Point Clouds
Sets of x, y, and z coordinates
Cloudism
Christophe Girot, 2019
Designers adopting cloudism will step into
an overwhelmingly convincing simulacra of physical reality, space and time;
this will enhance their understanding of site,
and yield a stronger awareness of ambient aspects and cues.
Medium
Point Clouds
are algorithmically mediated, indexical encodings
of space and color at an instance in time
A point cloud is an index; it is a trace of its subject, a record of light as a space. It can also be iconic, an immediately legible rendering of a scene. As the result of an algorithm encoded as numeric data, the point cloud is also a collection of symbols. It is contingent on framing and chance. The indexical, iconic, symbolic, and contingent nature of the point cloud give it meaning – connecting it to the scene, author, and algorithms. They are the foundation for aesthetic expression in this emerging medium.
Indexicality
Indices are physically and causally connected to what they represent
In Peirce's semiotics an index is a sign that is physically or causally connected to what it represents. As its physical imprint, a footprint is an index for a foot. Smoke is an index for a fire; if you see smoke, then you know that it has been caused by a fire. A photograph, as a record of light captured on photographic film or a digital image sensor, is an index that is both physically and causally connected by rays of light to the scene that it represents. Point clouds, like photographs, are indexical because they are physically and causally connected by pulses of light to what they represent.
Iconicity
Icons look like what they represent
Point clouds are iconic because they look like what they represent. Their iconic nature is unique because of their fidelity, inherent three dimensionality, and resulting immersivity. Their indexical link to the scene captured imbues them with a authenticity that grounds their iconic claim to faithfully represent reality.
Hyperreality
Point clouds' indexicality, their truth claim, seems so authoritative as to be hyperreal, displacing reality – they are used as definitive records, as evidence, as substitutes for things lost, and to fabricate replicas. What does it mean to "step into an overwhelmingly convincing simulacrum of physical reality, space and time?" (Girot) Will hyperreal simulacra really enhance landscape architects' "understanding of site, and yield a stronger awareness of ambient aspects and cues?" (Girot) Or will these simulacra usher in an aesthetic crisis of creativity, an "end of the imaginary" (Baudrillard 1994)?
Pointillist
Point clouds, however, are imperfect icons with noise, systematic errors, occlusions in capture, and voids. By their very nature as a collection of points in space, there is empty space between each point, a discontinuity that does not exist in reality. Their incompleteness makes point clouds a fundamentally abstract medium, countering their hyperrealism. There is rich aesthetic tension between the point cloud’s hyperreal detail and depth and the painterly abstraction with which it is rendered.
Symbolism
Symbols are given meaning by convention
Symbols are signs given meaning only by conventional standards. While the scene captured in a point cloud may have symbolic significance, point clouds themselves are inherently symbolic since they have been encoded as digital data. Scanned point clouds are transformed from light into waves of electrons, processed by algorithms, and stored as symbolic data structures.
The transcription of photons into bits algorithmically mediates meaning. The point cloud remains a pure perceptual index of the referent, while as a conceptual index it becomes symbolically mediated by algorithms. The conceptual index - the trace of ideas – becomes a mark not only of the scene, but also of the author and the algorithms. A mark to be decoded and interpreted.
Contingency
The contingent is subject to change and chance
In its uncertainty, the contingent connotes the freedom of possibility and thus the potential for creativity. When laser pulses trace the contours of a scene, what is captured is contingent upon atmospheric conditions, the play of the light and shadow, gusts of wind, the fall of rain, and chance passersby.
Contingent Indices
As indices, point clouds are contingent on what happened to be captured, their meaning relying in part on the accidental.
Contingent Icons
As icons, the meanings of point clouds are contingent upon aesthetic choices and artistic interpretations.
Contingent Symbols
As algorithmically mediated symbols, the meanings of point clouds are contingent upon their processing. This transforms point clouds into reflexive indices, that refer not only to the scene captured, but also to their author and algorithms.
Characteristics
Algorithmic: Transcribed from photons into bitsInteractive: Not just an image, but rather dataHyperreal: Displaces reality with authoritative dataAbstract: Pointillist in its incompletenessContingent Indices: Depend on what capturedContingent Symbols: Depend on digital processing
The indexical, iconic, and symbolic contingency of point clouds is generative, forming a creative space between author, algorithm, and environment. The generative potential of point clouds as a medium lies in creatively transgressing the constraints of the technology, for it is the algorithmic nature and limits of the medium that create room for meaning.
Landscape Dynamics
Events: A record of space during an instance of timeTime Series: A series of instances recording change over time
As a medium, point clouds can capture and represent change over time in unique ways. With point clouds time can be recorded and represented as either a discrete event or a series of events.
Phenomena & Ephemera
The point cloud of Rosedown captures the complexity and temporality of landscape in the convoluted branching structure of live oaks, drapery of Spanish moss, seasonality of flowers, texture of bark, and a profusion of leaves. With its overwhelming detail and noticeable occlusions, this point cloud highlights a relationship between the noisy and imperfect, but rigidly structured formal gardens and the natural logic of irregular plants, all abstracted in a pointillist style that coheres noisy details into an impressionistic rendering of light. These details begin to show the landscape’s dynamic character, its semiotic richness beyond its tectonic form and symbolism.
Time Series
A time series of point clouds can be collected by repeatedly surveying a site with lidar or photogrammetry. Change over time can be represented visually through a sequence of iconic, yet analytic point cloud renderings such as transects, each representing a discrete event in time.
Time Series
Time series of point clouds can also be numerically analyzed to study spatiotemporal phenomena. Analyses can be rendered as three-dimensional scatter plots, another type of point cloud.
Time Series
These scatterplots are purely conceptual indices whose perceptual links to the landscape have disappeared. Despite their intensive algorithmic processing, they still retain indexical links to the landscape as data that record traces of its processes. They represent a trace of an idea such as fluxes of carbon, rather than the look and physical structure of the landscape. By modeling scatter plots as point clouds, abstract data visualizations can be combined with reductive, yet iconic graphics to form a composition in a single medium.
