Abstract

We present an approach to dense depth estimation froma single monocular camera that is moving through a dy-namic scene. The approach produces a dense depth mapfrom two consecutive frames. Moving objects are recon-structed along with the surrounding environment. We pro-vide a novel motion segmentation algorithm that segmentsthe optical flow field into a set of motion models, each withits own epipolar geometry. We then show that the scenecan be reconstructed based on these motion models by optimizing a convex program. The optimization jointly reasons about the scales of different objects and assembles the scene in a common coordinate frame, determined up to a global scale. Experimental results demonstrate that the presented approach outperforms prior methods for monocular depth estimation in dynamic scenes.