SatUnreal: A High-Precision Synthetic Dataset for Satellite Stereo Matching via Unreal Engine

📄 arXiv: 2609.27442v1 📥 PDF

作者: Han-Gyeol Kim, JaeWan Park, Junmin Park, Darongsae Kwon

分类: cs.CV, eess.IV

发布日期: 2026-09-23

备注: Accepted at CVPR 2026 Workshop on EarthVision (CVPRW 2026), pp. 7990-7999. Code and dataset: https://github.com/jmp-Telepix/SatUnreal_A_High-Precision_Synthetic_Dataset_for_Satellite_Stereo_Matching_via_UnrealEngine Supplementary material: https://openaccess.thecvf.com/content/CVPR2026W/EarthVision/supplemental/Kim_SatUnreal_A_High-Precision_CVPRW_2026_supplemental.pdf

期刊: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2026, pp. 7990-7999

🔗 代码/项目: GITHUB

💡 一句话要点

提出SatUnreal以解决卫星立体匹配数据集不足问题

🎯 匹配领域: 支柱一：机器人控制 (Robot Control) 支柱三：空间感知与语义 (Perception & Semantics) 支柱九：具身大模型 (Embodied Foundation Models)

关键词: 卫星影像 3D重建 合成数据集 时空一致性 遮挡标签 Unreal Engine 几何特征学习

📋 核心要点

  1. 现有卫星影像数据集在遮挡标签和时空一致性方面存在显著不足，限制了3D重建的精度。
  2. 论文提出SatUnreal数据集，通过Unreal Engine模拟生成高精度的立体图像对，解决了现有数据集的缺陷。
  3. 实验表明，基于SatUnreal训练的模型在真实数据集上表现优越，展示了合成数据在几何特征学习中的潜力。

📝 摘要（中文）

从卫星影像进行3D重建对大规模地形分析至关重要，但缺乏高保真训练数据集及准确的遮挡标签仍是主要瓶颈。现有基准如US3D和WHU-Stereo在时空不匹配方面存在固有挑战，并且由于LiDAR稀疏性在遮挡区域提供模糊的真实值。本文提出SatUnreal，一个基于Unreal Engine的高精度合成数据集，旨在根本上克服这些限制。SatUnreal提供10,000对高分辨率立体图像，具有物理几何模拟、时空一致性、地形多样性和数学标签完整性等特点。实验结果表明，专门在SatUnreal上训练的模型在真实基准上实现了优越的零-shot迁移性能，证明了物理准确的合成数据在学习几何特征方面的有效性。

🔬 方法详解

问题定义：本文旨在解决卫星影像3D重建中缺乏高保真训练数据集及准确遮挡标签的问题。现有方法如US3D和WHU-Stereo在时空不匹配和遮挡区域的真实值模糊性上存在显著挑战。

核心思路：论文提出的SatUnreal数据集通过Unreal Engine的模拟生成高精度的立体图像对，系统性地克服了现有数据集的局限性，提供了更为准确的训练信号。

技术框架：SatUnreal的整体架构包括物理几何模拟、时空一致性维护、地形多样性覆盖和数学标签完整性生成四个主要模块。通过这些模块，数据集能够提供高质量的立体图像对。

关键创新：最重要的创新点在于采用了一种新颖的两步线追踪算法，生成无瑕疵的遮挡掩码，这在现有方法中是未曾实现的。

关键设计：在数据生成过程中，系统性地调整基线和方位角以模拟真实卫星轨道，同时通过固定虚拟环境消除时间噪声，确保数据的时空一致性。

🖼️ 关键图片

fig_0
fig_1
fig_2

📊 实验亮点

实验结果显示，基于SatUnreal训练的模型在US3D和WHU-Stereo等真实基准上实现了显著的性能提升，零-shot迁移性能优于仅在真实数据集上训练的模型，验证了合成数据在几何特征学习中的有效性。

🎯 应用场景

该研究的潜在应用领域包括城市规划、环境监测和灾害管理等。通过提供高质量的合成数据，SatUnreal能够帮助研究人员和工程师更有效地进行地形分析和3D重建，推动相关领域的发展。

📄 摘要（原文）

3D reconstruction from satellite imagery is essential for large-scale topographic analysis, yet the lack of high-fidelity training datasets with accurate occlusion labels remains a primary bottleneck. Existing benchmarks, such as US3D and WHU-Stereo, face inherent challenges in spatio-temporal mismatch -- environmental changes and shadow displacements between multi-view acquisitions -- and provide ambiguous ground truth in occluded regions due to LiDAR sparsity. In this paper, we propose SatUnreal, a high-precision synthetic dataset designed to fundamentally overcome these limitations through an Unreal Engine-based simulation pipeline. SatUnreal provides 10,000 stereo pairs with high resolution (0.3m GSD) and is characterized by: (1) Physical Geometry Simulation, replicating realistic satellite orbits by systematically varying baselines and azimuths; (2) Spatio-temporal Consistency, eliminating temporal noise through fixed virtual environments; (3) Topographic Diversity, spanning dense urban canyons to low-texture natural terrains; and (4) Mathematical Label Integrity, utilizing a novel two-step linetrace algorithm to generate flawless occlusion masks. Experimental results using SOTA iterative models demonstrate that models trained exclusively on SatUnreal achieve superior zero-shot transfer performance on real-world benchmarks (US3D, WHU-Stereo) compared to those trained on real datasets. Our findings prove that physically accurate synthetic data provides a more effective supervisory signal for learning geometric features than complex real-world observations, establishing a new paradigm for Sim-to-Real transfer in Earth Observation. Code and dataset are available at https://github.com/jmp-Telepix/SatUnreal_A_High-Precision_Synthetic_Dataset_for_Satellite_Stereo_Matching_via_UnrealEngine