# SUM: A Benchmark Dataset of Semantic Urban Meshes

arXiv: https://arxiv.org/abs/2103.00355 · HTML: https://mlsanity.com/paper/2103.00355/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 55 references that are in the map.
- First posted on arXiv 27 Feb 2021.

## Builds on (most cited first, 14 of 14)

- [PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation](https://mlsanity.com/paper/1612.00593/) (2016)
- [PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space](https://mlsanity.com/paper/1706.02413/) (2017)
- [KPConv: Flexible and Deformable Convolution for Point Clouds](https://mlsanity.com/paper/1904.08889/) (2019)
- [RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds](https://mlsanity.com/paper/1911.11236/) (2019)
- [Large-scale Point Cloud Semantic Segmentation with Superpoint Graphs](https://mlsanity.com/paper/1711.09869/) (2017)
- [Semantic3D.net: A new Large-scale Point Cloud Classification Benchmark](https://mlsanity.com/paper/1704.03847/) (2017)
- [Towards Semantic Segmentation of Urban-Scale 3D Point Clouds: A Dataset, Benchmarks and Challenges](https://mlsanity.com/paper/2009.03137/) (2020)
- [Toronto-3D: A Large-scale Mobile LiDAR Dataset for Semantic Segmentation of Urban Roadways](https://mlsanity.com/paper/2003.08284/) (2020)
- [DALES: A Large-scale Aerial LiDAR Data Set for Semantic Segmentation](https://mlsanity.com/paper/2004.11985/) (2020)
- [Paris-Lille-3D: a large and high-quality ground truth urban point cloud dataset for automatic segmentation and classification](https://mlsanity.com/paper/1712.00032/) (2017)
- [The Hessigheim 3D (H3D) Benchmark on Semantic Segmentation of High-Resolution 3D Point Clouds and Textured Meshes from UAV LiDAR and Multi-View-Stereo](https://mlsanity.com/paper/2102.05346/) (2021)
- [DublinCity: Annotated LiDAR Point Cloud and its Applications](https://mlsanity.com/paper/1909.03613/) (2019)
- [Campus3D: A Photogrammetry Point Cloud Benchmark for Hierarchical Understanding of Outdoor Scene](https://mlsanity.com/paper/2008.04968/) (2020)
- [Semantic Segmentation on Swiss3DCities: A Benchmark Study on Aerial Photogrammetric 3D Pointcloud Dataset](https://mlsanity.com/paper/2012.12996/) (2020)

## Built on by (most cited first, 23 of 23)

- [EarthNets: Empowering AI in Earth Observation](https://mlsanity.com/paper/2210.04936/) (2022)
- [UrbanBIS: a Large-scale Benchmark for Fine-grained Urban Building Instance Segmentation](https://mlsanity.com/paper/2305.02627/) (2023)
- [NuiScene: Exploring Efficient Generation of Unbounded Outdoor Scenes](https://mlsanity.com/paper/2503.16375/) (2025)
- [3D Instance Segmentation of MVS Buildings](https://mlsanity.com/paper/2112.09902/) (2021)
- [Synthetic 3D Data Generation Pipeline for Geometric Deep Learning in Architecture](https://mlsanity.com/paper/2104.12564/) (2021)
- [PSSNet: Planarity-sensible Semantic Segmentation of Large-scale Urban Meshes](https://mlsanity.com/paper/2202.03209/) (2022)
- [Learnable Earth Parser: Discovering 3D Prototypes in Aerial Scans](https://mlsanity.com/paper/2304.09704/) (2023)
- [TUM2TWIN: Introducing the Large-Scale Multimodal Urban Digital Twin Benchmark Dataset](https://mlsanity.com/paper/2505.07396/) (2025)
- [Human Vision Based 3D Point Cloud Semantic Segmentation of Large-Scale Outdoor Scene](https://mlsanity.com/paper/2301.12972/) (2023)
- [PG-SAG: Parallel Gaussian Splatting for Fine-Grained Large-Scale Urban Buildings Reconstruction via Semantic-Aware Grouping](https://mlsanity.com/paper/2501.01677/) (2025)
- [AutoEncoding Tree for City Generation and Applications](https://mlsanity.com/paper/2309.15941/) (2023)
- [Beyond Grid Data: Exploring Graph Neural Networks for Earth Observation](https://mlsanity.com/paper/2411.03223/) (2024)
- [CitySeg: A 3D Open Vocabulary Semantic Segmentation Foundation Model in City-scale Scenarios](https://mlsanity.com/paper/2508.09470/) (2025)
- [SUM Parts: Benchmarking Part-Level Semantic Segmentation of Urban Meshes](https://mlsanity.com/paper/2503.15300/) (2025)
- [Turin3D: Evaluating Adaptation Strategies under Label Scarcity in Urban LiDAR Segmentation with Semi-Supervised Techniques](https://mlsanity.com/paper/2504.05882/) (2025)
- [BuildingWorld: A Structured 3D Building Dataset for Urban Foundation Models](https://mlsanity.com/paper/2511.06337/) (2025)
- [Effective Utilisation of Multiple Open-Source Datasets to Improve Generalisation Performance of Point Cloud Segmentation Models](https://mlsanity.com/paper/2211.15877/) (2022)
- [Semantic Segmentation of Urban Textured Meshes Through Point Sampling](https://mlsanity.com/paper/2302.10635/) (2023)
- [LMSeg: An end-to-end geometric message-passing network on barycentric dual graphs for large-scale landscape mesh segmentation](https://mlsanity.com/paper/2407.04326/) (2024)
- [OpenUrban3D: Annotation-Free Open-Vocabulary Semantic Segmentation of Large-Scale Urban Point Clouds](https://mlsanity.com/paper/2509.10842/) (2025)
- [Semantic Segmentation of Textured Non-manifold 3D Meshes using Transformers](https://mlsanity.com/paper/2604.01836/) (2026)
- [COSTA: A Cluster-Centric Paradigm for Annotation-Free Open-Set Semantic Segmentation of Aerial Point Clouds with Domain Shifts](https://mlsanity.com/paper/2608.18479/) (2026)
- [Polis: 3D Self-Supervision at City Scale](https://mlsanity.com/paper/2608.29426/) (2026)

## Abstract (arXiv)

Recent developments in data acquisition technology allow us to collect 3D texture meshes quickly. Those can help us understand and analyse the urban environment, and as a consequence are useful for several applications like spatial analysis and urban planning. Semantic segmentation of texture meshes through deep learning methods can enhance this understanding, but it requires a lot of labelled data. The contributions of this work are threefold: (1) a new benchmark dataset of semantic urban meshes, (2) a novel semi-automatic annotation framework, and (3) an annotation tool for 3D meshes. In particular, our dataset covers about 4 km2 in Helsinki (Finland), with six classes, and we estimate that we save about 600 hours of labelling work using our annotation framework, which includes initial segmentation and interactive refinement. We also compare the performance of several state-of-theart 3D semantic segmentation methods on the new benchmark dataset. Other researchers can use our results to train their networks: the dataset is publicly available, and the annotation tool is released as open-source.