Sparse-to-Dense: Depth Prediction from Sparse Depth Samples and a Single Image
This addresses the problem of unreliable monocular depth estimation for robotics and computer vision applications, offering a practical solution with incremental improvements.
The paper tackles dense depth prediction by combining sparse depth samples with a single RGB image, using a deep regression network, resulting in a 50% reduction in root-mean-square error on the NYU-Depth-v2 dataset and an increase in reliable predictions from 59% to 92% on the KITTI dataset.
We consider the problem of dense depth prediction from a sparse set of depth measurements and a single RGB image. Since depth estimation from monocular images alone is inherently ambiguous and unreliable, to attain a higher level of robustness and accuracy, we introduce additional sparse depth samples, which are either acquired with a low-resolution depth sensor or computed via visual Simultaneous Localization and Mapping (SLAM) algorithms. We propose the use of a single deep regression network to learn directly from the RGB-D raw data, and explore the impact of number of depth samples on prediction accuracy. Our experiments show that, compared to using only RGB images, the addition of 100 spatially random depth samples reduces the prediction root-mean-square error by 50% on the NYU-Depth-v2 indoor dataset. It also boosts the percentage of reliable prediction from 59% to 92% on the KITTI dataset. We demonstrate two applications of the proposed algorithm: a plug-in module in SLAM to convert sparse maps to dense maps, and super-resolution for LiDARs. Software and video demonstration are publicly available.