The Earth ain't Flat: Monocular Reconstruction of Vehicles on Steep and Graded Roads from a Moving Camera

Accurate localization of other traffic participants is a vital task in autonomous driving systems. State-of-the-art systems employ a combination of sensing modalities such as RGB cameras and LiDARs for localizing traffic participants, but most such demonstrations have been confined to plain roads. We demonstrate, to the best of our knowledge, the first results for monocular object localization and shape estimation on surfaces that do not share the same plane with the moving monocular camera. We approximate road surfaces by local planar patches and use semantic cues from vehicles in the scene to initialize a local bundle-adjustment like procedure that simultaneously estimates the pose and shape of the vehicles, and the orientation of the local ground plane on which the vehicle stands as well. We evaluate the proposed approach on the KITTI and SYNTHIA-SF benchmarks, for a variety of road plane configurations. The proposed approach significantly improves the state-of-the-art for monocular object localization on arbitrarily-shaped roads. *The first two authors contributed equally to this work. 1Junaid Ahmed Ansari, Sarthak Sharma, Anshuman Majumdar, and K. Madhava Krishna are with the Robotics Research Center, KCIS, IIIT Hyderabad, India. [email protected] 2J. Krishna Murthy is with Montreal Institute of Learning Algorithms (MILA), Universite de Montreal, Canada.