Virtual top-view camera calibration for accurate object representation

Simple perspective transforms are used in Advanced Driver Assistance Systems (ADAS) in vehicles for converting captured images of lateral, front, and rear views into a synthesized top-view image, which displays the scene as viewed from above the vehicle. These transforms, however, are typically deri...

Full description

Saved in:
Bibliographic Details
Published in:2016 IEEE Southwest Symposium on Image Analysis and Interpretation (SSIAI) pp. 21 - 24
Main Authors: Pan, Janice, Appia, Vikram, Bovik, Alan C.
Format: Conference Proceeding
Language:English
Published: IEEE 01-03-2016
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Simple perspective transforms are used in Advanced Driver Assistance Systems (ADAS) in vehicles for converting captured images of lateral, front, and rear views into a synthesized top-view image, which displays the scene as viewed from above the vehicle. These transforms, however, are typically derived using a basic calibration procedure that is only capable of correctly mapping ground-plane points in captured images to their corresponding locations in the top-view image, and any off-the-ground points look distorted. We present a new method, which uses four orthogonally-placed fisheye cameras, for calibrating a top-view image, in which objects and off-the-ground points are accurately represented.
DOI:10.1109/SSIAI.2016.7459165