Detecting Road Obstacles by Erasing Them [article]

Krzysztof Lis, Sina Honari, Pascal Fua, Mathieu Salzmann
2021 arXiv   pre-print
Vehicles can encounter a myriad of obstacles on the road, and it is impossible to record them all beforehand to train a detector. Instead, we select image patches and inpaint them with the surrounding road texture, which tends to remove obstacles from those patches. We then uses a network trained to recognize discrepancies between the original patch and the inpainted one, which signals an erased obstacle. We also contribute a new dataset for monocular road obstacle detection, and show that our
more » ... pproach outperforms the state-of-the-art methods on both our new dataset and the standard Fishyscapes Lost \& Found benchmark.
arXiv:2012.13633v2 fatcat:df7z43iofjanpafl7gebk7mt7u