Cited 0 time in webofscience Cited 17 time in scopus

Depth completion with deep geometry and context guidance

Title
Depth completion with deep geometry and context guidance
Authors
Lee, Byeong-UkJeon, Hae-GonIm, SunghoonKweon, In So
DGIST Authors
Im, Sunghoon
Issue Date
2019-05-20
Citation
2019 International Conference on Robotics and Automation, ICRA 2019, 3281-3287
Type
Conference
ISBN
9781538660263
Abstract
In this paper, we present an end-to-end convolutional neural network (CNN) for depth completion. Our network consists of a geometry network and a context network. The geometry network, a single encoder-decoder network, learns to optimize a multi-task loss to generate an initial propagated depth map and a surface normal. The complementary outputs allow it to correctly propagate initial sparse depth points in slanted surfaces. The context network extracts a local and a global feature of an image to compute a bilateral weight, which enables it to preserve edges and fine details in the depth maps. At the end, a final output is produced by multiplying the initially propagated depth map with the bilateral weight. In order to validate the effectiveness and the robustness of our network, we performed extensive ablation studies and compared the results against state-of-the-art CNN-based depth completions, where we showed promising results on various scenes.
URI
http://hdl.handle.net/20.500.11750/14458
DOI
10.1109/ICRA.2019.8794161
Publisher
Institute of Electrical and Electronics Engineers Inc.
Related Researcher
  • Author Im, Sunghoon Computer Vision Lab.
  • Research Interests Computer Vision; Deep Learning; Robot Vision
Files:
There are no files associated with this item.
Collection:
Department of Electrical Engineering and Computer ScienceComputer Vision Lab.2. Conference Papers


qrcode mendeley

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE