Current: A Platform for Urban Archive

From Design Computation
Jump to: navigation, search

DC I/O 2020 poster by PROVIDES NG, ELI JOTEVA, YA NZI.


Abstract

The quality of our built environment is often assessed through records in architectural archives. Yet, traditional architectural archives do not include records of occupants and events but only drawings and models of buildings, which does not give a comprehensive overview of the qualities and impacts of the design. Livestream data from media platforms often consist of information about our built environment and its events. How can architects utilise livestream data to archive, reconstruct and preserve moments of our built environment in real-time to support future architectural research and studies?

Provoked by related institutional works, ‘Current’ focuses on democratising reconstruction techniques to facilitate a collective contribution to urban archives. Instead of using high end technologies and softwares that are only available to institutions and corporates, ‘Current’ tested a number of low-end sensors (i.e. mobile phones, Kinect, etc.), open-sourced AI algorithms and photogrammetry frameworks that are readily available to any architects or individuals for environment/event reconstruction. The aim is to formulate a platform that allows individuals to collectively gather and process data. ‘Current’ records the workflow and the final pipeline used to produce the 12-min prototype of the urban reconstruction.

Keywords

Digital Platform, Urban Archive, Environment reconstruction, AI image processing, Livestream.

Topics

3D Sensing Cameras, Artificial Intelligence in Design, Climate change, Future of the internet, Generative Adversarial Networks, Media and Telecommunications, Planning, Urban Design, Visual and Spatial Modelling, Visualization & Communication;

Reference

DOI: https://doi.org/10.47330/DCIO.2020.VWEG6251

Video Presentation: https://youtu.be/WH5nZrVfCPQ

Full text: Maciel, A. (Ed.), 2020. Design Computation Input/Output 2020, 1st ed. Design Computation, London, UK. ISBN: 978-1-83812-940-8, DOI:10.47330/DCIO.2020.QPRF9890