34 papers with code • 5 benchmarks • 4 datasets
The Depth Completion task is a sub-problem of depth estimation. Instead of knowing nothing about the scene, the Depth Completion task has strong priors on scene depth. The goal of Depth Completion is to fill in the depth on pixels where there is no valid depth. The pixels where the input sparse depth map has a valid value should remain unchanged during the process.
Depth completion, the technique of estimating a dense depth image from sparse depth measurements, has a variety of applications in robotics and autonomous driving.
Many standard robotic platforms are equipped with at least a fixed 2D laser range finder and a monocular camera.
However, we additionally propose a fusion method with RGB guidance from a monocular camera in order to leverage object information and to correct mistakes in the sparse input.
Ranked #3 on Depth Completion on KITTI Depth Completion
With the rise of data driven deep neural networks as a realization of universal function approximators, most research on computer vision problems has moved away from hand crafted classical image processing algorithms.
In this paper, we propose a deep learning architecture that produces accurate dense depth for the outdoor scene from a single color image and a sparse depth.
To address these challenges, we present ClearGrasp -- a deep learning approach for estimating accurate 3D geometry of transparent objects from a single RGB-D image for robotic manipulation.