Enhancing Photogrammetric Digital Surface Models with Pretrained Diffusion Models and Multimodal Conditioning

• 类型：arxiv
• 标识：2609.31199
• 链接：https://arxiv.org/abs/2609.31199
• 主分类：multimodal
• 形态：method
• TLDR：Large-scale Digital Surface Models (DSMs) can be produced cost-effectively from satellite images via stereo-photogrammetry. However, the resulting 3D maps are often contaminated by noise, outliers, and voids. On the other hand, aerial LiDAR provides high-accuracy elevation measurements at a substantially higher cost. In this work, we study diffusion models conditioned both on photogrammetric DSMs and Pléiades imagery to refine vertically co-registered DSMs. We introduce a modified Stable Diffusion 3 architecture with a pruned text stream and a patch-wise normalization strategy, enabling stable
• 副分类：engineering
• 待LLM分类：否
• 标题中文：[标题中文] 利用预训练扩散模型与多模态条件增强摄影测量数字表面模型
• TLDR中文：[TLDR中文] 大规模数字表面模型（DSM）可通过摄影测量方法从卫星影像中以较低成本生成。然而，由此得到的 3D 地图常受到噪声、离群点和空洞的污染。另一方面，机载 LiDAR 能以高得多的成本提供高精度的高程测量。在本工作中，我们研究以摄影测量 DSM 和 Pléiades 影像同时作为条件的扩散模型，用于精细化处理垂直配准的 DSM。我们提出一种改进的 Stable Diffusion 3 架构，采用剪枝后的文本流和逐 patch 的归一化策略，从而实现稳定的
• 来源文件：
• /inbox/tom/_candidates/2026-09-28-agent-rag-longcontext-candidates.json