Use of First and Third Person Views for Deep Intersection Classification
This work addresses road topology classification for autonomous driving systems, representing an incremental improvement by integrating existing view types.
The paper tackles intersection classification using monocular on-board vision by combining first-person and third-person views into a unified deep learning framework, resulting in a scheme that outperforms previous methods with minimal measurements.
We explore the problem of intersection classification using monocular on-board passive vision, with the goal of classifying traffic scenes with respect to road topology. We divide the existing approaches into two broad categories according to the type of input data: (a) first person vision (FPV) approaches, which use an egocentric view sequence as the intersection is passed; and (b) third person vision (TPV) approaches, which use a single view immediately before entering the intersection. The FPV and TPV approaches each have advantages and disadvantages. Therefore, we aim to combine them into a unified deep learning framework. Experimental results show that the proposed FPV-TPV scheme outperforms previous methods and only requires minimal FPV/TPV measurements.