Journal of Digital Landscape Architecture

Seite drucken

Automated Recording of Human Movement Using an Artificial Intelligence Identification and Mapping System

Few designers engage in post-occupancy evaluation of built works and those that do have limited tools available for capturing robust and unbiased data. This paper proposes a system in development to record continuous and unbiased site usage inventories with high accuracy and rich metadata for further analysis. To date the system can use multiple cameras with time synced video feeds to reidentify people in the landscape and project their locations onto a two-dimensional map. The unique multi-camera system avoids accuracy and identification issues inherent in single camera systems while recording unbiased date without privacy concerns. The inventory programming system is discussed, and preliminary outputs and data are presented along with issues encountered and future goals for the project.

Autor / Author: Barbarash, David; Rasheed, Moiz; Gupta, Apoorva; Wang, Tong
Institution / Institution: Purdue University, Indiana/USA; Purdue University, Indiana/USA; Purdue University, Indiana/USA; Purdue University, Indiana/USA
Seitenzahl / Pages: 12
Sprache / Language: Englisch
Veröffentlichung / Publication: JoDLA – Journal of Digital Landscape Architecture, 7-2022
Tagung / Conference: Digital Landscape Architecture 2022 – Hybrid Landscapes
Veranstaltungsort, -datum / Venue, Date: Harvard University, Cambridge Mass, USA 09-06-22 - 10-06-22
Schlüsselwörter (de):
Keywords (en): Post-occupancy evaluation, behavioral mapping, artificial intelligence, site inventory
Paper review type: Full Paper Review
DOI: doi:10.14627/537724007
7391 -