Human to Machine Scene Understanding

Ceylan S., EFE M. Ö.

6th Mediterranean Conference on Pattern Recognition and Artificial Intelligence, MedPRAI 2024, İstanbul, Turkey, 18 - 19 October 2024, vol.1393 LNNS, pp.631-643, (Full Text) identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 1393 LNNS
  • Doi Number: 10.1007/978-3-031-90893-4_42
  • City: İstanbul
  • Country: Turkey
  • Page Numbers: pp.631-643
  • Keywords: Saliency, Scene understanding, Visual attention
  • Hacettepe University Affiliated: Yes

Abstract

Scene understanding is a complex but essential task for humans and machines. Humans have a unique ability to analyze objects in their environment and relate these objects to their background. However, understanding objects and establishing a relationship with background is challenging for machines. They require large amounts of data and complex calculations. This study examines how humans and machines understand visual scenes. The features that humans and machines find saliency during scene understanding are emphasized.