Automatically generating structured information on the as-is status of facilities from visual data

Despite the large production and collection of digital data, the AEC/FM industry lacks accurate, detailed, consistent and up-to-date knowledge of facilities. Advances in sensing technology allow the acquisition of high-fidelity 3D digital replicas of the as-is status. However, the output of such systems does not provide information on the spatial components, attributes and in-between relationships, thus cannot support downstream tasks (e.g., construction progress monitoring, maintenance activities, etc.). In this talk I will describe recent work on the automated extraction and knowledge-based representation of this information from visual data using Machine Vision and AI, as well as how it improves existing processes and enables novel workflows.

Iro Armeni is a PhD candidate at Stanford University, conducting interdisciplinary research between Civil Engineering and Machine Vision. Her area of focus is on automated semantic and operational understanding of buildings throughout their life cycle using visual data. Prior to enrolling in PhD program, Iro received an MEng in Architecture and Digital Design (University of Tokyo-2011), an MSc in Computer Science (Ionian University-2013), and a Diploma in Architectural Engineering (National Technical University of Athens-2009). She is the recipient of the Google PhD Fellowship on Machine Perception and the Japanese Government (MEXT) scholarship, among other awards. She is also a co-instructor in the class "AI applications in the AEC industry" at Stanford University. Iro has worked as an architect and consultant for both the private and public sector.