Instance Segmentation and Fine-grained Classification for Urban Buildings with Adaptive Region Dividing and Spatially-Supervised Contrastive Learning
作者: Weiyuan Zhang, Qi Zhang, Hui Huang
分类: cs.CV
发布日期: 2026-09-17
备注: 10 pages, 4 figures
💡 一句话要点
提出自适应区域划分与空间监督对比学习以解决城市建筑实例分割与细粒度分类问题
🎯 匹配领域: 支柱二:RL算法与架构 (RL & Architecture)
关键词: 城市建筑 实例分割 细粒度分类 自适应区域划分 空间监督对比学习 点云处理 数字城市建模
📋 核心要点
- 现有方法依赖于预定义区域进行训练和评估,导致建筑结构碎片化且不适用于真实场景。
- 提出自适应区域划分策略,通过鸟瞰图平面检测建筑区域,构建动态分区,避免手动设计。
- 在UrbanBIS和STPLS3D数据集上进行实验,结果显示所提方法在实例分割和细粒度分类上优于现有最先进方法。
📝 摘要(中文)
准确理解城市建筑的实例级和功能性信息对于数字城市建模和城市分析至关重要。然而,现有方法通常依赖于预定义的区域进行训练和评估,这在实际应用中较为稀缺,并且会导致额外的预处理和建筑结构的碎片化。为此,本文提出了一种自适应区域划分策略,并结合统一的场景级评估。具体而言,3D点云被投影到鸟瞰图平面,利用预训练的分割模型检测建筑区域,并将检测到的边界框回投影到原始点云中,构建结构对齐的自适应训练块。此外,本文还提出了一种细粒度分类模型,采用空间监督对比损失来增强类别间的可区分性。通过在UrbanBIS和STPLS3D数据集上的广泛实验,验证了所提方法在建筑实例分割和细粒度分类方面的优势。
🔬 方法详解
问题定义:本文旨在解决城市建筑在大规模点云中的实例分割与细粒度分类问题。现有方法通常依赖于手动定义的区域进行训练和评估,导致建筑结构的碎片化和不适用性。
核心思路:提出自适应区域划分策略,通过将3D点云投影到鸟瞰图平面,利用预训练的分割模型自动检测建筑区域,构建结构对齐的训练块,避免手动设计。
技术框架:整体流程包括三个主要模块:首先,3D点云投影到鸟瞰图平面;其次,使用分割模型检测建筑区域并回投影;最后,采用点变换分类器进行细粒度分类,并引入空间监督对比损失。
关键创新:最重要的创新在于自适应区域划分与空间监督对比学习的结合,能够在没有手动设计的情况下实现动态分区,并增强类别间的可区分性。
关键设计:采用类平衡加权交叉熵损失来缓解类别不平衡问题,同时引入空间监督对比损失,赋予空间上相近的同类建筑更大的权重,以鼓励紧凑的功能表示。
🖼️ 关键图片
📊 实验亮点
在UrbanBIS和STPLS3D数据集上的实验结果表明,所提方法在建筑实例分割和细粒度分类方面相较于现有最先进方法有显著提升,具体性能数据未提供,但提升幅度显著,验证了方法的有效性和实用性。
🎯 应用场景
该研究在城市建筑的数字建模、城市规划和智能城市建设等领域具有广泛的应用潜力。通过提高建筑实例分割和细粒度分类的准确性,能够为城市分析、环境监测和基础设施管理提供更为精准的数据支持,进而推动智慧城市的发展。
📄 摘要(原文)
Accurate instance-level and functional understanding of urban buildings in large-scale point clouds is essential for digital city modeling and urban analysis. However, the extensive spatial coverage of urban scenes leads most existing methods to rely on predefined blocks for training and evaluation, although such partitions are rarely available in real-world applications and introduce additional preprocessing while fragmenting complete building structures. To address this issue, we propose an adaptive region-dividing strategy with unified scene-level evaluation. Specifically, the 3D point cloud is projected onto a bird's-eye-view (BEV) plane, where a pretrained segmentation model is used to detect building regions. The detected bounding boxes are then back-projected to the original point cloud to construct structure-aligned adaptive training blocks, enabling semantically guided dynamic partitioning without manual design. Furthermore, beyond instance-level understanding, few methods have explored fine-grained classification for urban buildings, and thus we also put forward a fine-grained classification model for urban buildings with a spatially-supervised contrastive loss. First, for each segmented building, a point transformer classifier jointly encodes its body and local context using geometric, color, and core-context information. Then, the class-balanced weighted cross-entropy is used to alleviate severe class imbalance. The proposed spatially-supervised contrastive loss further enhances inter-class discriminability by assigning greater weight to spatially proximate, same-category buildings, encouraging compact functional representations while separating easily confused categories. Extensive experiments on UrbanBIS and STPLS3D demonstrate the advantages of the proposed method in building instance segmentation and fine-grained classification compared to existing SOTA methods.