BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:Australia/Canberra
BEGIN:DAYLIGHT
DTSTART:20251005T030000
TZOFFSETFROM:+1000
TZOFFSETTO:+1100
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=10
TZNAME:AEDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20260405T020000
TZOFFSETFROM:+1100
TZOFFSETTO:+1000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=4
TZNAME:AEST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20251110T225143Z
UID:8ADC5F7E-8EF7-470B-8135-1FABE3C5703A
DTSTART;TZID=Australia/Canberra:20251110T140000
DTEND;TZID=Australia/Canberra:20251110T150000
DESCRIPTION:Title: Weakly Supervised Learning for Remote Sensing Object Seg
 mentation\nDate: Monday\, 10 November 2025\nTime: 2:00 PM to 3:00 PM AEDT\
 nVenue: LT04\, Building 30\, UNSW Canberra (ADFA)\nOnline Access:\n[Weakly
  Supervised Learning for Remote Sensing Object Segmentation | Meeting-Join
  | Microsoft Teams](https://teams.microsoft.com/l/meetup-join/19%3ameeting
 _MWI1NWRhMTgtYzc1Mi00M2QyLWE3Y2EtZWU3ZDFlZmM2M2Jl%40thread.v2/0?context=%7
 b%22Tid%22%3a%223ff6cfa4-e715-48db-b8e1-0867b9f9fba3%22%2c%22Oid%22%3a%223
 b29eade-2027-49a1-aaeb-cace27fee532%22%7d)\n\nAbstract:\nWeakly supervised
  object segmentation such as photovoltaic panel mapping in remote sensing 
 suffers from noisy pseudo labels that erode feature discrimination and red
 uce accuracy. This talk presents three complementary advances to counterac
 t label noise and strengthen generalization. Together\, these methods deli
 ver consistent gains and translate to practical benefits for energy GIS\, 
 renewable energy siting optimization\, and ecological and environmental as
 sessment.\n\nSpeaker(s): Jue Zhang\n\nRoom: LT04\, Bldg: Building 30\, Nor
 thcott Drive\, UNSW Canberra\, Canberra\, Australian Capital Territory\, A
 ustralia\, 2600\, Virtual: https://events.vtools.ieee.org/m/512266
LOCATION:Room: LT04\, Bldg: Building 30\, Northcott Drive\, UNSW Canberra\,
  Canberra\, Australian Capital Territory\, Australia\, 2600\, Virtual: htt
 ps://events.vtools.ieee.org/m/512266
ORGANIZER:aiairen.mail@unsw.edu.au
SEQUENCE:67
SUMMARY:Weakly Supervised Learning for Remote Sensing Object Segmentation 
URL;VALUE=URI:https://events.vtools.ieee.org/m/512266
X-ALT-DESC:Description: <br /><p>Title: Weakly Supervised Learning for Remo
 te Sensing Object Segmentation &nbsp\;<br>Date: Monday\, 10 November 2025 
 &nbsp\;<br>Time: 2:00 PM to 3:00 PM AEDT &nbsp\;<br>Venue: LT04\, Building
  30\, UNSW Canberra (ADFA) &nbsp\;<br>Online Access:&nbsp\;</p>\n<div><a d
 ir="auto" tabindex="0" href="https://teams.microsoft.com/l/meetup-join/19%
 3ameeting_MWI1NWRhMTgtYzc1Mi00M2QyLWE3Y2EtZWU3ZDFlZmM2M2Jl%40thread.v2/0?c
 ontext=%7b%22Tid%22%3a%223ff6cfa4-e715-48db-b8e1-0867b9f9fba3%22%2c%22Oid%
 22%3a%223b29eade-2027-49a1-aaeb-cace27fee532%22%7d" target="_blank" rel="n
 oopener noreferrer">Weakly Supervised Learning for Remote Sensing Object S
 egmentation &nbsp\; | Meeting-Join | Microsoft Teams</a></div>\n<p>Abstrac
 t: &nbsp\;<br>Weakly supervised object segmentation such as photovoltaic p
 anel mapping in remote sensing suffers from noisy pseudo labels that erode
  feature discrimination and reduce accuracy. This talk presents three comp
 lementary advances to counteract label noise and strengthen generalization
 . Together\, these methods deliver consistent gains and translate to pract
 ical benefits for energy GIS\, renewable energy siting optimization\, and 
 ecological and environmental assessment.</p>
END:VEVENT
END:VCALENDAR