We present a novel algorithm for self-supervised monocular depth completion. Our approach is based on training a neural network that requires only sparse depth measurements and corresponding monocular video sequences without dense depth labels. Our self-supervised algorithm is designed for challenging indoor environments with textureless regions, glossy and transparent surface, non-Lambertian surfaces, moving people, longer and diverse depth ranges and scenes captured by complex ego-motions. Our novel architecture leverages both deep stacks of sparse convolution blocks to extract sparse depth features and pixel-adaptive convolutions to fuse image and depth features. We compare with existing approaches in NYUv2, KITTI, and NAVERLABS indoor datasets, and observe 5-34 % improvements in root-means-square error (RMSE) reduction.
@article{arxiv.2011.04977,
title = {SelfDeco: Self-Supervised Monocular Depth Completion in Challenging Indoor Environments},
author = {Jaehoon Choi and Dongki Jung and Yonghan Lee and Deokhwa Kim and Dinesh Manocha and Donghwan Lee},
journal= {arXiv preprint arXiv:2011.04977},
year = {2021}
}