The Right (Angled) Perspective: Improving the Understanding of Road Scenes Using Boosted Inverse Perspective Mapping
This work addresses a specific bottleneck in perception for autonomous vehicles, offering an incremental improvement over existing IPM methods.
The paper tackles the problem of unnatural blurring and stretching in bird's-eye-view images generated by Inverse Perspective Mapping for autonomous vehicles, presenting an adversarial learning approach that produces sharper features and more homogeneous illumination while removing dynamic objects, resulting in improved scene understanding tasks as demonstrated on the Oxford RobotCar Dataset.
Many tasks performed by autonomous vehicles such as road marking detection, object tracking, and path planning are simpler in bird's-eye view. Hence, Inverse Perspective Mapping (IPM) is often applied to remove the perspective effect from a vehicle's front-facing camera and to remap its images into a 2D domain, resulting in a top-down view. Unfortunately, however, this leads to unnatural blurring and stretching of objects at further distance, due to the resolution of the camera, limiting applicability. In this paper, we present an adversarial learning approach for generating a significantly improved IPM from a single camera image in real time. The generated bird's-eye-view images contain sharper features (e.g. road markings) and a more homogeneous illumination, while (dynamic) objects are automatically removed from the scene, thus revealing the underlying road layout in an improved fashion. We demonstrate our framework using real-world data from the Oxford RobotCar Dataset and show that scene understanding tasks directly benefit from our boosted IPM approach.