BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Virtualizing existing industrial faciltiies - Eva Agapaki\, PhD Ca
 ndidate\, Engineering Department
DTSTART:20180309T150000Z
DTEND:20180309T160000Z
UID:TALK102211@talks.cam.ac.uk
CONTACT:Karen Mitchell
DESCRIPTION:The cost of modelling existing industrial facilities is curren
 tly considered to counteract the benefits of the model in managing and ret
 rofitting the facility. 90% of the modelling cost is typically spent on la
 bour for converting point cloud data to the final model\, hence reducing t
 he cost is only possible by automating this step. Previous research has su
 ccessfully validated methods for modelling specific object types such as c
 ylinders. Yet modelling is still prohibitively expensive. During this talk
 \, the most important object types of industrial facilities will be identi
 fied by ranking them according to their frequency of appearance and the ma
 n-hours required for modelling in a state of the art software\, EdgeWise. 
 This work is the first to rank objects according to their priority for aut
 omated modelling. These are straight pipes\, electrical conduit and circul
 ar hollow sections and constitute more than 80 % of industrial plants on a
 verage. This is significant because state-of-the-art practice has achieved
  semi-automated cylinder detection saving 64 % of their manual modelling t
 ime for the case studies investigated. Automated detection and semantic cl
 assification methods for the recognition of the abovementioned objects wil
 l be analyzed.
LOCATION: Cambridge University Engineering Department\, LT6
END:VEVENT
END:VCALENDAR