@inproceedings{fd5354749e1e4f97b2e4ab0ad6bd04e8,
title = "Bridging Depth Estimation and Completion for Mobile Robots Reliable 3D Perception",
abstract = "Autonomous mobile robots rely on knowing the 3D structure of their environment in order to plan and operate safely. This paper addresses the problem of estimating depth maps from monocular RGB images and any potentially available sparse depth measurements—as would be the case of using inexpensive RGBD or LIDAR sensors. Our approach bridges depth estimation and depth completion by introducing a novel lightweight two-stream encoder-decoder neural network that exploits late fusion and additional output refinements. The novelty of our network is that it can benefit from any available depth data, while not making any assumptions about their actual availability, density or distribution. Our proposed architecture outperforms models with equivalent number of operations in terms of depth accuracy and performs on par with much larger models.",
keywords = "3D perception, Depth completion, Depth estimation",
author = "Dimitrios Arapis and Milad Jami and Lazaros Nalpantidis",
note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.; 10<sup>th</sup> International Conference on Robot Intelligence Technology and Applications, RiTA 2022 ; Conference date: 07-12-2022 Through 09-12-2022",
year = "2023",
doi = "10.1007/978-3-031-26889-2_16",
language = "English",
isbn = "9783031268885",
series = "Lecture Notes in Networks and Systems",
publisher = "Springer",
pages = "169--179",
editor = "Jun Jo and Han-Lim Choi and Marde Helbig and Hyondong Oh and Jemin Hwangbo and Chang-Hun Lee and Bela Stantic",
booktitle = "Robot Intelligence Technology and Applications 7",
}