An Efficient Vehicle Localization Method by Using Monocular Vision

Accurate localization of surrounding vehicles helps drivers to perceive surrounding environment, which can be obtained by two parameters: depth and direction angle. This research aims to present a new efficient monocular vision based pipeline to get the vehicle's location. We proposed a plug-and-play convolutional block combination with a basic target detection algorithm to improve the accuracy of vehicle's bounding boxes. Then they were transformed to actual depth and angle through a conversion method which was deduced by monocular imaging geometry and camera parameters. Experimental results on KITTI dataset showed the high accuracy and efficiency of the proposed method. The mAP increased by about 2% with an additional inference time of less than 5 ms. The average depth error was about 4% for near distance objects and about 7% for far distance objects. The average angle error was about two degrees.

Automated summary

This research presents a new monocular vision-based pipeline for accurate vehicle localization, combining convolutional blocks and target detection algorithms to improve accuracy. By converting bounding boxes to actual depth and angle, the method shows high accuracy and efficiency on the KITTI dataset.