Shape Priors for Real-Time Monocular Object Localization in Dynamic Environments

Download Video


AbstractóReconstruction of dynamic objects in a scene is a highly challenging problem in the context of SLAM. In this paper, we present a real-time monocular object localization system that estimates the shape and pose of dynamic objects in real-time, using video frames captured from a moving monocular camera. Although the problem seems to be ill-posed, we demonstrate that, by incorporating prior knowledge of the object category, we can obtain more detailed instance-level reconstructions. As opposed to earlier object model specifications, the proposed shape-prior model leads to the formulation of a Bundle Adjustment-like optimization problem for simultaneous shape and pose estimation. Leveraging recent successes of Convolutional Neural Networks (CNNs) for object keypoint localization, we present a CNN architecture that performs precise keypoint localization. We then demonstrate how these keypoints can be used to recover 3D object properties, while accounting for any 2D localization errors and self-occlusion. We show significant performance improvements compared to state-of-the-art monocular competitors for 2D keypoint detection, as


    Main Paper