BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:GStreamer NVIDIA memory in Jetson: looking for an efficient way to
  process video frames when handling raw CUDA kernels
DTSTART;VALUE=DATE-TIME:20241007T145000Z
DTEND;VALUE=DATE-TIME:20241007T151000Z
DTSTAMP;VALUE=DATE-TIME:20260717T233232Z
UID:indico-contribution-346@indico.freedesktop.org
DESCRIPTION:Speakers: Diego Nieto Munoz (Fluendo)\nMixing NVIDIA Jetson me
 mory and the raw CUDA one is not straightforward. In this talk\, we presen
 t the steps we follow to develop GStreamer plugins based on CUDA  frame pr
 ocess. In order to work with this kind of memory\, several data conversion
 s and mappings are required. These mappings are expensive\, so we based ou
 r filters on a custom CUDA mapped memory pool to get a noticeable speed-up
 .\n\nhttps://indico.freedesktop.org/event/9/contributions/346/
LOCATION: Room 2
URL:https://indico.freedesktop.org/event/9/contributions/346/
END:VEVENT
END:VCALENDAR