marn Thoms. 2007.
“Geomorphology and Ecosystems: Challenges and Keys for
Success in Bridging Disciplines.” Geomorphology 89 (1–2): 1–8.
Responsive Arfacts and Environments Lab. 2017. “Cyborg
Ecologies.” hps://vimeo.com/134275416.
Robinson, Alexander. 2014. “Calibrang Agencies in a Territory
of Instrumentality: Rapid Landscape Prototyping for the Owens
Lake Dust Control Project.” In ACADIA 14: Projects of the 34th
Annual Conference of the Associaon for Computer Aided
Design in Architecture (ACADIA), 143–46. Los Angeles, CA.
Schubert, Gerhard, Eva Arnger, Violin Yanev, Frank Petzold,
and Gudrun Klinker. 2012. “3D Virtuality Sketching: Interacve
3D-Sketching Based on Real Models in a Virtual Scene.” In
Proceedings of the 32nd Annual Conference of the Associaon
for Computer Aided Design in Architecture (ACADIA), 32:409–
18. San Francisco, CA.
Tateosian, Laura G, Helena Mitasova, Brendan A Harmon,
Brent Fogleman, Katherine Weaver, and Russell S Harmon.
2010. “TanGeoMS: Tangible Geospaal Modeling System.” IEEE
Transacons on Visualizaon and Computer Graphics
Tabrizian, Payam, Anna Petrasova, Brendan A Harmon, Vaclav
Petras, Helena Mitasova, and Ross K Meentemeyer. 2016.
“Immersive Tangible Geospaal Modeling.” In Proceedings of the
24th ACM SIGSPATIAL Internaonal Conference on Advances in
Geographic Informaon Systems. San Francisco, CA: ACM.
Underkoer, John, and Hiroshi Ishii. 1999. “Urp: A Luminous-
Tangible Workbench for Urban Planning and Design.” In CHI ’99
Proceedings of the SIGCHI Conference on Human Factors in
Compung Systems, 386–93. New York, New York, USA: ACM
Press.
Xfrog. “Xfrog.” Accessed May 8, 2017. hp://xfrog.com/.
Payam Tabrizian is a PhD student in college of Design and Center
for Geospaal Analycs at NC State University. He received a Master
of Urbanism and Strategic Planning from K.U Leuven, Belguim, and a
Bachleor in Architecture from Azad University, Iran. Payam is a Virtual
reality developer with specec interest on integrang geospaal model-
ling and realme 3D visualizaon to enhance urban design and research.
Brendan Harmon is an assistant professor of landscape architec-
ture at Louisiana State University’s Robert Reich School of Landscape
Architecture. He received a Master of Landscape Architecture from the
Harvard Graduate School of Design, a Master of Philosophy in Geography
and the Environment from the University of Oxford, and a PhD in Design
from NC State University.
Anna Petrosova is a PhD Candidate at the Center for Geospaal
Analycs at NC State University focusing on integrang dynamic geospa-
al modeling with tangible user interfaces. Anna is an acve member of
GRASS Development team developing and maintaining GRASS GIS, an
open source GIS.
Vaclav Petras Vaclav Petras has a a master degree in geoinformacs
from Czech Technical University and is currently pursuing PhD in Marine,
Earth, and Atmospheric Sciences at NC State University. His research
revolves around lidar point clouds, terrain, geospaal simulaons and
modeling. He is an advocate for open science and an open source devel-
oper and enthusiast.
12
13
ACADIA 2017
| DISCIPLINES + DISRUPTION
TOPIC (ACADIA team will ll in)