用Transformer直接生成高精度建筑轮廓，端到端训练更简单高效。
Pix2Poly: A Sequence Prediction Method for End-to-end Polygonal Building Footprint Extraction from Remote Sensing Imagery
- 基于Transformer的编码器-解码器结构，直接输出顶点序列
- 在多个数据集上优于现有方法，矢量形状质量指标领先
- 无需复杂损失函数，适合城市建模与地图更新任务
从遥感影像中提取建筑轮廓多边形对城市重建、导航和制图等任务至关重要。尽管已有显著进展，精确提取多边形建筑轮廓仍是开放难题。本文提出Pix2Poly，一种基于注意力机制的端到端可训练、可微分深度神经网络，能直接以环状图形式生成高质量建筑轮廓。该模型采用生成式编码器-解码器Transformer生成顶点序列，通过最优匹配网络学习连接关系。相比以往图学习方法，Pix2Poly为真正端到端可训练方案，无需复杂的栅格损失函数和繁琐训练流程即可提取高精度建筑轮廓与道路网络。在多个复杂挑战性数据集上的评估表明，其在多项矢量形状质量指标上超越当前最优方法，且为完全显式方法。代码已开源：https://github.com/yeshwanth95/Pix2Poly。
原文摘要 · Abstract (English)
Extraction of building footprint polygons from remotely sensed data is essential for several urban understanding tasks such as reconstruction, navigation, and mapping. Despite significant progress in the area, extracting accurate polygonal building footprints remains an open problem. In this paper, we introduce Pix2Poly, an attention-based end-to-end trainable and differentiable deep neural network capable of directly generating explicit high-quality building footprints in a ring graph format. Pix2Poly employs a generative encoder-decoder transformer to produce a sequence of graph vertex tokens whose connectivity information is learned by an optimal matching network. Compared to previous graph learning methods, ours is a truly end-to-end trainable approach that extracts high-quality building footprints and road networks without requiring complicated, computationally intensive raster loss functions and intricate training pipelines. Upon evaluating Pix2Poly on several complex and challenging datasets, we report that Pix2Poly outperforms state-of-the-art methods in several vector shape quality metrics while being an entirely explicit method. Our code is available at https://github.com/yeshwanth95/Pix2Poly.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。