Unsupervised Depth Completion from Visual Inertial Odometry
This addresses depth completion for robotics and autonomous systems using low-cost sensors, but it is incremental as it builds on existing unsupervised methods with a new dataset.
The paper tackles the problem of inferring dense depth from sparse depth points and camera motion using visual-inertial odometry, achieving state-of-the-art performance on the unsupervised KITTI depth completion benchmark.
We describe a method to infer dense depth from camera motion and sparse depth as estimated using a visual-inertial odometry system. Unlike other scenarios using point clouds from lidar or structured light sensors, we have few hundreds to few thousand points, insufficient to inform the topology of the scene. Our method first constructs a piecewise planar scaffolding of the scene, and then uses it to infer dense depth using the image along with the sparse points. We use a predictive cross-modal criterion, akin to `self-supervision,' measuring photometric consistency across time, forward-backward pose consistency, and geometric compatibility with the sparse point cloud. We also launch the first visual-inertial + depth dataset, which we hope will foster additional exploration into combining the complementary strengths of visual and inertial sensors. To compare our method to prior work, we adopt the unsupervised KITTI depth completion benchmark, and show state-of-the-art performance on it. Code available at: https://github.com/alexklwong/unsupervised-depth-completion-visual-inertial-odometry.