Unsupervised Stereo Matching Network For VHR Remote Sensing Images Based On Error Prediction
This addresses the challenge of expensive and scarce Lidar data for remote sensing applications, offering a more scalable solution, though it is incremental as it builds on existing unsupervised learning approaches.
The paper tackles the problem of limited ground truth data for stereo matching in remote sensing by proposing an unsupervised network for very-high-resolution images, achieving superior accuracy compared to other unsupervised methods and better generalization than supervised models on US3D and WHU-Stereo datasets.
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.