Saved in:
Bibliographic Details
Main Authors: Jiang, Liting, Xiang, Yuming, Wang, Feng, You, Hongjian
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2408.07419
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917748949385216
author Jiang, Liting
Xiang, Yuming
Wang, Feng
You, Hongjian
author_facet Jiang, Liting
Xiang, Yuming
Wang, Feng
You, Hongjian
contents Stereo matching in remote sensing has recently garnered increased attention, primarily focusing on supervised learning. However, datasets with ground truth generated by expensive airbone Lidar exhibit limited quantity and diversity, constraining the effectiveness of supervised networks. In contrast, unsupervised learning methods can leverage the increasing availability of very-high-resolution (VHR) remote sensing images, offering considerable potential in the realm of stereo matching. Motivated by this intuition, we propose a novel unsupervised stereo matching network for VHR remote sensing images. A light-weight module to bridge confidence with predicted error is introduced to refine the core model. Robust unsupervised losses are formulated to enhance network convergence. The experimental results on US3D and WHU-Stereo datasets demonstrate that the proposed network achieves superior accuracy compared to other unsupervised networks and exhibits better generalization capabilities than supervised models. Our code will be available at https://github.com/Elenairene/CBEM.
format Preprint
id arxiv_https___arxiv_org_abs_2408_07419
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Unsupervised Stereo Matching Network For VHR Remote Sensing Images Based On Error Prediction
Jiang, Liting
Xiang, Yuming
Wang, Feng
You, Hongjian
Computer Vision and Pattern Recognition
Stereo matching in remote sensing has recently garnered increased attention, primarily focusing on supervised learning. However, datasets with ground truth generated by expensive airbone Lidar exhibit limited quantity and diversity, constraining the effectiveness of supervised networks. In contrast, unsupervised learning methods can leverage the increasing availability of very-high-resolution (VHR) remote sensing images, offering considerable potential in the realm of stereo matching. Motivated by this intuition, we propose a novel unsupervised stereo matching network for VHR remote sensing images. A light-weight module to bridge confidence with predicted error is introduced to refine the core model. Robust unsupervised losses are formulated to enhance network convergence. The experimental results on US3D and WHU-Stereo datasets demonstrate that the proposed network achieves superior accuracy compared to other unsupervised networks and exhibits better generalization capabilities than supervised models. Our code will be available at https://github.com/Elenairene/CBEM.
title Unsupervised Stereo Matching Network For VHR Remote Sensing Images Based On Error Prediction
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2408.07419