loading
Papers Papers/2022 Papers Papers/2022

Research.Publish.Connect.

Paper

Paper Unlock

Authors: Yong Deng 1 ; Jimin Xiao 2 and Steven Zhiying Zhou 1 ; 3

Affiliations: 1 Department of Electrical and Computer Engineering, National University of Singapore, 117583, Singapore ; 2 Department of Electrical and Electronic Engineering, Xi’an Jiaotong-Liverpool University, Suzhou, Jiangsu, 215123, P.R.China ; 3 National University of Singapore Suzhou Research Institute, Suzhou, Jiangsu, 215123, P.R.China

Keyword(s): Stereo Matching, Depth Estimation, Deep Learning, Dynamic Upsampling.

Abstract: Deep learning based stereo matching networks achieve great success in the depth estimation from stereo image pairs. However, current state-of-the-art methods usually are computationally intensive, which prevents them from being applied in real-time scenarios or on mobile platforms with limited computational resources. In order to tackle this shortcoming, we propose a lightweight real-time stereo matching network for disparity estimation. Our network adopts the efficient hierarchical Coarse-To-Fine (CTF) matching scheme, which starts matching from the low-resolution feature maps, and then upsamples and refines the previous disparity stage by stage until the full resolution. We can take the result of any stage as output to trade off accuracy and runtime. We propose an efficient hourglass-shaped feature extractor based on the latest MobileNet V3 to extract multi-resolution feature maps from stereo image pairs. We also propose to replace the traditional upsampling method in the CTF m atching scheme with the learning-based dynamic upsampling modules to avoid blurring effects caused by conventional upsampling methods. Our model can process 1242 x 375 resolution images with 35-68 FPS on a GeForce GTX 1660 GPU, and outperforms all competitive baselines with comparable runtime on the KITTI 2012/2015 datasets. (More)

CC BY-NC-ND 4.0

Sign In Guest: Register as new SciTePress user now for free.

Sign In SciTePress user: please login.

PDF ImageMy Papers

You are not signed in, therefore limits apply to your IP address 18.191.9.9

In the current month:
Recent papers: 100 available of 100 total
2+ years older papers: 200 available of 200 total

Paper citation in several formats:
Deng, Y.; Xiao, J. and Zhou, S. (2021). A Lightweight Real-time Stereo Depth Estimation Network with Dynamic Upsampling Modules. In Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) - Volume 5: VISAPP; ISBN 978-989-758-488-6; ISSN 2184-4321, SciTePress, pages 701-710. DOI: 10.5220/0010197607010710

@conference{visapp21,
author={Yong Deng. and Jimin Xiao. and Steven Zhiying Zhou.},
title={A Lightweight Real-time Stereo Depth Estimation Network with Dynamic Upsampling Modules},
booktitle={Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) - Volume 5: VISAPP},
year={2021},
pages={701-710},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010197607010710},
isbn={978-989-758-488-6},
issn={2184-4321},
}

TY - CONF

JO - Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) - Volume 5: VISAPP
TI - A Lightweight Real-time Stereo Depth Estimation Network with Dynamic Upsampling Modules
SN - 978-989-758-488-6
IS - 2184-4321
AU - Deng, Y.
AU - Xiao, J.
AU - Zhou, S.
PY - 2021
SP - 701
EP - 710
DO - 10.5220/0010197607010710
PB - SciTePress