BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Phoenix
X-LIC-LOCATION:America/Phoenix
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0700
TZNAME:MST
DTSTART:19700101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20241014T203103Z
LOCATION:FLW Salon J
DTSTART;TZID=America/Phoenix:20240913T083000
DTEND;TZID=America/Phoenix:20240913T093000
UID:HFESAM_ASPIRE - Presented by HFES_sess288_DEMO105@linklings.com
SUMMARY:Discriminator: Exploring Highlighting Techniques in Vast Natural I
 mage Databases
DESCRIPTION:Demonstration\n\nZeth duBois and Steffen Werner (University of
  Idaho)\n\nThe number of available natural images for machine learning is 
 staggering. Programs like Microsoft’s AI For Earth “MegaDetector” can pre-
 filter large datasets by reliably detecting the presence of wildlife, huma
 ns, or vehicles and determine bounding boxes for targets. Outputs can then
  guide further classification. \n\nOften human observers are providing the
  ground truths which are then used as input for machine learning algorithm
 s. Given the large number of image classifications humans might be involve
 d in, optimization of the visual presentation of target AOIs is paramount.
  To explore this design space we developed Discriminator, an interactive w
 ebapp that allows users to dynamically filter images based on MegaDetector
  output and parametrically explore different presentation techniques. By d
 irectly applying image manipulation to random selections of natural images
  we can test visualization strategies on the fly. As an extensible tool, D
 iscriminator can provide additional services, including gathering online i
 mage ratings, asset sorting, and image manipulation.\n\nTrack: Perception 
 and Performance
END:VEVENT
END:VCALENDAR