4 ms·
Show HN: Unsupervised Depth Completion from Visual-Inertial Odometry
- xfei91 7y agoWe trained a deep learning model to densify a sparse point cloud of a 3-D scene given an RGB image of the scene. Compared to other learning methods, ours has 80% fewer parameters while outperforming others thanks to the mesh triangulation and linear interpolation used as pre-processing steps. The paper describing our method has been accepted by the International Conference on Robotics and Automation (ICRA), 2020.