Learning what is above and what is below: horizon approach to monocular obstacle detection
This addresses obstacle detection for robots or vehicles using monocular cameras, but it appears incremental as it builds on horizon-based methods with uncertainty.
The paper tackles monocular obstacle detection by using self-supervised learning to classify pixels above and below the horizon line, leveraging classifier uncertainty for detection. Preliminary results show it works in different environments, such as segmenting road and sky on KITTI and floor on a flying dataset.
A novel approach is proposed for monocular obstacle detection, which relies on self-supervised learning to discriminate everything above the horizon line from everything below. Obstacles on the path of a robot that keeps moving at the same height, will appear both above and under the horizon line. This implies that classifying obstacle pixels will be inherently uncertain. Hence, in the proposed approach the classifier's uncertainty is used for obstacle detection. The (preliminary) results show that this approach can indeed work in different environments. On the well-known KITTI data set, the self-supervised learning scheme clearly segments the road and sky, while application to a flying data set leads to the segmentation of the flight arena's floor.