Top-level Scene Information Extraction from Eye-level View Images

Autor:	Adina Magda Florea, Mihai Nan, Alexandra Stefania Ghita, David-Traian Iancu
Rok vydání:	2021
Předmět:	Computer science business.industry Image segmentation Prediction system computer.software_genre Tracking (particle physics) Perspective transformation Information extraction Control system Trajectory Computer vision Segmentation Artificial intelligence business computer
Zdroj:	CSCS
DOI:	10.1109/cscs52396.2021.00029
Popis:	People tracking and trajectory prediction generate more research interest recently given the advancement of autonomous devices. To obtain accurate results, the behaviour modelling needs to take into consideration the interactions people have with the environment. This paper proposes a method to extract scene information from 2D eye-level images which can then be passed to a trajectory prediction system. It combines a bird-eye view perspective transformation with a segmentation technique to identify global obstacles and possible walking areas.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::f2eaca8b760044ca5771a7708ae04669 https://doi.org/10.1109/cscs52396.2021.00029 Zobrazit plný text záznamu