BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pyconde-pydata-2024//speaker//TCFUAP
BEGIN:VTIMEZONE
TZID:Europe/Berlin
BEGIN:DAYLIGHT
DTSTART:20230425T000000
TZNAME:CEST
TZOFFSETFROM:+0200
TZOFFSETTO:+0200
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20231029T030000
RDATE:20241027T030000
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20240331T030000
RDATE:20250330T030000
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
SUMMARY:FlixBus CitySnap: How we use GenAI and not only to collect captiva
 ting images for cities and confirm their locations - Andrei Chernov
DTSTART;TZID=Europe/Berlin:20240424T103000
DTEND;TZID=Europe/Berlin:20240424T110000
DTSTAMP:20260815T200208Z
UID:pretalx-pyconde-pydata-2024-ECCJAG@pretalx.com
DESCRIPTION:Have you ever wondered how travel e-commerce companies gather 
 photos of cities? While I can't speak for everyone\, I will demonstrate th
 e innovative approach we are using at Flix.\n\nIn recent years\, text-to-t
 ext models like ChatGPT and text-to-image models such as DALL-E 3 have bec
 ome increasingly integrated into various industries. The main aim of these
  initiatives is typically to generate text or images. In our presentation\
 , we propose a slightly different approach to leveraging these models comm
 ercially. Our objective is to gather images for thousands of cities that i
 nspire travel. We utilize ChatGPT to tailor prompts for our business requi
 rements\, enabling efficient image retrieval through API queries from free
  stock image services. Then we apply image-to-text models to confirm the i
 mages' locations. Finally\, we need to adjust the resolution of images for
  display across various platforms\, such as social media campaigns on Inst
 agram\, email marketing\, and on our website. To achieve this\, we have us
 ed an automated cropping service to get images in the required aspect rati
 os\, followed by Lanczos sampling for downscaling the images. This integra
 tion of cutting-edge models has resulted in an automated\, highly flexible
  process that aligns with varied business needs. Our approach is cost-effi
 cient\; processing several hundred cities amounts to only a few euros\, an
 d we have utilized commonly available services\, making replication easy f
 or everyone.
LOCATION:A1
URL:https://pretalx.com/pyconde-pydata-2024/talk/ECCJAG/
END:VEVENT
END:VCALENDAR