Stereo matching of high-resolution satellite images (HRSI) is still a fundamental but challenging task in the field of photogrammetry and remote sensing. Recently, deep learning (DL) methods have demonstrated the potential for stereo matching on public benchmark datasets. Currently, mainstream stereo matching models depend on quantities of training data with ground truth. However, datasets for stereo matching of satellite images are scarce, which profoundly blocks the application of DL in this field. To facilitate further research, this paper publishes a large-scale dataset, termed WHU-Stereo, for stereo matching DL network training and testing. This dataset is created by using airborne LiDAR point clouds and high-resolution stereo imageries taken from the Chinese GaoFen-7 satellite (GF-7). Occlusions should be seriously considered in the generation of ground-truth disparities. We propose an occlusion removal technique, which can adapt to point clouds with different densities and shows high potential in training data preparation. The WHU-Stereo dataset contains more than 1700 epipolar rectified image pairs, which cover six areas in China and includes various kinds of landscapes. We have assessed the accuracy of ground-truth disparity maps, and it is shown that our dataset achieves comparable precision compared with existing state-of-the-art stereo matching datasets. To verify its feasibility, in experiments, the hand-crafted SGM algorithm and recent DL networks have been tested on the dataset. Experimental results show that the WHU-Stereo dataset can serve as a challenging benchmark for stereo matching of high-resolution satellite images and performance evaluation of deep learning models. Our dataset is available at https://github.com/Sheng029/WHU-Stereo.

As an essential task in remote sensing, disparity estimation of high-resolution stereo images is still confronted with intractable problems due to extremely complex scenes and dynamically changing disparities. Especially in areas containing texture-less regions, repetitive patterns, disparity discontinuities, and occlusions, stereo matching is difficult. Recently, convolutional neural networks have provided a new paradigm for disparity estimation, but it is difficult for current models to consider both accuracy and speed. This paper proposes a novel end-to-end network to overcome the aforementioned obstacles. The proposed network learns stereo matching at dual scales, in which the low one captures coarse-grained information while the high one captures fine-grained information, helpful for matching structures of different scales. Moreiver, we construct cost volumes from negative to positive values to make the network work well for both negative and nonnegative disparities since the disparity varies dramatically in remote sensing stereo images. A 3D encoder-decoder module formed by factorized 3D convolutions is introduced to adaptively learn cost aggregation, which is of high efficiency and able to alleviate the edge-fattening issue at disparity discontinuities and approximate the matching of occlusions. Besides, we use a refinement module that brings in shallow features as guidance to attain high-quality full-resolution disparity maps. The proposed network is compared with several typical models. Experimental results on a challenging dataset demonstrate that our network shows powerful learning and generalization abilities. It achieves convincing performance on both accuracy and efficiency, and improvements of stereo matching in these challenging areas are noteworthy.

The Markov Random Field (MRF) energy function, constructed by existing OpenMVS-based 3D texture reconstruction algorithms, considers only the image label of the adjacent triangle face for the smoothness term and ignores the planar-structure information of the model. As a result, the generated texture charts results have too many fragments, leading to a serious local miscut and color discontinuity between texture charts. This paper fully utilizes the planar structure information of the mesh model and the visual information of the 3D triangle face on the image and proposes an improved, faster, and high-quality texture chart generation method based on the texture chart generation algorithm of the OpenMVS. This methodology of the proposed approach is as follows: (1) The visual quality on different visual images of each triangle face is scored using the visual information of the triangle face on each image in the mesh model. (2) A fully automatic Variational Shape Approximation (VSA) plane segmentation algorithm is used to segment the blocked 3D mesh models. The proposed fully automatic VSA-based plane segmentation algorithm is suitable for multi-threaded parallel processing, which solves the VSA framework needed to manually set the number of planes and the low computational efficiency in a large scene model. (3) The visual quality of the triangle face on different visual images is used as the data term, and the image label of adjective triangle and result of plane segmentation are utilized as the smoothness term to construct the MRF energy function. (4) An image label is assigned to each triangle by the minimizing energy function. A texture chart is generated by clustering the topologically-adjacent triangle faces with the same image label, and the jagged boundaries of the texture chart are smoothed. Three sets of data of different types were used for quantitative and qualitative evaluation. Compared with the original OpenMVS texture chart generation method, the experiments show that the proposed approach significantly reduces the number of texture charts, significantly improves miscuts and color differences between texture charts, and highly boosts the efficiency of VSA plane segmentation algorithm and OpenMVS texture reconstruction.

Semantic segmentation of LiDAR point clouds in powerline corridor environments is crucial for the intelligent inspection and maintenance of power infrastructure. However, existing deep learning methods often underperform in such scenarios due to severe class imbalance, sparse and long-range structures, and complex elevation variations. We propose EMPower-Net, an Elevation-Aware Multi-Resolution Network, which integrates an Elevation Distribution (ED) module to enhance vertical geometric awareness and a Multi-Resolution (MR) module to enhance segmentation accuracy for corridor structures with varying object scales. Experiments on real-world datasets from Yunnan and Guangdong show that EMPower-Net outperforms state-of-the-art baselines, especially in recognizing power lines and towers with high structural fidelity under occlusion and dense vegetation. Ablation studies confirm the complementary effects of the MR and ED modules, while transfer learning results reveal strong generalization with minimal performance degradation across different powerline regions. Additional tests on urban datasets indicate that the proposed elevation features are also effective for vertical structure recognition beyond powerline scenarios.

Driven by the growing demand for high-precision 3D spatial data in smart cities, accurate building instance segmentation from airborne LiDAR has become critical. Traditional methods often struggle in complex urban environments, while existing deep learning approaches primarily rely on local geometry and neglect explicit building morphology. To address these limitations, we propose a Contour-Guided Hybrid Graph Network (CG-HGN) for building instance segmentation. CG-HGN incorporates contour to explicitly encode building morphology and leverages a hybrid graph architecture: a superpoint graph captures semantic features, and an edge graph models inter-superpoint boundary connectivity. By formulating instance segmentation as a graph connectivity inference problem, CG-HGN improves boundary precision and alleviates over-segmentation. Experiments on four datasets with distinct characteristics demonstrate that CG-HGN consistently outperforms existing methods in terms of AP and mCov, particularly in sparse, heavily occluded, or structurally complex areas. Cross-dataset evaluations further confirm the strong generalization ability. CG-HGN provides an effective solution for high-quality building instance segmentation in real-world airborne LiDAR applications.