BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//talks.geoinfo-lab.org//UNZN7S
BEGIN:VTIMEZONE
TZID:+07
BEGIN:STANDARD
DTSTART:20000101T000000
RRULE:FREQ=YEARLY;BYMONTH=1
TZNAME:+07
TZOFFSETFROM:+0700
TZOFFSETTO:+0700
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-foss4g-asia-2024-UNZN7S@talks.geoinfo-lab.org
DTSTART;TZID=+07:20241217T111500
DTEND;TZID=+07:20241217T113000
DESCRIPTION:The rapid development of autonomous vehicles (AVs) demands stro
 ng perception systems capable of reliably recognizing and classifying obje
 cts in complicated urban environments. To improve the reliability and prec
 ision of 3D object identification\, combining camera and LiDAR sensors has
  emerged as a potential solution. This research describes a multimodal fus
 ion framework that uses camera pictures and LiDAR point clouds to accompli
 sh high-performance 3D object recognition in urban circumstances. Camera s
 ensors provide precise color and texture information necessary for identif
 ying traffic signs\, pedestrians\, and cars\, whereas LiDAR provides exact
  depth measurements required for interpreting object geometry and spatial 
 relationships. The indicated fusion technique improves detection accuracy 
 by using these sensors' enhancing strengths\, especially in difficult sett
 ings such as occlusions and fluctuating illumination conditions taking the
  KITTI open dataset. Here the OpenPCDet is used for 3D object detection an
 d MMDetection for 2D detection as open library. Beside this the Facebook A
 I Research\, Detectron2 flexible framework for 2D and 3D object detection 
 tasks is more popular too. \n\nFusion is accomplished via a well built arc
 hitecture that aligns and combines data from both modalities at various st
 ages of the detection pipeline\, such as feature extraction\, region propo
 sal\, and classification. Advanced deep learning techniques\, such as conv
 olutional neural networks\, are used to process and integrate multimodal i
 nput. Experimental results show that 3D object detection outperforms singl
 e-modality techniques in terms of robustness and precision\, particularly 
 when recognizing small and partially occluded objects.
DTSTAMP:20260416T201422Z
LOCATION:Auditorium Hall 1
SUMMARY:Camera-LiDAR Fusion for multimodal 3D Object detection in Autonomou
 s Vehicles - Badri Raj Lamichhane
URL:https://talks.geoinfo-lab.org/foss4g-asia-2024/talk/UNZN7S/
END:VEVENT
END:VCALENDAR