Extracts of Local Distance

Generating New Spaces From Existing Architecture

Date:

2014

Using a vast archive of fragmented architectural photographs, Extracts of Local Distance (ELD) constructs multilayered collages with a unique, abstract point of view. The ELD algorithm digitally analyzes the original images according to their vanishing points and reassembles them in an entirely new aesthetic beyond the perspectives of photographer and architect.

Project Phases:

Strategy | Concept | Design | Execution

Role:

Research | Digital Strategy | Interaction and Information Design | Art Direction | Graphic Design | Software Development | Print Production Coordination

Visual data challenges:

How can image recognition algorithms create new insights and perspectives from existing photographic archives?

Extracts of Local Distance is a research project that demonstrates a new approach to processing and visualizing already existing data sets – in this case, a strategy to compose synthetic architectural photographs which allow the viewer to gain a fresh perspective on familiar places and iconic buildings. The resulting images shake off the restrictions placed on our perception of space by creating the illusion of an entirely new spatial reality.

Based on the initial analysis of the source material, the images are digitally sliced into pieces which retain the original information of their positions relative to the vanishing point and lines of perspective. Forming a large data pool, the fragments can be newly aligned in endless variations, establishing a unique viewpoint each time.

To compose a collage, a perspective grid is defined and covered with matching image segments. Utilizing the database structure already in place in many archives adds another contextual layer: by defining keywords which describe the content of the original photographs – e.g. materials, function or colors –, the selection of segments for the final composition can be influenced.