BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//programme.europython.eu//europython-2024//speaker//PRFLC
 Z
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-europython-2024-T3KP3H@programme.europython.eu
DTSTART;TZID=CET:20240711T160500
DTEND;TZID=CET:20240711T165000
DESCRIPTION:Ever wondered how location planning is done to build city infra
 structure? Or when there is a disaster\, how do we determine the possible 
 affected areas and send reinforcements there? We require overhead imagery 
 for that\, which we mainly obtain from satellites.\nEuropean Space Agency 
 has sent various satellites however\, the dataset from these satellites is
  huge and may even contain multiple bands from the electromagnetic spectru
 m. Large AI models have a huge potential in this domain\, if they are deve
 loped to work well with this dataset.\nThere are a lot of pre-trained Gene
 rative & Large Vision models on platforms like HuggingFace\, Kaggle\, etc.
 \, but these models do not integrate well with a specific domain like sate
 llite datasets\, hence the need to train or fine-tune them. In this talk\,
  we are going to see from where we can access open satellite datasets\, fi
 ne-tune various Vision Models and Multimodals on it\, and examine the foll
 owing applications:\n\n- Perform Zero-Shot classification and object detec
 tion on satellite images with human language input using Multimodal models
 .\n- Image-to-image translation on satellite imagery using generative visi
 on models.
DTSTAMP:20260617T073920Z
LOCATION:North Hall
SUMMARY:Earth Observation through Large Vision Models - Mayank Khanduja
URL:https://programme.europython.eu/europython-2024/talk/T3KP3H/
END:VEVENT
END:VCALENDAR